=========================================================================== COMPARATIVE STUDY OF ELECTORAL SYSTEMS (CSES) - MODULE 3 (2006-2011) CODEBOOK: INTRODUCTION FULL RELEASE - DECEMBER 15, 2015 VERSION CSES Secretariat www.cses.org =========================================================================== HOW TO CITE THE STUDY: The Comparative Study of Electoral Systems (www.cses.org). CSES MODULE 3 FULL RELEASE [dataset]. December 15, 2015 version. doi:10.7804/cses.module3.2015-12-15 These materials are based on work supported by the American National Science Foundation (www.nsf.gov) under grant numbers SES-0451598 , SES-0817701, and SES-1154687, the GESIS - Leibniz Institute for the Social Sciences, the University of Michigan, in-kind support of participating election studies, the many organizations that sponsor planning meetings and conferences, and the many organizations that fund election studies by CSES collaborators. Any opinions, findings and conclusions, or recommendations expressed in these materials are those of the author(s) and do not necessarily reflect the views of the funding organizations. =========================================================================== IMPORTANT NOTE REGARDING FULL RELEASES: This dataset and all accompanying documentation is the "Full Release" of CSES Module 3 (2006-2011). Users of the Final Release may wish to monitor the errata for CSES Module 3 on the CSES website, to check for known errors which may impact their analyses. To view errata for CSES Module 3, go to the Data Center on the CSES website, navigate to the CSES Module 3 download page, and click on the Errata link in the gray box to the right of the page. =========================================================================== TABLE OF CONTENTS =========================================================================== ))) CSES PROJECT PROFILE >>> CSES MODULE 3 STUDY DESCRIPTION >>> CSES MODULE 3 PLANNING COMMITTEE >>> CSES MODULE 4 PLANNING COMMITTEE >>> CSES MODULE 3 COLLABORATORS >>> CSES MODULE 3 SECRETARIAT >>> MICRO-LEVEL (SURVEY) COMPONENT >>> CSES MODULE 3 COLLABORATOR INSTRUCTIONS FOR THE ADMINISTRATION OF THE CSES QUESTIONNAIRE >>> DISTRICT-LEVEL COMPONENT >>> MACRO-LEVEL COMPONENT ))) HOW TO USE THE CSES MODULE 3 DOCUMENTATION >>> TYPES OF DOCUMENTATION AVAILABLE >>> CODEBOOK CONVENTIONS ))) HOW TO USE THE CSES MODULE 3 DATA FILES ))) SPECIAL DATA NOTES >>> IDENTIFICATION VARIABLES >>> MISSING DATA >>> WEIGHTS >>> FREEDOM STATUS OF ELECTIONS ))) CSES MODULE 3 ELECTION STUDIES >>> LIST OF ELECTION STUDIES INCLUDED IN CSES MODULE 3 >>> OTHER ELECTION STUDIES INCLUDING CSES MODULE 3 ))) ELECTION SUMMARIES AND GENERAL NOTES >>> ELECTION SUMMARIES AND GENERAL NOTES - AUSTRALIA (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - AUSTRIA (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - BELARUS (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - BRAZIL (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - BRAZIL (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - CANADA (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - CROATIA (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - CHILE (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - CZECH REPUBLIC (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - CZECH REPUBLIC (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - DENMARK (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - ESTONIA (2011) >>> ELECTION SUMMARIES AND GENERAL NOTES - FINLAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - FINLAND (2011) >>> ELECTION SUMMARIES AND GENERAL NOTES - FRANCE (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - GERMANY (2005) >>> ELECTION SUMMARIES AND GENERAL NOTES - GERMANY (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - GREECE (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - HONG KONG (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - ICELAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - ICELAND (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - IRELAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - ISRAEL (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - JAPAN (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - LATVIA (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - MEXICO (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - MEXICO (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - NETHERLANDS (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - NETHERLANDS (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - NEW ZEALAND (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - NORWAY (2005) >>> ELECTION SUMMARIES AND GENERAL NOTES - NORWAY (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - PERU (2011) >>> ELECTION SUMMARIES AND GENERAL NOTES - PHILIPPINES (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - POLAND (2005) >>> ELECTION SUMMARIES AND GENERAL NOTES - POLAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - PORTUGAL (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - ROMANIA (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - SLOVAKIA (2010) >>> ELECTION SUMMARIES AND GENERAL NOTES - SLOVENIA (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - SOUTH AFRICA (2009) >>> ELECTION SUMMARIES AND GENERAL NOTES - SOUTH KOREA (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - SPAIN (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - SWEDEN (2006) >>> ELECTION SUMMARIES AND GENERAL NOTES - SWITZERLAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - TAIWAN (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - THAILAND (2007) >>> ELECTION SUMMARIES AND GENERAL NOTES - TURKEY (2011) >>> ELECTION SUMMARIES AND GENERAL NOTES - UNITED STATES (2008) >>> ELECTION SUMMARIES AND GENERAL NOTES - URUGUAY (2009) ))) BIBLIOGRAPHY >>> WEBSITES >>> ARTICLES AND BOOKS =========================================================================== ))) CSES PROJECT PROFILE =========================================================================== The Comparative Study of Electoral Systems (CSES) is a collaborative program of research among election study teams from around the world. Participating countries include a common module of survey questions in their post-election studies. The resulting data are deposited along with voting, demographic, district and macro variables. The studies are then merged into a single, free, public dataset for use in comparative study and cross-level analysis. The research agenda, questionnaires, and study design are developed by an international committee of leading scholars of electoral politics and political science. The design is implemented in each country by their foremost social scientists. By collaborating in this way, the CSES community hopes to forward scientific inquiry into the relationship between electoral institutions and political behavior. --------------------------------------------------------------------------- >>> CSES MODULE 3 STUDY DESCRIPTION --------------------------------------------------------------------------- CSES Module 3 is scheduled to be in use from 2006 through 2011. It is based on the stimulus paper by Hermann Schmitt and Bernhard Wessels titled "Meaningful Choices: Under what conditions do general elections provide a meaningful choice set, and what happens if they don't?" The Module focuses on voters' perceptions of, assessments of, and responses to the variety and quality of political choices in an election. It considers several aspects of voter perceptions of the choices they are offered. The first involves retrospective evaluations of candidates and parties. Second, the module incorporates ideology, party image, and policy differences between parties as bases for prospective evaluations. Third, the module incorporates respondents' summary perceptions of their own political choices. The module also provides instrumentation to address the potential consequences of political choice sets of varying quality. (excerpted from Burns and McAllister, 2008) --------------------------------------------------------------------------- >>> CSES MODULE 3 PLANNING COMMITTEE --------------------------------------------------------------------------- The CSES Module 3 Planning Committee was responsible for the design of CSES Module 3, and shared responsibility for its implementation. The following persons were members of the CSES Module 3 Planning Committee: Ian McAllister, chair Australian National University Australia Bernt Aardal The Norwegian Election Studies Norway Kees Aarts University of Twente The Netherlands John Aldrich Duke University United States Ulises Beltrán CIDE (Centro de Investigacion y Docencia Economica) Mexico André Blais Université de Montréal Canada Yun-Han Chu Academia Sinica Taiwan Russell Dalton University of California, Irvine United States Juan Díez-Nicolás ASEP (Analisis Sociológicos Económicos y Políticos) Spain David A. Howell, ex officio Director of Studies, CSES Secretariat University of Michigan United States Ken'ichi Ikeda The University of Tokyo Japan Christiaan Keulder Institute for Public Policy Research (IPPR) Namibia Marta Lagos Latinobarómetro Chile Radoslaw Markowski Polish Academy of Sciences Poland Ekkehard Mochmann Zentralarchiv für Empirische Sozialforschung Germany Hans Rattinger, ex officio GESIS - Leibniz Institute for the Social Sciences Germany Hermann Schmitt Mannheimer Zentrum für Europäische Sozialforschung (MZES) Germany Michal Shamir Tel Aviv University Israel Sandeep Shastri MATS University India Gábor Tóka Central European University Hungary Jack Vowles University of Exeter United Kingdom Bernhard Weßels Wissenschaftszentrum Berlin für Sozialforschung (WZB) Germany --------------------------------------------------------------------------- >>> CSES MODULE 4 PLANNING COMMITTEE --------------------------------------------------------------------------- The CSES Module 4 Planning Committee shared responsibility with the CSES Module 3 Planning Committee for the implementation of CSES Module 3. The following persons were members of the CSES Module 4 Planning Committee: André Blais, chair Université de Montréal Canada Bernt Aardal University of Oslo Norway Kees Aarts University of Twente The Netherlands John Aldrich Duke University United States Ulises Beltrán Centro de Investigacion y Docencia Economica (CIDE) Mexico Goran Cular University of Zagreb Croatia Rachel Gibson University of Manchester United Kingdom Elisabeth Gidengil McGill University Canada Sara Hobolt Oxford University United Kingdom David Howell, ex officio Director of Studies, CSES Secretariat University of Michigan United States Chi Huang National Chengchi University Taiwan Ken'ichi Ikeda University of Tokyo Japan Pedro Magalhaes Lisbon University Portugal Radoslaw Markowski Polish Academy of Sciences Poland Rachel Meneguello Universidade de Campinas Brazil David Sanders University of Essex United Kingdom Nicolas Sauger Sciences Po France Michal Shamir Tel Aviv University Israel Jack Vowles University of Exeter United Kingdom Bernhard Weßels Wissenschaftszentrum Berlin für Sozialforschung (WZB) Germany --------------------------------------------------------------------------- >>> CSES MODULE 3 COLLABORATORS --------------------------------------------------------------------------- The CSES project is extremely grateful to our Module 3 collaborators, who raised their own funds to include CSES Module 3 in a nationally representative post-election study in their country or province. Listed collaborators are our contact persons for election studies that appear in the CSES Module 3 full release dataset - they are not necessarily the parties who collected the data. Most election studies benefitted from the scientific input and data preparation skills of additional persons not listed here. Within each election study, collaborators are presented in alphabetical order. The affiliations listed are current as of the date when each election study's Design Report was deposited with CSES. Listed collaborators are our contact persons for election studies that appear in the CSES Module 3 full release dataset - they are not necessarily the parties who collected the data, and they are not necessarily the only investigators on each study. Within each election study, collaborators are presented in the order in which they appear in their election study's Design Report. Organizational affiliation for each person is from the date when the election study first appeared in the CSES dataset. - Australia (2007) Clive Bean Queensland University of Technology Australia Rachel Gibson University of Manchester United Kingdom Ian McAllister The Australian National University Australia - Austria (2008) Fritz Plasser University of Innsbruck Austria Sylvia Kritzinger Universität Wien Austria Wolfgang C. Müller Universität Wien Austria Günther Lengauer University of Innsbruck Austria - Belarus (2008) David Rotman Belarussian State University Belarus Ian McAllister The Australian National University Australia Stephen White University of Glasgow Scotland - Brazil (2006) Rachel Meneguello University of Campinas Brazil Clifford Young IPSOS-Opinion Brasil Ltda. Brazil Alberto Carlos Melo de Almeida IPSOS-Opinion Brasil Ltda. Brazil - Brazil (2010) Rachel Meneguello University of Campinas Brazil Marta Maia Vox Populi Brazil - Canada (2008) Elisabeth Gidengil McGill University Canada Patrick Fournier Université de Montréal Canada Joanna Everitt University of New Brunswick (Saint John) Canada Neil Nevitte University of Toronto Canada - Chile (2009) Marta Lagos MORI Chile SA Chile - Croatia (2007) Nenad Zakosek University of Zagreb Croatia Goran Cular University of Zagreb Croatia - Czech Republic (2006) Lukás Linek Czech Academy of Sciences Czech Republic - Czech Republic (2010) Lukás Linek Czech Academy of Sciences Czech Republic - Denmark (2007) Jřrgen Goul Andersen Aalborg University Denmark Rune Stubager Aarhus University Denmark Kasper Mřller Hansen University of Copenhagen Denmark - Estonia (2011) Piret Ehin University of Tartu Estonia Mihkel Solvak University of Tartu Estonia - Finland (2007) Heikki Paloheimo University of Tampere Finland Lauri Karvonen Ĺbo Akademi University Finland - Finland (2011) Sami Borg University of Tampere Finland Kimmo Grönlund Ĺbo Akademi University Finland - France (2007) Nicolas Sauger Sciences Po France - Germany (2005) Bernhard Weßels Wissenschaftszentrum Berlin für Sozialforschung (WZB) Germany Hermann Schmitt Mannheimer Zentrum für Europäische Sozialforschung (MZES) Germany - Germany (2009) Bernhard Weßels Wissenschaftszentrum Berlin für Sozialforschung (WZB) Germany Hans Rattinger Universität Mannheim Germany Sigrid Roßteutscher Universität Frankfurt Germany Rüdiger Schmitt-Beck Universität Mannheim Germany - Greece (2009) Ioannis Andreadis Aristotle University of Thessaloniki Greece Theodore Chadjipadelis Aristotle University of Thessaloniki Greece Eftichia Teperoglou Aristotle University of Thessaloniki Greece - Hong Kong (2008) Li Pang-kwong, Ph.D. Lingnan University Hong Kong - Iceland (2007) Olafur Th. Hardarson University of Iceland Iceland Eva Heiđa Önnudóttir Bifrost University Iceland Einar Mar Ţórđarson University of Iceland Iceland - Iceland (2009) Olafur Th. Hardarson University of Iceland Iceland Eva Heiđa Önnudóttir Bifrost University Iceland Einar Mar Ţórđarson University of Iceland Iceland - Ireland (2007) Michael Marsh Trinity College Dublin (TCD) Ireland Richard Sinnott University College Dublin (UCD) Ireland - Israel (2006) Michal Shamir Tel Aviv University Israel Asher Arian The Israel Democracy Institute Israel - Japan (2007) Ken'ichi Ikeda The University of Tokyo Japan Masahiro Yamada Kwansei-Gakuin University Japan Yoshitaka Nishizawa Doshisha University Japan Kazunori Inamasu The University of Tokyo Japan - Latvia (2010) Janis Ikstens University of Latvia Latvia - Mexico (2006) Ulises Beltrán Ugarte División de Estudios Políticos (CIDE) Mexico - Mexico (2009) Ulises Beltrán Ugarte División de Estudios Políticos (CIDE) Mexico - Netherlands (2006) Kees Aarts University of Twente Netherlands Henk van der Kolk University of Twente Netherlands - Netherlands (2010) Henk van der Kolk University of Twente Netherlands Kees Aarts University of Twente Netherlands - New Zealand (2008) Jack Vowles University of Exeter United Kingdom - Norway (2005) Bernt Aardal Institute for Social Research Norway - Norway (2009) Bernt Aardal Institute for Social Research Norway - Peru (2011) David Sulmont Pontificia Universidad Católica del Perú Peru Vania Martínez Pontificia Universidad Católica del Perú Peru - Philippines (2010) Linda Luz Guerrero Social Weather Stations Philippines Vladymir Joseph Licudine Social Weather Stations Philippines - Poland (2005) Radoslaw Markowski Polish Academy of Sciences Poland Pawel Grzelak Polish Academy of Sciences Poland Mikolaj Czesnik Polish Academy of Sciences Poland Michal Kotnarowski Polish Academy of Sciences Poland - Poland (2007) Radoslaw Markowski Polish Academy of Sciences Poland Pawel Grzelak Polish Academy of Sciences Poland Mikolaj Czesnik Polish Academy of Sciences Poland Michal Kotnarowski Polish Academy of Sciences Poland - Portugal (2009) António Barreto University of Lisbon Portugal Pedro Magalhăes University of Lisbon Portugal Marina Costa Lobo University of Lisbon Portugal - Romania (2009) Mircea Comsa Babes-Bolyai University of Cluj-Napoca Romania Gabriel Badescu Babes-Bolyai University of Cluj-Napoca Romania Andrei Gheorghita Lucian Blaga University of Sibiu Romania Cristina Stanus Babes-Bolyai University of Cluj-Napoca Romania Claudiu D. Tufis Romanian Academy Romania Coca Vieru Babes-Bolyai University of Cluj-Napoca Romania Bogdan Voicu Romanian Academy Romania Ovidiu Voicu Soros Foundation Romania Romankia - Slovakia (2010) Olga Gyarfasova Institute for Public Affairs Slovakia Zora Butorova Institute for Public Affairs Slovakia Vladimír Krivy Slovak Academy of Sciences Slovakia - Slovenia (2008) Niko Tos CJMMK Slovenia Brina Malnar CJMMK Slovenia Slavko Kurdija CJMMK Slovenia - South Africa (2009) Robert Mattes University of Cape Town South Africa Collette Schulz-Herzenberg University of Cape Town South Africa - South Korea (2008) Nam Young Lee Sejong University South Korea Wook Kim Paichai University South Korea - Spain (2008) Juan Díez-Nicolás Complutense University and ASEP Spain Antonio M. Jaime-Castillo University of Grenada Spain - Sweden (2006) Sören Holmberg University of Gothenburg Sweden Henrik Oscarsson University of Gothenburg Sweden - Switzerland (2007) Georg Lutz Université de Lausanne, Vidy Switzerland - Taiwan (2008) Chi Huang National Chengchi University Taiwan Ching-hsin Yu National Chengchi University Taiwan - Thailand (2007) Thawilwadee Bureekul King Prajadhipok’s Institute Thailand Robert B. Albritton University of Mississippi United States Tharada Tangkananurak King Prajadhipok’s Institute Thailand - Turkey (2011) Ali Carkoglu Koc University Turkey Ersin Kalaycioglu Sabanci University Turkey - United States (2008) Darrell Donakowski University of Michigan United States Arthur Lupia University of Michigan United States Jon Krosnick Stanford University United States - Uruguay (2009) Oscar Bottinelli Instituto Factum Uruguay Eduardo Bottinelli Instituto Factum Uruguay Nadia Mateo Instituto Factum Uruguay --------------------------------------------------------------------------- >>> CSES MODULE 3 SECRETARIAT --------------------------------------------------------------------------- While the CSES project and its governance are international in nature, the CSES Secretariat for CSES Module 3 was a cooperation between two organizations: the Center for Political Studies, Institute for Social Research, University of Michigan, United States; and GESIS - Leibniz Institute for the Social Sciences, Germany. Support was received from various sources for the activities of the CSES Secretariat during the period of CSES Module 3: 1. American National Science Foundation (NSF) grant SES-0451598, "The Comparative Study of Electoral Systems" with Principal Investigators Nancy Burns (University of Michigan), Donald Kinder (University of Michigan), and Ian McAllister (Australian National University), provided support for the entire CSES Secretariat from the beginning of CSES Module 3 through 2008. 2. American National Science Foundation (NSF) grant SES-0817701, "The Dynamics of Political Choice: the Third Module of the Comparative Study of Electoral Systems (CSES)" with Principal Investigators Nancy Burns (University of Michigan) and Ian McAllister (Australian National University), supported CSES Secretariat activities at the University of Michigan from 2008-2012. 3. American National Science Foundation (NSF) grant SES-1154687, "Distributional Politics and Social Protection: the Fourth Module of the Comparative Study of Electoral Systems (CSES)" with Principal Investigators Nancy Burns (University of Michigan) and André Blais (University of Montreal), supported CSES Secretariat activities at the University of Michigan beginning in 2012. 4. The work by the GESIS - Leibniz Institute for the Social Sciences was funded by the German Federal Ministry of Education and Research and governments of several German Laender. 5. The Center for Political Studies at the University of Michigan provided additional financial support. Professor Ian McAllister of The Australian National University, and chair of the CSES Module 3 Planning Committee, oversaw operations of the CSES Secretariat through 2009. Professor André Blais of the University of Montreal and chair of the CSES Module 4 Planning Committee, was responsible for oversight of operations through the remainder of CSES Module 3. Various persons staffed the CSES Secretariat throughout the Module 3 period. David Howell served as the Director of Studies. Matias Bargsted, Kathrin Busch, Jessica Fortin-Rittberger, Sebastian Netscher, Bojan Todosijevic, and Jill Wittrock provided research support, documentation, data preparation, and other services. Laurie Winslow served as project administrator. Ryan Burton provided information technology support. --------------------------------------------------------------------------- >>> MICRO-LEVEL (SURVEY) COMPONENT --------------------------------------------------------------------------- The core questionnaire ("Module") of CSES Module 3 was intended to be administered as a single, uninterrupted block of questions in a nationally representative post-election survey in each country. A. The question text is included in the variable documentation of this codebook. The questions are reported in the order in which they appear in the CSES questionnaire. B. Where there are known differences in the way a particular question was administered in an election study, this is noted in the "Election Study Notes" following the documentation of the corresponding variable. C. There are several sets of party and leader evaluation items included in the module. These correspond to parties labeled A-F, in descending order of vote share, of the six most popular parties in the lower house elections (or presidential elections if legislative elections were not held). Where respondents were asked to evaluate other parties, these evaluations have been included where possible and are labeled parties G-I, regardless of their vote shares. The parties and leaders to which these evaluations apply are identified in Appendix I. D. There are several questions (including the vote-choice and party identification items) that ask the respondents to specify a political party. The codes for these items are also reported in Appendix I. --------------------------------------------------------------------------- >>> CSES MODULE 3 COLLABORATOR INSTRUCTIONS FOR THE ADMINISTRATION OF THE CSES QUESTIONNAIRE --------------------------------------------------------------------------- The following instructions appeared in the header to the questionnaire for CSES Module 3, as instructions to collaborators regarding the implementation of the questionnaire. INSTRUCTIONS FOR THE ADMINISTRATION OF THE CSES QUESTIONNAIRE: ( 1) Following these instructions, this document is comprised of three sections: ))) CSES MODULE 3 QUESTIONNAIRE: ADMINISTRATIVE VARIABLES The "Administrative Variables" section is a list of common administrative variables that, if possible, should be provided at the time data are deposited with the CSES Secretariat. ))) CSES MODULE 3 QUESTIONNAIRE: CSES MODULE This is the CSES Module itself, a common module of survey questions for researchers to include in their national post-election survey. The CSES Module is intended to be administered exactly as it is specified in this document. ))) CSES MODULE 3 QUESTIONNAIRE: DEMOGRAPHIC VARIABLES Collaborators are asked to provide data on background (demographic) characteristics of respondents, coded to an agreed upon set of standards as indicated in this section. There is great international variation in the ways that collaborators will go about soliciting information on the background characteristics of their respondents. The objective here is not standardization of the way collaborators ask these background questions, but instead, standardization to a common, cross-national scheme for coding each variable. ( 2) The CSES Module is intended to be administered in its entirety as a single, uninterrupted block of questions. In most cases, the CSES Module is included as part of a larger study. For reliable comparisons to be made, it is important that any additional items investigators may wish to include do not interrupt the CSES Module. ( 3) The CSES module should be administered as a post-election interview. ( 4) Where the CSES module is included in a larger study, to ensure that question-ordering effects are minimized, it is most preferable for the CSES Module to be administered at the beginning of the survey instrument. Where this is not possible, collaborators should be sensitive to the effects questions asked immediately prior to the module may have. ( 5) NOTES often precede the question TEXT, and provide instructions for the administration of the item. Where no question TEXT is provided, collaborators should provide documentation of the question used. ( 6) The response options that should be read to the respondent are contained in the body of the question. ( 7) Where lower-case words appear in brackets [ ] collaborators should select the words that are most appropriate. For example: [party/presidential candidate] ...indicates that either the word "party" or the phrase "presidential candidate" should be read, but not both. ( 8) Where upper-case words appear in brackets [ ] collaborators should substitute the words that are most appropriate. For example: [COUNTRY] ...should be replaced with the name of the country where the election was held (perhaps "Canada" or "the Philippines"). Another example: [NUMBER OF YEARS BETWEEN THE PREVIOUS AND THE PRESENT ELECTION OR CHANGE IN GOVERNMENT] ...should be replaced with a number that indicates the amount of years that have passed between the previous election and either the current election or recent change in government. ( 9) Phrases that appear in parentheses ( ) contain words that are optional -- that collaborators (or their interviewers) can decide to read or not read to respondents as needed. (10) Several of the items in the CSES Module follow skip patterns that are noted by braces < >. When a question is inappropriate for a particular respondent because of the skip pattern, code the response 'MISSING'. (11) Respondents who volunteer the response "Don't know" (or who refuse to answer a question) should be coded appropriately. Interviewers should accept this response and should not probe for additional information or force a respondent to use one of the response categories provided in the text of the question. (12) Showcards may be helpful for the administration of some questions. For this reason, a Respondent Booklet is available for download from the CSES website. This Respondent Booklet contains showcards for select questions. (13) Special care should be taken in the administration of the Vote Choice items. For countries where voters have two votes for the same institution (e.g., parallel and mixed member proportional systems; double ballot systems), please ensure that both/all votes are supplied. For countries where more than one institution is elected on the same day (e.g., president and legislature) using different votes, please ensure all votes are supplied. For countries using preferential systems (e.g., STV, AV) please provide first and second preference vote. (14) For questions asking about parties, collaborators should be advised that they may add one or several party blocs to a list of individual parties if they feel that it will be difficult for respondents to recognize individual parties. (15) Collaborators in the Comparative Study of Electoral Systems shall adhere to the following standards of data quality: a. Mode of interviewing: Interviews should be conducted face-to-face, unless local circumstances dictate that telephone or mail surveys will produce higher quality data. b. Timing of interviewing: We strongly recommend that collaborators in the Comparative Study of Electoral Systems conduct their interviews in the weeks following their national election. Out of concern for data quality, data collection should be completed in as timely a fashion as possible. In the event of a runoff election, interviewing shall be conducted after the first round election. The date of interview shall be provided for each respondent. c. Placement of module in post-election questionnaire: The questionnaire module should be asked as a single, uninterrupted block of questions. We leave it to each collaborator to select an appropriate location for the module in their national survey instrument. Collaborators should take steps to ensure that questions asked immediately prior to the questionnaire module do not contaminate the initial questions in the module. Collaborators are also free to select an appropriate place in their survey instrument to ask the turnout, vote choice, and demographic questions. d. Population to be sampled: National samples should be drawn from all age-eligible citizens. When non-citizens (or other non-eligible respondents) are included in the sampled, a variable should be provided to permit the identification of those non-eligible respondents. When a collaborator samples from those persons who appear on voter registration lists, the collaborator should quantify the estimated degree of discrepancy between this population and the population of all age-eligible citizens. e. Sampling procedures: We strongly encourage the use of random samples, with random sampling procedures used at all stages of the sampling process. Collaborators should provide detailed documentation of their sampling practices. f. Sample Size: We strongly recommend that no fewer than 1,000 age-eligible respondents be interviewed. g. Interviewer training: Collaborators should pre-test their survey instrument and should train interviewers in the administration of the questionnaire. The Planning Committee will provide each collaborator with documentation that clarifies the purposes and objectives of each item and with rules with respect to probing "don't know" responses. h. Field practices: Collaborators should make every effort to ensure a high response rate. Investigators should be diligent in their effort to reach respondents not interviewed on the initial contact with the household and should be diligent in their effort to convert respondents who initially refuse to participate in the study. Data on the number of contact attempts, the number of contacts with sample persons, and special persuasion or conversion efforts undertaken should be coded for each respondent. i. Strategies for translation (and back-translation): Each collaborator should translate the questionnaire module into their native language(s). To ensure the equivalence of the translation, collaborators shall perform an independent re-translation of the questionnaire back into English. Collaborators engaged in translation of the questionnaire module into the same language (e.g., Spanish, French, English, German, and Portuguese) should collaborate on the translation. --------------------------------------------------------------------------- >>> DISTRICT-LEVEL COMPONENT --------------------------------------------------------------------------- The district-level variables report the returns of the lower house (first segment) election for each respondent's district. Wherever possible, these data were collected from official electoral commissions (see Bibliography for details). In other cases, CSES has been grateful for the compilations of these data provided by third-party sources. --------------------------------------------------------------------------- >>> MACRO-LEVEL COMPONENT --------------------------------------------------------------------------- To supplement the micro (survey) data, the teams of researchers responsible for the collection of the public opinion data also compiled and deposited the following types of data: electoral legislation, political party platforms, and official electoral returns. To facilitate this process, a detailed questionnaire was constructed to serve as a framework for the macro component of the project. The Macro Data Reports, completed by the CSES collaborators, can be found on the CSES website in the Module 3 section under the "Data Center". Additional measures thought pertinent to the micro-district-macro design are also compiled and available in the CSES data files. A bibliography of the sources consulted during the compilation of macro data follows the main body of these introductory materials. =========================================================================== ))) HOW TO USE THE CSES MODULE 3 DOCUMENTATION =========================================================================== --------------------------------------------------------------------------- >>> TYPES OF DOCUMENTATION AVAILABLE --------------------------------------------------------------------------- There are several components to the CSES documentation. Analysts will want to become familiar with all of them. For most election studies, collaborators have provided documentation to accompany their election studies, including the source macro reports, design reports, and original-language questionnaire(s). These documents, where available, can be found on the CSES website (www.cses.org) by visiting the "Data Center" and then clicking on the CSES Module 3 download page. The CSES Module 3 questionnaire is also available from the website or by referencing the corresponding variables in this codebook. The codebook consists of three files: "cses3_codebook_part1_introduction.txt" is the codebook introduction, the file you are reading now; it includes an overview of the study and data, information about use of the files, general election study descriptions and notes, and a bibliography "cses3_codebook_part2_variables.txt" is the variable descriptions' file, including questions, code frames, general notes, and notes specific to an election study (by variable) "cses3_codebook_part3_appendices.txt" contains Appendix I (Parties and Leaders) and Appendix II (Primary Electoral Districts). Analysts will also want to become familiar with the CSES Module 3 errata page, which is accessible from the CSES Module 3 download page on the website. Updates, error notifications and corrections are posted there, often in real time, as they become available. Please check the errata page regularly for new notifications. --------------------------------------------------------------------------- >>> CODEBOOK CONVENTIONS --------------------------------------------------------------------------- In the CSES Module 3 dataset, all variables begin with the letter "C". This convention helps reduce the possibility of overwriting data when merging with other CSES data sets. Variables are presented in five groupings: C1001-C1999 Identification, weight, and election study variables C2001-C2999 Demographic variables C3001-C3999 Micro-level (survey) data (the CSES Module 3 questionnaire) C4001-C4999 District-level data C5001-C5999 Macro-level data In the Variable Descriptions portion of the codebook, the headers for individual variables are surrounded by two lines of dashes. Variable names do not exceed eight characters in length. Most sections of the codebook can be navigated in the electronic files by searching for the characters ">>>" or ")))" as appropriate. =========================================================================== ))) HOW TO USE THE CSES MODULE 3 DATA FILES =========================================================================== We recommend that PC users create the following directory on their hard drive, and to download their files from this Module 3 release to that location: c:/cses/module3/20151215/ The subdirectory value "20151215" represents the version (release date) of the dataset - this being the 2015, December 15th version of CSES Module 3. This organization method allows users with multiple CSES datasets and/or versions to stay organized and not over-write their other files. Users of other computer types (Macs, Unix, etc.) are recommended to use a similar directory structure to organize their CSES files. The following ZIP files are available to download from the Module 3 download page under the Data Center on the CSES website. All users should download the codebook file: cses3_codebook.zip Contains the three codebook files, including this one, in text format. Users should also download one or more of the following six files, depending on which statistical package(s) they intend to use with the data, and how: cses3_csv.zip Contains a CSV file with variables names as column headers but no additional metadata (for instance, no code labels are included). cses3_syntax.zip Contains a raw data file and syntax statements to read the dataset into SAS, SPSS, and STATA. The instructions for doing so are found in the headers of the syntax files for each statistical package: cses3.sas for SAS, cses3.sps for SPSS, and cses3.do for STATA. This ZIP file also contains optional missing data statements which can be applied to the dataset in SAS, SPSS, or STATA. cses3_r.zip Contains a R Workplace system file (.rdata), with the dataset already prepared and ready to be loaded into R. Missing data statements are not applied. cses3_sas.zip Contains a SAS 7-8 system file (.sas7bdat), with the dataset already prepared and ready to be loaded into SAS. Missing data statements are not applied. cses3_spss.zip Contains a SPSS system file (.sav), with the dataset already prepared and ready to be loaded into SPSS. Missing data statements are not applied. cses3_stata.zip Contains a STATA 13 system file (.dta), with the dataset already prepared and ready to be loaded into STATA. Missing data statements are not applied. =========================================================================== ))) SPECIAL DATA NOTES =========================================================================== --------------------------------------------------------------------------- >>> IDENTIFICATION VARIABLES --------------------------------------------------------------------------- Each record in CSES Module 3 contains a number of identification variables with which the analysis will wish to become familiar. These three variables identify the dataset, version, and DOI: C1001 >>> DATASET C1002 >>> DATASET VERSION C1002_DOI >>> DIGITAL OBJECT IDENTIFIER This variable uniquely identifies an election study across time. It appears in two variations: C1003 >>> ID VARIABLE - ELECTION STUDY (NUMERIC POLITY) C1004 >>> ID VARIABLE - ELECTION STUDY (ALPHABETIC POLITY) This variable uniquely identifies a respondent across time: C1005 >>> ID VARIABLE - RESPONDENT This variable uniquely identifies a polity (country, nation, etc.). It appears in three variations: C1006 >>> ID COMPONENT - POLITY CSES CODE C1006_UN >>> ID COMPONENT - POLITY UN CODE C1006_NAM >>> ID COMPONENT - POLITY NAME Notes on their creation and use of these variables are available in the Variable Descriptions portion of the codebook. --------------------------------------------------------------------------- >>> MISSING DATA --------------------------------------------------------------------------- (1) Researchers should be aware that sometimes there are multiple response categories included in the code "missing" (the code "missing" is usually indicated with a last digit of 9). For some election studies in which we could not distinguish among various answers, the code "missing" may include cases where respondents refused to answer the question, "don't know" responses, and cases where there a particular question went unanswered for other reasons. (2) While CSES guidelines request that the response categories "Refused" and "Don't Know" be volunteered responses, this was not always consistently applied. For instance, sometimes the options were offered explicitly to respondents in mail-back surveys, which do not have the benefit of an interviewer being present. To identify whether the response options were volunteered (or not) in a particular election study, please refer to the original questionnaire provided by the collaborator. The original questionnaires, where available, are accessible from the CSES Module 3 download page under "Data Center" on the CSES website. --------------------------------------------------------------------------- >>> WEIGHTS --------------------------------------------------------------------------- Because of the variance in the sample designs used in the election studies included in this project, the weights provided by the collaborators also vary greatly. They are described in detail in variables C1010-C1014. Analysts are advised to read the weight documentation carefully to ensure that their analyses are weighted appropriately. The original weights provided by the collaborators, where available, are: C1010_1 >>> ORIGINAL WEIGHT: SAMPLE C1010_2 >>> ORIGINAL WEIGHT: DEMOGRAPHIC C1010_3 >>> ORIGINAL WEIGHT: POLITICAL The remainder of the weight variables in the dataset are derivative variables, constructed from the original weights: C1011_1 >>> FACTOR: MEAN OF SAMPLE WEIGHT C1011_2 >>> FACTOR: MEAN OF DEMOGRAPHIC WEIGHT C1011_3 >>> FACTOR: MEAN OF POLITICAL WEIGHT C1012_1 >>> POLITY WEIGHT: SAMPLE C1012_2 >>> POLITY WEIGHT: DEMOGRAPHIC C1012_3 >>> POLITY WEIGHT: POLITICAL C1013 >>> FACTOR: SAMPLE SIZE ADJUSTMENT C1014_1 >>> DATASET WEIGHT: SAMPLE C1014_2 >>> DATASET WEIGHT: DEMOGRAPHIC C1014_3 >>> DATASET WEIGHT: POLITICAL --------------------------------------------------------------------------- >>> FREEDOM STATUS OF ELECTIONS --------------------------------------------------------------------------- The majority of studies that comprise CSES are collected in countries that have free or partly free elections. However, sometimes a collaborator will include the CSES module in a study of a country that is a developing democracy or that is considered not free. If the data collection is judged to be of sufficiently high quality, the study is included in CSES datasets even if the country is considered to be not free. The decision is left to individual users as to whether such countries should be included in their analyses of CSES datasets. The CSES Module 3 dataset includes freedom ratings from Freedom House, whose website as of this dataset release is: http://www.freedomhouse.org/ Freedom House is not affiliated with the CSES project. =========================================================================== ))) CSES MODULE 3 ELECTION STUDIES =========================================================================== --------------------------------------------------------------------------- >>> LIST OF ELECTION STUDIES INCLUDED IN CSES MODULE 3 --------------------------------------------------------------------------- The Full Release of CSES Module 3 contains data from the following 50 election studies (in alphabetic order): Election Study Cases Australia (2007) 1,873 Austria (2008) 1,165 Belarus (2008) 1,000 Brazil (2006) 1,000 Brazil (2010) 2,000 Canada (2008) 4,495 Croatia (2007) 1,004 Chile (2009) 1,200 Czech Republic (2006) 2,002 Czech Republic (2010) 1,857 Denmark (2007) 1,442 Estonia (2011) 1,000 Finland (2007) 1,283 Finland (2011) 1,298 France (2007) 2,000 Germany (2005) 2,018 Germany (2009) 2,095 Greece (2009) 1,022 Hong Kong (2008) 815 Iceland (2007) 1,595 Iceland (2009) 1,385 Ireland (2007) 1,435 Israel (2006) 1,200 Japan (2007) 1,373 Latvia (2010) 1,005 Mexico (2006) 1,591 Mexico (2009) 2,400 Netherlands (2006) 2,359 Netherlands (2010) 2,153 New Zealand (2008) 1,149 Norway (2005) 2,012 Norway (2009) 1,782 Peru (2011) 1,570 Philippines (2010) 1,200 Poland (2005) 2,402 Poland (2007) 1,817 Portugal (2009) 1,316 Romania (2009) 1,403 Slovakia (2010) 1,203 Slovenia (2008) 1,055 South Africa (2009) 1,200 South Korea (2008) 1,000 Spain (2008) 1,204 Sweden (2006) 1,547 Switzerland (2007) 3,164 Taiwan (2008) 1,905 Thailand (2007) 1,990 Turkey (2011) 1,109 United States (2008) 2,102 Uruguay (2009) 968 TOTAL 80,163 For election studies that are embedded in multi-wave panel studies, only those respondents who participated in the wave of the study that included CSES Module 3 are included in the CSES Module 3 dataset. --------------------------------------------------------------------------- >>> OTHER ELECTION STUDIES INCLUDING CSES MODULE 3 --------------------------------------------------------------------------- The following studies collected CSES data during the CSES Module 3 field period, but were unable to be included in the CSES Module 3 Full Release for the listed reasons. We are highlighting the studies here to promote their efforts, and as a service to CSES users who might wish to contact the collaborators directly to inquire about obtaining the data for use in their own analyses. If you are interested in obtaining these data, please contact the listed collaborators directly. CSES is unable to distribute or support the files. - Denmark (2007 Internet Study) CSES collaborators Jřrgen Goul Andersen, Rune Stubager, and Kasper Mřller Hansen ran two election studies in Denmark after the 2007 election. The first election study, a mixed mode survey, included 1,442 completions and appears in the CSES Module 3 Full Release dataset. The second election study, collected on the Internet, does not appear in the CSES Module 3 Full Release dataset. While preliminary research indicated that the web panel might have used probability-based sampling, this could not be verified by the CSES Secretariat staff members with certainty prior to the dissemination of the CSES Module 3 Full Release. The Denmark (2007) Internet Study was run on the Internet panel named Gallup@Forum. The Denmark (2007) Design Report indicates that the Gallup@Forum Internet panel consisted of 45,000 Danes at the age of 15 and above. The Design Report went on to describe that the sample was stratified by sex * age, region, and education, and consisted of 6,100 individuals, of which 2,576 persons completed an interview via the Internet. - Great Britain (2010 Internet Study) The British Election Study (BES), under the direction of Principal Investigators Harold D. Clarke, David Sanders, Marianne C. Stewart, and Paul Whiteley, included CSES Module 3 in a Internet survey run by YouGov after the 2010 election in Great Britain. A total of 927 interviews were collected. As of publication time, the dataset was publicly available from the BES website at address: http://bes2009-10.org/ The BES survey that included CSES Module 3 drew respondents from a panel built using a non-probability sample, not a probability sample, and thus the study was unable to be included in the CSES Module 3 Full Release. =========================================================================== ))) ELECTION SUMMARIES AND GENERAL NOTES =========================================================================== The following section provides: (1) General information about the election after which the CSES Module 3 was administered. In several cases, the election followed the establishment of new electoral rules. In several others, the election marks a dramatic change in government. This information is provided with the intention of alerting the analyst to interesting features of the election. For more details, please refer to the Macro Reports prepared by the collaborators, available on the CSES website. (2) General information about the research and sample designs of the component election studies. For example, in some cases, the CSES module was administered in a later wave of a multi-wave study. Additionally, in several countries, portions of the population were excluded from the sample frame, usually because of geographic isolation. (3) Additional information on survey weights, where available, is provided for some election studies. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - AUSTRALIA (2007) --------------------------------------------------------------------------- (1) On 14 October 2007, Prime Minister John Howard called elections for 24 November. At stake were all 150 seats in the House of Representatives, and 40 of the 76 seats in the Senate. The 2007 elections saw once again a duel between the Liberal Party and the ALP. Prime Minister Howard (Liberal Party) was seeking a fifth consecutive term in office. Prime Minister Howard urged voters to support his party for further economic growth. The Liberal leader insisted that the country should maintain its 1,500 troops in Iraq and 1,000 in Afghanistan as long as needed, and maintain a close alliance with the U.S. The ALP leader argued that 550 combat troops should be withdrawn from Iraq by mid-2008, while more troops should be sent to Afghanistan. Prime Minister Howard continued to oppose the Kyoto Protocol on climate change and its carbon emission targets, arguing they would damage the country's economy. Mr. Rudd promised to ratify the Kyoto Protocol and pledged to cut carbon emissions by 60 per cent by 2050. In 2007, Australia suffered its worst drought in 100 years. Global warming was reportedly a major concern for many citizens. During the election campaign, the Liberal Party was damaged by two incidents. On 7 November, the Reserve Bank raised interest rates to 6.75 per cent, the sixth increase since the 2004 elections, when Prime Minister Howard had promised to keep the interest rate low. Two days before the polling day, further controversy was caused by leaflets circulated in Sydney by volunteers and members of the Liberal Party, which linked the ALP to Muslim extremists. The Prime Minister condemned the act and emphasized he had not authorized the leaflets. The final results gave a major victory to the ALP, which won 83 seats in the House of Representatives, while the Liberal Party took 55. Howard became the first incumbent prime minister to be voted out since Mr. Stanley Bruce in 1929. In the Senate elections, the ALP won 18 seats, controlling 32 seats in all. The Liberal Party won 15 seats and also holds 32 seats in all. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES Module was included in the Australian Post-Election Study, running from November 23, 2007 until March 14, 2008 as a self-completion supplement by mail. Because of this chosen mode of survey, interviewers were not employed. The sample of the survey was restricted to persons aged 18 or older being registered as voters. No social group such as institutionalized persons and military personnel were excluded from sampling. The sampling procedure itself was a stratified systematic random sampling. Electors were drawn from the Commonwealth Electoral Roll by the Australian Electoral Commission, following the close of rolls for the 2007 election. The sample was selected to be proportional to the population, on a state by state basis. The entire electorate was randomly selected, therefore involving no further stages of selection, or any kind of stratification, clustering or quota sampling. On average, selected persons have been contacted three times. (3) The original data, provided by Australia, did not include any kind of weights. Furthermore, due to the sample procedure, weights were not necessary to make the sample representative for the whole Australian population. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - AUSTRIA (2008) --------------------------------------------------------------------------- (1) Elections were held on 28 September 2008 for all the seats in the National Council following premature dissolution of this body on 9 July 2008. Elections to the National Council had previously taken place on 1 October, 2006. The collapse of the coalition government in July 2008 triggered early elections to the 183-member National Council in September, nearly 25 months before they were due. Inflation and high fuel prices reportedly disrupted the country's economy prior to the 2008 elections. Following a reform of the electoral law in 2007, the voting age was lowered from 18 to 16 years, and people over 18 years of age (instead of 19) became eligible to run for elections to the National Council. In the same vein, the term of the National Council was extended from four to five years starting from the new legislature. Both the SPÖ and the ÖVP recorded their worst results since World War II. They won 57 and 51 seats respectively. On the contrary, the FPÖ and the BZÖ increased their strength to 34 and 21 seats respectively. The Greens took the remaining 20 seats. In all 50 women were elected. On 8 October, President Heinz Fischer (SPÖ) asked Mr. Faymann (SPÖ) to form a new government. The newly elected National Council held its first session on 28 October and re-elected Ms. Barbara Prammer (SPÖ) as its Speaker. On 2 December, Mr. Faymann's government comprising the SPÖ and the ÖVP was sworn in by President Fischer. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The sample is meant to be representative of the Austrian Population, persons aged 17 and older, all persons eligible to vote for the national election 2008. The sample was drawn using multiple stratification with a clustered address random procedure: a. Austria was divided into nine provinces proportional to their population (age 17 and above) b. These provinces were then divided in 121 administrative districts, again proportional to their population, c. These districts were finally divided in 9 different town sizes. d. Communities were then randomly selected, and within each community a PAC (post-certified address code) and then 5 addresses, again randomly. e. The Kish Method was used to reach younger people in each household. The sampling frame for stratification is based on census data provided by Statistics Austria; the selection of households is based on the Austrian telephone book (Österreichische Post AG). Every interviewer was given a second address in case the target respondent was not contactable or the address was not correct. Only then could the second address be used. A postcard was sent to potential respondents prior to the study. Note that although the survey organization had been requested to interview only Austrian citizens who are eligible to vote, some 38 non-citizens were first included in the sample. All of those have been removed from the final sample. (3) A demographic sample weight is included that adjusts sample characteristics to those of the population. The sample selection procedure was designed to ensure random sampling at the last stage. Because of non- response the sample is not an exact representation in terms of equal distribution of the known demographic characteristics of the Austrian population. Hence weights are used to adapt to those known distributions. The weighting was computed as follows. Gender x province Gender x age Gender x occupation Gender x education Gender x province x town size Note that the weight variable was not recalculated after the removal of those 38 non-naturalized immigrant respondents who had been sampled only by mistake, also see above. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - BELARUS (2008) --------------------------------------------------------------------------- (1) On 24 June 2008, President Alexander Lukashenko issued a presidential decree calling elections to the House of Representatives on 28 September. A total of 263 candidates and eight parties contested the 2008 elections. The United Democratic Force, the opposition coalition formed during the 2006 presidential elections, remained the main opposition force in 2008. In August 2008, former Hramada leader, Mr. Alexander Kozulin who had been sentenced to five and a half years' imprisonment in 2006 after staging a protest against President Lukashenko was released from prison. In mid August, five UCP candidates in the capital Minsk announced that they would withdraw from the elections, citing obstacles encountered in their campaigning. On 30 August, the BPF, which had seen the candidatures of its two deputy chairmen rejected, announced that the party would withdraw all its candidates before polling day, criticizing the "undemocratic nature" of the elections. The government stated that the electoral process was 'open and transparent'. The candidature of the leader of an unregistered opposition coalition, the European Coalition, Mr. Mikalay Statkevich, was rejected in a Supreme Court ruling of 9 September due to a previous conviction for unauthorized demonstrations against the results of the 2004 parliamentary elections. The Organisation for Security and Co operation in Europe (OSCE) and the Commonwealth of Independent States (CIS) sent 465 and 368 observers respectively to monitor the polls. The OSCE declared that the elections fell short of OSCE commitments for democratic elections. However, it noted some improvements such as the slightly increased access of opposition representatives to election commissions and the decision to allow fairer access for all candidates to prime time television..It nevertheless criticized a "barely visible election campaign", which it deemed to be strictly controlled by the government. The CIS and the Central Election Commission said the elections had been "free and open". Final results gave six seats to the Communist Party and one to the Agrarian Party. In all, 30 outgoing members were returned. Other elected candidates included 32 managers of State owned companies and institutions, and 17 State officials. No opposition members were elected. The UCP rejected the election results. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented between February 2 2009 and February 24 2009, following the September 28 legislative elections. The mode of interview was face to face. The sample frame covered the totality of the population of the Republic of Belarus 18 years or older and did not exclude non-citizens, non-registered voters, institutionalized people or military personnel. The exact percentage of eligible population excluded from the sample frame is not available. Multi-stage stratified sampling was used in the research. The territorial principle of the sample design underlies the choice, in accordance with which the selection is realized through a number of stages (5 in total). At every stage different territorial units are used: larger ones - at the initial stage; the observation unit corresponds to the selection unit. At the first stage the stratification of the primary selection units was realized across all the six administrative territorial district of the Republic of Belarus. Urban settlements are grouped according to the number of dwellers, 4 strata were formed. Rural settlements are grouped according to the regions of each administrative district of the country. Settlements for the survey were selected by probability, proportional to the number of people residing there. In total 60 settlements were included into the sample. At the second stage a method of random sampling without replacement was applied to select the streets in the settlements. At further stages the route modeling method was used to perform the selection. At the third stage, the unit of selection is the house, then at the fourth stage, the unit of selection is the household. At the final stage of the respondent selection the method of the «closest birth date» was applied and a single respondent was selected. Urban settlements and rural administrative districts were used as primary sampling units (PSU), and these were selected by means of random sampling without replacement. The sample design did not involve clustering at any stage, or quota sampling. Substitution was permitted if the respondent who got into the sample refused to take part in the interview, or if after three visits the respondent was not reached. The response rate was 50,5%. The sample contains 2494 households for 1000 valid interviews. (3) There is no weight variable in this election study given that weights are not necessary to make the population under study representative of the general population of Belarus. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - BRAZIL (2006) --------------------------------------------------------------------------- (1) On 1 October 2006, parliamentary elections were held for all 513 seats in the Chamber of Deputies and for 27 seats in the Senate, concurrently with presidential elections. In June 2005, the government was weakened by corruption scandals involving campaign donations by the PT to coalition partners. President da Silva, who had been elected on an anti-corruption platform, publicly apologized to his fellow citizens in August 2005. In the 2006 election campaign, President da Silva argued that his government had helped millions of poor Brazilians to join the middle class, and insisted on the necessary co-existence of social and economic policy. The main opposition parties, the PSDB and the PFL, formed an electoral coalition. The PSDB camp presented more liberal policies that included the promotion of free trade agreements, in particular with the United States. The PFL did not field a presidential candidate, while the PSDB endorsed the former governor of Sao Paulo, Mr. Geraldo Alckmin, who pledged to enhance the country's economy by fighting corruption. The PMDB became the largest party in the Chamber of Deputies, winning 89 seats, followed by the PT with 83 and the PSDB with 66 seats. The PFL lost 33 seats, retaining 65. It nevertheless remained the largest party in the Senate, winning six seats in the 2006 partial elections, to control 18 seats in total. President da Silva was re-elected with more than 60 per cent of the votes in a run-off election on 29 October. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between December 17, 2006 and December 27, 2006. The questionnaire was translated in Portuguese by the national survey team. The sample includes persons aged 16 or older with Brazilian citizenship. The study employs a three stage stratified probability sample of Brazilian adults. In the first stage, 79 primary sampling units (PSU), or municipals, were selected with probability proportional to size (PPS). In the second stage, 100 secondary units (census tracts) were selected PPS within each of the PSUs. In the third stage, the households were selected within census tracts with one adult respondent being selected by quota by age, sex, education level, working/not working population, according to 2003 PNAD (IBGE). Note that the electoral districts in Brazil overlap with state boundaries, with the exception of one: the Federal District. For procedural reasons, five electoral districts in the region North (Acre, Amapa, Rondonia, Roraima) and Center (Tocantins) were left out of the data sampling procedure. According to the National Household Survey PNAD2003 these 5 states represent 2% of the population. (3) A demographic weight is available. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - BRAZIL (2010) --------------------------------------------------------------------------- (1) The 2010 elections were the first to be held since electoral amendments intended to promote women's political participation. Among other things, parties must allocate at least 30 per cent of their parliamentary candidatures to women. At stake were all 513 seats in the Chamber of Deputies and two thirds of the 81 seats in the Senate (54 seats). The parliamentary elections were held in parallel with the presidential elections. In 2010, for the first time since the end of military rule in 1985, President Lula was not a presidential candidate. He was constitutionally barred from seeking a third consecutive term. The international media focused on the presidential election, paying only secondary attention to the parliamentary polls. The PT pledged to continue the "My Home, My Life" (Minha Casa, Minha Vida) programme, which aims to provide up to 3 million new homes to low-income families and single mothers. Ms. Rousseff, who was keen to become the country's first female President, pledged to carry on President Lula's policies in order to continue to provide stability and prosperity to all Brazilians. The PT was challenged by the PSDB. The PSDB promised to lead a pro-business government and increase the minimum wage from R$ 500 (US$ 300) to R$ 600 (US$ 360) starting from 2011. The PSDB's coalition partner, the DEM, endorsed Mr. Indio da Costa (DEM) as Mr. Serra's running mate (in March 2007, the PFL renamed itself The Democrats - DEM). Mr. Costa had co-sponsored the "clean record" (ficha limpa) bill, toughening the eligibility requirements for public office. Candidates who are found guilty of committing electoral violations and crimes involving the use of public funds are barred from running for office for eight years. President Lula signed the bill into law on 4 June 2010. The left-wing parties, including the PT and the PSB, gained 310 of the 513 seats in the Chamber of Deputies and 50 of the 81 seats in the Senate. The opposition parties, including the PSDB and the DEM, took 138 seats in the Chamber of Deputies and 24 seats in the Senate. On 31 October, in the run-off presidential elections, Ms. Rousseff (PT) triumphed over Mr. Serra (PSDB), thereby becoming the first female President of Brazil. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between November 4, 2010 and November 20, 2010. The questionnaire was translated in Portuguese and pre-tested by the national survey team. There were 27 different versions of the questionnaire filled with different names of local politicians (governors, senators) for each state of federation. The sample includes persons aged 16 or older with Brazilian citizenship. The study employs a three stage stratified probability sample of Brazilian adults. In the first stage, 149 primary sampling units (PSU), or municipals, were selected probability proportional to size (PPS). Each PSU consists of a single municipals as defined by IBGE. For the 2009 Projeçăo da Populaçăo, IBGE divided Brazil into 5 census region with 27 states and 5566 municipals. In the second stage, 500 secondary units (census tracts) were selected PPS within each of the PSUs. In the third stage, the households were selected within census tracts with one adult respondent being selected by quota by situaçăo (urban/rural) age, sex, education level, working/not working population, according to 2008 PNAD (IBGE). The interviewer has specific procedures to complete the quotas: in sum, the interviewer enumerates the blocs within census tracts, walks the blocs oriented by hour direction, and at each three households he runs the interview. Inside the household the quotas are applied to choose the respondent. These procedures allow checking the interviews and limit the interviewer to run the interviews within the selected census tract. The sample design included stratification based on administrative division. Brazil is divided in 5 administrative regions. The population distribution is the following, according to 2009 Population Estimation: North: 8,0%; Northeast: 28,0%; Southeast: 42,3%; South: 14,5%; Center-west: 7,2%. (3) There is no weight variable for this election study. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - CANADA (2008): --------------------------------------------------------------------------- (1) On 7 September 2008 Prime Minister Stephen Harper asked Governor General Michaëlle Jean to dissolve Parliament and call new elections to the House of Commons for 14 October. This would be the third elections in four years. According to an amendment to the Elections Act in November 2006 elections were not due until 19 October 2009. However the amendment did not divest the Governor General of the power to dissolve parliament before the end of its term. Prime Minister Harper used this legal loophole to call for early elections in a bid to secure a parliamentary majority arguing that the parliament was "dysfunctional". In all 1 601 candidates including 445 women were vying for the 308 seats at stake. Although no specific issues dominated the political agenda at the time of the dissolution the global financial crisis pushed the economy to the forefront of the election campaign. Major parties presented various tax- reduction plans to boost the economy trying to win over the provinces of Ontario and Québec which account for a total of 181 out of the 308 seats in the House of Commons. The campaign was also dominated by Canada's ongoing involvement in the war in Afghanistan. The final results gave 143 seats to the Conservative Party. The Liberal Party's share dropped to 76. The Bloc Québécois remained the third largest party with 50 seats. The New Democrats increased its strength to 37. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) Interviewing for the campaign study (CPS) began on September 26, 2008. Excepting the first day of calling and Thanksgiving Day (October 10), between 130 and 260 interviews were completed each day until midnight of the evening before the October 14 election. In total 3,257 CPS interviews were completed. The sample for the post-election surveys was comprised of respondents to the CPS. For the 2008 survey this included not only the new 2008 RDD sample (n = 2,451) but also the 2006 panel sample (n = 1,238). At the end of the CPS in 2008, as in 2006, the interviewer ensured that they had a first name or some other identifier (such as respondent's position in the household, e.g., mother). This information, as well as the sex and year of birth of the CPS respondent, and the respondent's telephone number, was recorded on a cover sheet. The interviewer called and asked for the person by name or identifier. If there was any concern about reaching the correct person the interviewer also checked age and gender. Calling for the post-election study (PES) started on October 15 the day after vote and all of the campaign respondents were called back within three days of the vote. After 10 days of calling 50 percent of the PES interviews were completed. By day 20 almost 70 percent of the interviews were completed but mall numbers of interviews were completed until late December 23. In total 3,689 PES interviews were completed of which 2,451 were from the new 2008 CPS and the remaining 1,238 were respondents who were interviewed in the 2006 CPS. Many of the panel respondents were also interviewed in the 2006 PES and a small number were interviewed in 2004 but not 2006. At the end of the post-election survey, respondents were asked to provide their address so they could be sent the mail-back survey. Mail-back information was provided by 76 percent of the PES respondents. The sample for the CPS was designed to represent the adult population of Canada: Canadian citizens 18 years of age or older who speak one of Canada's official languages, English or French, and reside in private homes in the ten Canadian provinces (thus excluding the territories). Because the survey was conducted by telephone, the small proportion of households in Canada without telephones were excluded from the sample population. Residents of old age homes, group homes, and educational and penal institutions were excluded from the sample (about 0.3%). To select individual survey respondents for the CPS, a two-stage probability selection process was utilized. The first stage involved the selection of households by randomly selecting telephone numbers. The second stage of the sample selection process was the random selection of a respondent from the selected household. To be eligible for the interview the household member had to be an adult (18 years of age or older) and a Canadian citizen. If there was more than one eligible person in the household, the person with the next birthday was selected as the survey respondent. The design of the CES included a longitudinal component as the CPS respondents were asked to complete the PES and respondents to the PES were asked to complete the MBS. There was no substitution of respondents for those who declined to participate in the second or third wave of the study. The number of completed interviews, the sample size, for the three studies was 4,323, 3,138 and 1,674 respectively. The response rate to the CPS was 55 percent and the re-interview rate for the PES was 73 percent and 53 percent of the PES respondents (representing 39 percent of the CPS respondents) completed the MBS. (3) In order to produce national estimates it is advisable to correct for both the unequal probabilities of selection at the household stage and the unequal probabilities of selection based on province of residence. The included sample weight (C1010_1) represents the original National Weight, constructed as the product of the household weight and the provincial weight and should be used with the National Sample when national estimates are required. The household weight compensates for the higher probability of households with smaller number of adults to be included in the sample. The Provincial weight compensates for over representation of the eight smaller provinces and a corresponding under-representation reduction in Ontario and Quebec. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - CHILE (2009): --------------------------------------------------------------------------- (1) The 2009 parliamentary and presidential elections were the first to be held since the death of Augusto Pinochet, the country's dictator between 1973 and 1990. The centre-left Coalition for Democracy (CPD) had led the country since 1990. In the Chamber of Deputies, the CC won 58 seats, one more than the CPD. Clean Chile-Vote Happy and the Pactfree Independents took three and two seats respectively. In the Senate, the CC and the CPD took nine seats each out of the 18 seats at stake. However, the CPD remained the largest force in the Senate with 19 seats, the CC holding 16. Mr. Pinochet failed to win the seat he was contesting, receiving only 10 per cent of the votes, far less than the two CC candidates, who were elected with over 30 per cent of the vote each. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Election Study was conducted between December 26th, 2009 and January, 7th, 2010 as a post-election study using face-to-face interviews. The sample frame included citizens eligible to vote, in continental Chile. It is thus meant to be representative of around 95% of the population. The sample was drawn as a random sample but a quota concerning age, sex and active/inactive was applied, using the proportions of the population of the electoral district. The sampling was done in three stages: a) Random selection of Primary Selection Units (of 2000 inhabitants each) from the whole country. B)Random selection of households with routes within the map of the 2000 inhabitants. C) Random selection of the interviewee - the last person to have had birthday (who is at least 18 years old). (3) The study has a multiplicative weight variable designed to match the distribution of sex, education and age in the Chilean population as measured by the latest census. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - CROATIA (2007) --------------------------------------------------------------------------- (1) On 12 October 2007, the parliament voted to dissolve itself in view of elections due in November 2007. President Stjepan Mesic subsequently called elections for 25 November. A total of 3,585 candidates from 251 lists (parties and coalitions) ran for the 2007 elections. The 2007 elections saw once again a duel between Prime Minister Ivo Sanader's HDZ, and the SDP of Mr. Zoran Milanovic. The main issues were economic policy and the country's accession to the European Union (EU) and NATO. Both parties were favourable to accede to the EU before 2010. While the HDZ pledged to accelerate the accession process, the SDP argued the conditions of accession were more important than speed. Prime Minister Sanader advocated a liberal economy with less state involvement. The SDP insisted on more state control over the economy. The HDZ had strong support from Croats in Bosnia and Herzegovina who are eligible to vote. The SDP vowed to strip Bosnian Croats of the right to vote in Croatian elections if it was elected to office. The Organization for Security and Co-operation in Europe (OSCE) praised the elections for being transparently and professionally administered. It declared that they "represented further progress in fully meeting OSCE commitments for democratic elections". The final results gave 66 seats to the ruling HDZ. The SDP came in second with 56 seats. Both the HDZ and SDP leaders claimed the right to form a government. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES was implemented between March 10, 2008 and April 3, 2008 as a post-election study in face-to-face interviews. The sample of origin included 2060 persons aged 18 years and older living in Croatia. The selection process was a two-way stratified multistage sampling. Stratification was performed by region and settlement size (including four groups: up to 2.000 inhabitants, 2.001-10.000 inhabitants, 10.001-100.000 inhabitants, more than 100.000 inhabitants). Within each stratum, settlements were chosen with probability proportionate to size, according to the size of population, based on Croatian Census data of 2001. That implies random selection of settlement with population less than 2000, while the largest cities in all regions are mandatory part of the sample. The household within the chosen settlements was selected using "random starting point method" and "random walk method". Finally, within the household the concert respondent was selected using the Troldahl-Carter- Bryant method. The sample selection process excluded institutionalized persons in health care and elderly care institutes, in religious organizations and so on. Military personnel were excluded as long as they live in military barracks, but not if they live in private households. Moreover, people with visual and hearing disabilities were excluded generally. In total about 0.7% of the overall population (aged 18 upwards) were not part of the sampling frame. The total amount of completed interviews in the sample are 1004 respondents. (3) The Croatian data includes two different weights, to control a) for the demographic marginal (region, settlement, size, sex, age, education) and b) to controls for the political marginal (region, settlement, size, sex, age, education) according to the election results. Both weights were calculated on the basis of the Croatian Census data of 2001. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - CZECH REPUBLIC (2006) --------------------------------------------------------------------------- (1) The 2006 elections were the first parliamentary elections since the country's accession to the European Union (EU) in May 2004. Close to 5,000 candidates and 26 political groups contested the 2006 elections. The pro-EU CSSD formed an electoral coalition with the Communist Party of Bohemia and Moravia (KSCM) pledging to preserve a welfare state policy by raising pensions and the minimum wage while aiming to introduce the euro as the national currency in 2010. The Civic Democratic Party (ODS) led by Mr. Mirek Topolanek formed an electoral coalition with the KDU-CSL and the Greens (SZ). It favoured a more liberal economy while taking a more cautious approach to the euro. It pledged to introduce a flat 15 per cent rate for income tax and VAT and to abolish all other taxes. The elections ended in political deadlock with both coalitions winning exactly 100 seats. The ODS as the largest party with 81 seats was given a mandate to form a government. The newly-elected Chamber of Deputies was adjourned on several occasions due to disputes over the election of its new speaker and amid ongoing discussions about the formation of a new government. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The election study is a post-election study that began on 9th June 2006 and ended on 21th June, 2006. It was conducted face-to-face and was not part of a panel study. As a sampling method, quota sampling was used. Quotas were designed for each region (14 regions) based on age, sex and education, and the size of municipality. If a person refused to take part in interview, another person suitable to the quota was selected. There were around 900 interviewers in the CVVM polling agency all around the country. They were both men and women almost in the same ratio. Age varied in between 18 to 65 years with the average age 38 years. Education structure of interviewers is 80 % secondary school and 20 % university level. The experience with CVVM polling agency varied between 1 to 21 years. Average interviewer worked for CVVM 12 years. Before a person becomes an interviewer of CVVM s/he has to go through training where principles of interviewing and the administration around interviewing (filling the questionnaires, correspondence with public opinion headquarters etc.) are learned. After the training, the person goes through test interviews. Training of interviewers after becoming an interviewer of CVVM is organized every 5 years. There is a training letter for interviewers added to every questionnaire/survey and almost non-stop phone and e-mail consultations. (3) Since the sample is a quota sample, no weights were needed to make it representative of the population. ============================================================================= >>> ELECTION SUMMARIES AND GENERAL NOTES - CZECH REPUBLIC (2010) ============================================================================= (1) In January 2009, the Czech Republic took over the EU Presidency. However, in the middle of its term, on 24 March 2009, the Government lost a vote of non-confidence. It was the first time in the Czech Republic's history that the opposition succeeded in passing a non-confidence motion against the government. On 5 February 2010, President Václav Klaus set elections to the Chamber of Deputies for 28 and 29 May 2010. Early elections required a constitutional amendment, which President Klaus signed into law in September after its adoption by the Senate. However, several members of the Chamber of Deputies opposed the amendment, arguing that they had the right to serve the full four-year term. Later in the same month, the CSSD, the KSCM and the SZ voted against the Chamber's dissolution. Consequently, the elections were held on 28 and 29 May 2010, a few days before the term of outgoing members expired (on 2 June). The mandate of the caretaker government was extended until the new elections. The 2010 elections were held against the backdrop of the Greek and euro-zone crises. The dominant theme during the campaign in the Czech Republic was how to deal with the country's public deficit. The final result of the polls was another stalemate with no party securing a majority in the Chamber of Deputies. Both major parties ended up with fewer seats than in the outgoing legislature. The CSSD came in first with 56 seats (down from 74), narrowly followed by the ODS, which took 53 seats (down from 81). The KSCM took 26 seats. The KDU-CSL failed to win any seats in the Chamber of Deputies for the first time since the Czech Republic was founded in 1993, prompting its leader, Mr. Cyril Svoboda, to resign. The SZ also failed to retain its representation in the Chamber of Deputies. CSSD leader Paroubek took responsibility for his party's defeat and announced his resignation as well. Source: http://www.ipu.org/parline-e/reports/2083_E.htm (last accessed on 08/01/2013) (2) The sample from 2010 was implemented in the same way as the one from Czech Republic (2006). See above. (3) No weights were implemented. The sample is a quota sample. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - DENMARK (2007) --------------------------------------------------------------------------- (1) On 24 October 2007 Prime Minister Anders Fogh Rasmussen called early elections to the 179-member Danish Parliament (Folketing) for 13 November ahead of the official end of the four-year term in February 2009. In the 2007 elections Prime Minister Rasmussen was seeking a third consecutive term in office. A total of 808 candidates representing nine parties and including 12 independents contested the 2007 elections. Tax welfare and immigration were main issues during the election campaign. Prime Minister Rasmussen pledged to implement further tax-cutting measures referring to his government's proposal to lower income tax which was passed in parliament on 23 October. The Social Democrat leader Thorning-Schmidt promised to provide better welfare including higher wages for the lowest- paid public servants. Prime Minister Rasmussen also promised to soften asylum laws proposing that asylum seekers with families be housed outside reception centres. The Social Democratic Party suggested that rejected asylum seekers be allowed to work in Denmark. The New Alliance a new centrist party formed in May 2007 and led by Mr. Naser Khader a Syrian-born Palestinian immigrant also promised to work towards better treatment of refugees. The Danish People's Party led by Ms. Pia Kjaersgaard opposed all initiatives to soften asylum laws. The final results gave the outgoing coalition and its supporters a total of 89 seats. The Social Democratic Party took 45 seats. The New Alliance finally won 2.8 per cent of the votes or five seats. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The 1442 Danish respondents were interviewed in person or through self- administrated interviews (postal and internet). A number of address clusters were drawn from the Ministry of the Interior. These clusters were divided based on population density. Each cluster consisted of 4-12 addresses. A total of approximately 600 addresses were drawn (sampling points). The addresses were merged with a database consisting of 2.5 million landline phone numbers, as well as 2.1 million cellular phone numbers. Addresses where no phone number was found were contacted by CAPI interviewers. Addresses where a phone number was found were contacted by phone in order to recruit a respondent. These phone numbers were contacted up to 16 times at different times during the day and week. The person whose birthday was most recent in household were selected. Substitution within the household was not permitted. When the right respondent was met, the interviewer shortly explained the purpose of the survey and tried to arrange a personal interview. Respondents who did not wish to participate were instead offered to complete the survey via postal or web questionnaire. Respondents who did not complete the questionnaire within the given time frame were given two reminders. Respondents opting for the postal questionnaires were given a reminder after the deadline and 14 days later. Respondents opting for the web questionnaire were given a reminder four days after receiving the questionnaire and eight after the deadline. As the recruiting was an on- going process questionnaires and reminders were sent on an on-going basis. The overall response rate is about 41.6%. About 10% of these completed interviews are verified by a second phone call, to control the authenticity in terms of respondent’s age and gender. (3) The Danish data of 2007 include a sample weight to control for the representativeness of the population under study, taking age, gender, education and the electoral results into account. Researchers should care about the fact that the originally deposited Danish data included a second subsample, based on a web-panel. The web-panel was selected by probability methods and is being considered for inclusion in the upcoming final release of CSES Module 3, but it does not yet appear. However, the sample weight coded in C1010_1 depends on the complete Danish sample, existing of the subsample described before, as well as of the excluded web-panel. Consequently, the employed weight in C1010_1 might slightly be biased. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - ESTONIA (2011) --------------------------------------------------------------------------- (1) The March 2011 elections were the first to be held following the global economic crisis that began in 2008. In 2011, the Prime Minister's coalition government was challenged by the Centre Party, the SDE, the Estonian Greens and the Estonian People's Union. The country was severely hit by the global economic crisis, with gross domestic product (GDP) shrinking by 14 per cent in 2009. The country's economy showed signs of recovery in 2010. The leading parties focused on taxation. The two parties in the outgoing government pledged to maintain a single rate for value-added tax (VAT), income tax and corporate tax. The opposition parties promised to introduce a progressive taxation system, reducing the VAT on food. The SDE promised to levy a 26 per-cent income tax on persons earning more than 1,000 Euros per month while maintaining it at 21 per cent for persons earning less. The SDE also promised to triple the child allowance for all under-19s and guarantee each child a place in nursery, in addition to providing free education. The Centre Party argued that the VAT increase and budget cuts had weakened the economy and affected the financial situation of many Estonians. The parties in the outgoing government increased their presence, together securing a majority in the newly elected parliament. Prime Minister Ansip's Reform Party won 33 seats, up from 31; and the IRL took 23 seats, up from 19. The SDE also increased its seats to 19, up from 10, while the Centre Party fell from 29 to 26 seats. The Estonian People's Union and the Estonian Greens failed to surpass the 5 per-cent threshold required to win parliamentary representation. Source: . 2012/01/10. (2) The Estonian sample was a stratified random sample. In the first stage, sampling points were chosen in 15 counties, determined by the county’s share of the total eligible population according to information from the Population Register of the Ministry of Interior. The interviews were conducted by using the random route method. On average five interviews were achieved per sampling point. A household in every third house or apartment was chosen and the last birthday rule was used to pick a respondent from the household. Three visits were made before replacing a household with another in case the respondent was not available. Substituting within the household was not allowed. In sum, 2513 households were sampled, out of which 1453 households had been eligible for interviewing. According to the 1000 completed questionnaires, the Estonian response rate for 2011 was about 68.8%. All interviews were held between March, 7th and April, 18th 2011, as face-to-face interviews. On average, the interviewers were 52 years old, with an average interviewer experience of 4.5 years. All of them had secondary education (about 50%) or higher. Roughly 79% of interviewers conducted interviews in Estonian and 21% in Russian. The interviewers were trained individually by the Saar Poll company, an Estonian social and market research centre. In addition, interviewers received written instructions. About 10% of the completed interviews were verified through mail, by a shortened questionnaire. (3) Data on Estonia 2011 include a socio-demographic weight (C1010_2) to make the sample representative of the population being studied. The constructed weight is based on data from the Population Register of the Ministry of Interior, processed by AS Andmevara. It combines the overall distributions of gender, ethnicity, region, and rural-urban nature of settlement. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - FINLAND (2007) --------------------------------------------------------------------------- (1) The March 2007 elections came on the heels of the country's presidency of the European Union from July to December 2006. As a result, the election campaign started late compared to previous ones. In the 2007 elections, 1,205 men and 799 women were vying for the 200 seats in the parliament. As in previous elections, employment and taxation were the main issues in the 2007 elections. Prime Minister Vanhanen pledged to reduce VAT on food. The SDP said the VAT reduction must apply to ordinary goods so as to help people with the most modest and average incomes. The KOK pledged to cut the income tax of people with small and medium incomes. The biggest winner in the 2007 elections was the KOK, which gained ten more seats, winning 50 in all. Former KOK leader Sauli Niinistö was re-elected with over 60,000 votes or 10 per cent of the party's votes nationwide. The KESK remained the largest party in the parliament by winning 51 seats, although it lost four. The SDP lost eight seats, winning 45, reportedly suffering from the low turnout. The SDP lost its position as the country's second largest party for the first time since 1962. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between March 20, 2007 and May 20, 2007. The questionnaire was translated into Finnish and Swedish. The sample was meant to be representative of Finnish citizens aged 18 or over living in Finland (excluding the Ĺland Islands, which accounts for 0.48% of the population). There was a different sampling procedure for Finnish-speaking Finns and Swedish-speaking Finns. For the former, a quota sample was used based on gender, age, and province. The interviews were conducted by using the random walk method, in which the interviewers made five subsequent interviews proceeding from a randomly selected starting point. In cities, the interviewers were given specific starting addresses, whereas in rural areas they were allowed to select the starting points by themselves. For the Swedish-speaking Finns a quota sample was used based on gender, age, and electoral district, which includes only the districts with a significant population of Swedish-speaking Finns, i.e. Helsinki, Uusimaa, Varsinais-Suomi, and Vaasa. Furthermore, in the selected districts, only the municipalities with a significant population of Swedish-speaking Finns were included in the study. The quotas for the interviews of Swedish- speaking Finns were formed by ordering an especially large random sample from the population register. The sample consisted of enfranchised Swedish- speaking people in the selected municipalities, and it was ten times larger than the desired number of respondents. The original sample contained 1422 respondents which included an oversample (209 cases) of the Swedish-speaking Finns. In the final dataset, the number of cases is 1283, of which 5.5 % (71 cases) are Swedish-speaking Finns. The oversample was corrected by randomly dropping out 138 Swedish-speaking Finns. Average response rate was 24.25 % of all the contacted households, which includes households where no one was reached, households where no one belonged to the target group, and the households where none of the respondents matched the available control quotas. Note that the electoral district Ĺland was excluded from sampling. (3) There is no weight variable for this election study. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - FINLAND (2011) --------------------------------------------------------------------------- (1) Following amendments to the Electoral Act, which came into force in 2010, the 2011 elections were held in April instead of March. As a result of the global economic crisis, the country's gross domestic product (GDP) shrunk by 8.1 per cent in 2009 but grew by 3.1 per cent in 2010. In June 2010, Mr. Vanhanen stepped down as KESK leader and Prime Minister following funding scandals in his party. Ms. Mari Kiviniemi took over both posts, becoming the first woman Prime Minister of Finland. She led the same four-party coalition government until the 2011 elections. On 16 February 2011, the outgoing legislature voted to approve several constitutional amendments, with 144 votes in favour and 29 against. In order to take effect, however, they needed to be approved by the new legislature, to be elected in April. One of the amendments specifically states that Finland is a member of the European Union (EU). Another stipulates that any disagreement between the country's President and the Government must be resolved by the Parliament. Parliament was dissolved on 15 March in view of elections in which 2,315 candidates, including 903 women, were running. Major election issues at the beginning of the campaign included taxation and compulsory teaching of Swedish. The debates on taxation and the Swedish language were pushed to the backburner after 6 April, when Portugal officially requested a financial bailout from the EU. Finland's financial contribution to the EU became the focus of the election campaign. Parties in the outgoing government said they would support the bailout for Portugal. Opposition parties were more reluctant. The KOK came in first with 44 seats, followed by the SDP with 42. The True Finns more than quadrupled its share to 39. Prime Minister Kiviniemi's KESK came in fourth with 35 seats. The Left Alliance and the Green League took 14 and 10 seats respectively. The Swedish People's Party and the KD took nine and six seats respectively. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between April 18 and May 28, 2011. The questionnaire was translated into Finnish and Swedish. The sample was meant to be representative of Finnish citizens aged 18 or over living in Finland (excluding the Ĺland Islands, which accounts for 0.49% of the population). There was a different sampling procedure for Finnish-speaking Finns and Swedish-speaking Finns. For the former, a quota sample was used, based on gender, age, and the regional distribution of the target population. The interviews were conducted by using the random walk method, in which the interviewers made five subsequent interviews proceeding from a randomly selected starting point. For the Swedish-speaking Finns a quota sample was used, based on electoral districts with a significant population of Swedish-speaking Finns, i.e. Helsinki, Uusimaa, Finaldn Propoer, and Vaasa. Furthermore, in the selected districts, only the municipalities with a significant population of Swedish-speaking Finns were included in the study, based on information Finlandssvenskarna 2009 Report, compiled by The Swedish Assembly of Finland (Folktinget). In the first stage, a random sample of 1800 Swedish- speaking persons was drawn from the Finnish Population Register Centre. In the second stage, quotas were formed, based on the age, gender and constituency-specific distribution in the random sample. The final sample contained 1298 respondents, including 1223 Finnish- speaking and 75 Swedish-speaking Finns. The average response rate was 24.25% of all the contacted households, including households where no one was reached, households where no one belonged to the target group, and households where none of the respondents matched the available control quotas. (3) The Finnish data include a political weight. It controls for the electoral outcome of the parliament election in 2011, based on respondents vote choice in C3023_LH_PL. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - FRANCE (2007) --------------------------------------------------------------------------- (1) The 2007 elections to the National Assembly followed the election as President of the Republic of Mr. Nicolas Sarkozy, of the Union for a Popular Movement (UMP, the ruling centre-right party). In May, Mr. Sarkozy defeated Ms. Ségolčne Royal of the main opposition Socialist Party (PS) in a run-off election, and pledged to take the country into a new era, both economically and socially. The UMP held 359 seats in the outgoing legislature, while the PS had 149. The June parliamentary elections saw once again a duel between the UMP and the PS. President Sarkozy insisted that he needed a majority in the National Assembly to carry out his presidential programme. Pre-election polls predicted a major victory for the UMP. In all, 7,639 candidates contested 577 seats in the National Assembly. On 10 June, 60.42 per cent of registered voters turned out at the first round of the polls, the lowest figure since 1958. 110 candidates were elected in the first round, compared to 58 in the 2002 elections. The UMP and its allies took 98 of those 110 seats, securing 39.54 per cent of the valid votes while the PS obtained 24.73 per cent of the votes. Twelve women were elected. Fiscal issues took centre stage during the period between the first and second rounds. Prime Minister François Fillon came in for scathing criticism from the PS over the government's supposed plans to introduce a "social VAT" (value-added tax). On 17 June, the second round of elections for the remaining 467 seats involved candidates who had obtained over 12.5 per cent of the registered vote in the first round. Nearly 60 per cent of registered voters turned out at the polls. The UMP fared less well than opinion polls had predicted. The Energy and Environment Minister and former prime minister, Alain Juppé (UMP) was narrowly defeated and subsequently resigned as minister. The final results nevertheless confirmed a clear victory for the UMP, giving it a total of 313 of the 577 seats. The PS took 186 seats. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented between June 18 2007, and July 7 2007 September 13, 2005 and December 20, 2005, after the elections in telephone interviews. The sample consists of 2000 persons aged 18-96 years. The sample is selected from the mational metropolitan population over 18, registered on electoral lists; overseas territories were excluded from target population. Institutionalized people and military personnel were included in the sampling frame. Official statistics register 87.6% of households as having a phone (only land line phones), which excludes some 15% of households from the sampling frame. Furthermore unlisted numbers were not included, which represents some 16% of households with a land line phone. It was decided not to include unlisted numbers so that sampling frames make it possible to locate precisely households (so that their precise vote is recorded). The sampling procedure consists of a stratified three stage probability sampling. Stratification: Table with 7 ZEAT areas and 6 agglomeration classes, i.e. 42 cells. Each cell contains the corresponding population size. Stage 1: Allocation of 200 Primary Sampling Units (districts)) to the defined cells using Cox Method of controlled rounding. Selection of PSUs from the cells according to the allocation proportional to population size. Stage 2: Selection of a fixed number (10) of households from the selected PSUs via a random draw in phone book. Stage 3: Selection of an individual within a household via Last-Birthday- Method. The primary sampling units (PSU) were districts (circonscriptions) so that each PSU has an homogenous political offer, selected by the Cox Method of controlled rounding (randomly selected). Each unit has been randomly drawn through adapted techniques, without clustering, substitution, or quota sampling at any stage. Participants did not receive incentives for their participation. The telephone interviewers (mainly student and young educated adults from 20 to 30 -25.4 on average), had, on average one year of experience. Interviewer training was conducted in a half a day with thorough explanation of sampling strategy and method and overview of the questionnaire. Interviewers received special training in refusal conversion and reissue of refusals to and other interviewer. The average contact per household for the entire sample was 8.2 over 15 days. The response rate was 34.2%. The total number of household in the sample was 10469, out of which, 5849 valid households, for a total of 2000 completed interviews. Weights are necessary to make the sample representative for two reasons: a) unequal probabilities of selection, especially due to random draw in households of unequal size, and b) a posteriori observation of bias on a number of selected variables. (3) Weights: The French data of origin includes for weights: C1010_1 (design weight) refers to an unequal probabilities of selection and is defined as the product of the probability of selection of i-th observation in its PSU and the corresponding probability of selecting this observation in its household. C1010_2 corrects for a selections bias bases on gender, age, education, occupation and household size. All these variables have been considered in their French definition, which do not match international classifications. Existence of bias after weighting depends on these various definitions of categories. C1010_3 is due to control for political skewness, computed to correct for a bias on aggregate electoral results of legislative elections. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - GERMANY (2005) --------------------------------------------------------------------------- (1) Following his Social Democratic Party's heavy defeat in local elections in May 2005, Chancellor Gerhard Schröder orchestrated a negative vote of confidence in parliament on 1 July to pave the way for new parliamentary elections. He insisted he needed a fresh mandate in order to continue his reforms to revive the German economy. On 21 July 2005, President Horst Köhler announced the dissolution of the Bundestag and called early elections for 18 September 2005, one year ahead of schedule. The election campaign focused both on domestic and foreign policy issues. Mr. Schröder came to power in 1998 promising to lower the unemployment rate. His recent reforms, known as Hartz IV, which aimed to reduce unemployment and welfare benefits in order to make the labour market more flexible, had reportedly produced little impact and frustrated both the unemployed and employers. His rival, Ms. Merkel, argued that a faster pace of reform was necessary. Her programmes included a two per cent VAT increase to finance a reduction in social security costs for employers. The two leaders also held contrasting views over Turkish accession to the European Union (EU). While Chancellor Schröder insisted that Turkish membership would ensure long-term security in Europe, Ms. Merkel firmly opposed Turkey joining the EU and proposed a "privileged partnership" instead. Two major turning points in the campaign seem to have been the televised debate between the two leaders on 4 September and Ms. Merkel's appointment of Mr. Paul Kirchof as economic adviser. Mr. Kirchof, known as a radical thinker, advocated a simplification of the tax system and, in particular, a flat 25 per-cent income tax for all, which reportedly split the CDU supporters. His proposals were also criticized by the CSU. Official results, excluding Dresden, showed that neither the CDU nor the SPD had secured enough seats to easily form a ruling coalition. The CDU/CSU coalition won 225 seats, followed closely by the SPD with 222 seats. Voting in the 160th district in Dresden was held separately on 2 October due to the death of a candidate. The CDU won the seat, bringing the total seats for the CDU/CSU coalition to 226. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The German data for CSES Module 3 was realized by telephone interviews, focusing on the Bundestag elections of September 18, 2005. The interviews were conducted by telephone, starting on September 21. The new Laender (East Germany) were over-sampled. The survey includes 1014 respondents from East Germany and 1004 from West Germany. The sample can be analyzed either separately for East and West, or, if appropriately weighted, for Germany, as a whole. Interviewing period was completed at November 05, 2005 The sample is a single-stage random household sample. An initial sample was drawn from the Infratest Telephone Household Master Sample (ITMS), which comprises a multi-stratified, largely unclustered sample that is distributed in proportion to the number of private households in micro- cells, thereby compensating for regional or local differences in the telephone density of households. The estimated percentage of households without a telephone for the whole of Germany is about one percent. Institutionalized persons (about four percent) are not included in the sample. The existing sample frame covers all listed and unlisted numbers in Germany. To be interviewed, respondents must be at least 18 years old and German citizens. Finally, households were designated non-sample after 12 unsuccessful callbacks. On average, the interviewer were 27 years old, with 22 month of corresponding experience. Interviewers were trained in two seasons, according to the questionnaire and the type of questions, as well as to dialoguing and argumentation. The response rate of completed interviews was 65.6%. (3) Due to the over-sampling of East Germany, as explained before, the use of weights is necessary to study the German sample as a whole (weight provided). Furthermore a weight is needed to compensate differences in the distributions of relevant socio-demographic variables in the sample due to periodical population shifts. According to the federal structure of Germany, this weight, based on age groups and sex, corrects for the population within each region. Users, employing the original weights should take care about the fact that the weight to control the oversampling of East Germany also controls for the skewness of socio-demographics with each of the federal states. Consequently, a combination of both weights is not necessary. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - GERMANY (2009) --------------------------------------------------------------------------- (1) The 2009 parliamentary elections were held against the backdrop of the global economic crisis, which severely affected the country's automobile industry. In September 2008, the government announced a rescue package involving 4.5 billion Euros in loans and credit guarantees. The CDU and the SPD disagreed on many issues. In the 2009 elections, Chancellor Merkel's CDU sought to win more seats so as to form a new coalition government with the Free Democratic Party (FDP). The FDP, led by Mr. Guido Westerwelle, is known for its pro-business policies based on the free market economy. In July 2009, the CSU, led by Mr. Horst Seehofer, opposed the country's ratification of the Lisbon Treaty, aimed at improving the decision-making process of the European Union (EU). The CDU announced that it intended to maintain its partnership with the CSU after the 2009 elections. The CDU/CSU was challenged by Mr. Franz Müntefering's SPD. The major parties focused on the economy and nuclear energy in the 2009 election campaign. Chancellor Merkel argued that a CDU/CSU-FDP coalition was the only way to guarantee economic stability in Germany. The CDU pledged to revive the country's economy by creating more jobs and combating unemployment. It also promised tax cuts and greater investment in education. The FDP promised to reduce the State's involvement in economic policy and opposed the introduction of a minimum wage proposed by the SPD. The 2009 elections resulted in the creation of a record 24 overhang seats, up from 16 in the 2005 elections, bringing the total number of seats in the new legislature to 622. The final results gave 194 seats to the CDU, 14 more than it had in the 2005 elections. The CSU took 45 seats. The CDU/CSU's future coalition partner, the FDP, won 93, up from 61, giving Ms. Merkel's camp a total of 332 seats. In its worst showing since the end of World War II, the SPD lost 76 seats, dropping to 146. Its rival, the Left Party, won 76 seats, 22 more than in the outgoing legislature. The Green Party took 68, 17 more than in 2005. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The German data for CSES Module 3 was conducted as face to face interviews, focusing on the Bundestag elections of September 27, 2009. The interviews were held face to face, for a period running from September 29th to November 23rd, 2009. The new Laender (East Germany) were over-sampled. The survey includes 1569 respondents from West and 526 respondents from West Germany. The sample can be analyzed either separately for East and West, or, if appropriately weighted, for Germany, as a whole, explained in section 3), below. The sample was applied on the ADM-Design. The ADM-Design is a three-step random-design of face to face area sampling, which includes the whole populated area of Germany. It is based on a municipality -structure, which is lodged with data of communal statistics, and with the regional divisions prepared for the navigation systems. On the base of this data 53,000 areas are electronically defined, which include on average 700 private households. The Sampling Points and the Households to be interviewed have already been extracted, using the Adress-Random-Methode, for the Pre- Election Study Component of the German Longitudinal Election Study (GLES) of the GESIS. The institute extracted by chance twelve households out of the remaining addresses and transmitted them to the interviewer. The interviewer had to contact all the twelve addresses and should absolve at least six interviews within these addresses. The probability of interviewing in the post-election study was independent of the pre-election one. Households were designated as non-sample if they were defined as non- residential sample point, including vacant housing units, if all members of the household were ineligible to vote, including language problems, and if the household was reached after at least five contacts. To be interviewed, respondents must be at least 16 years old and German citizens. Institutionalized persons (about four percent of the eligible population) were excluded of the sample frame. However, interviews of respondents younger than 18 years had been deleted afterwards. The response rate of completed interviews was 51.4%, based on the number of valid households. (3) Due to the over-sampling of East Germany, as explained before, the use of weights is necessary to study the German sample as a whole (weight provided). Furthermore a weight is needed to compensate differences in the distributions of relevant socio-demographic variables in the sample due to periodical population shifts. According to the federal structure of Germany, this weight, based on age groups and sex, corrects for the population within each region. Users, employing the original weights should note that the weight to control the oversampling of East Germany also controls for the skewness of socio-demographics with each of the federal states. Consequently, a combination of both weights is not necessary. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - GREECE (2009) --------------------------------------------------------------------------- (1) On 3 September 2009, Prime Minister Kostas Karamanlis asked President Karolos Papoulias to dissolve Parliament with a view to holding early elections, arguing that he needed a new mandate to help steer the country out of the economic crisis. On 7 September, President Papoulias dissolved Parliament and called elections for 4 October, nearly two years before they were constitutionally due. The 2009 elections were held against the backdrop of the global economic crisis. The unemployment rate reached 8.5 per cent in May 2009. The government provided 28 billion euros in bailout packages to banks and privatized several public-sector companies in a bid to boost investment and create jobs. The country's public debt was expected to exceed 100 per cent of GDP in 2009. The budget deficit was expected to rise to 6.2 per cent of GDP in 2009 and 7.3 per cent in 2010. The European Union (EU) had set 2010 as the deadline for Greece to reduce its budget deficit. The 2009 elections once again saw a duel between the ND and the PASOK. Both the ND and the PASOK underscored the need to tackle the economic crisis, albeit with different approaches. The two parties also differed on immigration policy. The final results gave a resounding victory to the PASOK, which took 160 seats. The ND followed with 91 seats, losing 61. Mr. Karamanlis announced that he would step down as the leader of the ND. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Greek survey was realized as a post-election study by telephone interviews. To become part of the sample, respondents had to have the Greek citizenship, as well as the right to vote in national parliamentary elections. Finally, the sample was restricted to respondents aged 18 to 89 years, excluding politicians, political office persons and candidates. The sampling process based on a random selection of geographical areas, built on the total population possessing a phone, stratified by Attica, Thessaloniki and the rest of Greece. Sample was based on the national telephone database, whereas about 17% of the Greek population was excluded due to not having a telephone. Within the geographical units quota sampling was used, to control for the overall distribution of sex and age, according to the Greek Census data of 2001. For further information on the sampling process, as well as on the overall distribution of age and sex in the Greek society, we advise users to have a look at the Greek Design Report, available at the CSES-website. The interviewers were between 22 and 43 years old, with an average interview experience of about 3 years, referring to social and political surveys. About 80% of the interviewers were university students. The interviewers' course of instruction included 12 hours of training for beginners, a two hour briefing for experienced, as well as a written manual for all interviewers. The interviews took place between December 10th and December 19th, 2009. Each of the sampled households was contacted only once. In sum, out of the 7,812 households 1,022 interviews were successfully completed. About one third of these interviews have been verified due to quality control purposes. (3) There is no demographic weight variable for this election study. Given that the Greek sample was based on quotas (see former section), weights are not necessary to make the population under study representative . Instead, a political weight is included in C1010_3 to control for political skewness, computed to correct for a bias on aggregate electoral results of the Greek legislative election of 2009. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - HONG KONG (2008) --------------------------------------------------------------------------- (1) Election summary not available.. For the information about the conduct of the election see: http://www.eac.gov.hk/en/legco/2008lce_report1.htm. (2) The CSES questionnaire was implemented as a single telephone study. The sample population was defined as registered voters, depending on permanently citizens of Hong Kong, aged 18 or older. In contrast, an estimated amount of about one percent of the sample population was excluded due to not having a telephone. The same is true for military persons, who are Chinese soldiers and hence no registered voters of Hong Kong. The sampling process was a two stage random sample, based on the Hong Kong Telephone Directory. In a first stage, the last two digits of each phone number were replaced at haphazard. In a second stage, individual respondents were randomly selected from all eligible voters within a given household. On average, each of the sampled households had been contacted 1.96 times. The maximal number of contacts had been five, at different days and day times, before a household was defined as non-sample. Moreover, households, which refused in the first contact, were called a second time at a different day. In total, the sample included 17,504 first stage units, out of which 16,154 were at least of an unknown validity. From the remaining 1,350 households another 535 units disappeared due to non-contact, refusals or break-offs and partial interviews. All in all, 815 interviews were successfully completed, referring to a responds rate of 60.4%, based on the amount of valid telephone numbers. Interviews were held by undergraduate students of the Lingnan University, who had received basic training in surveying and interviewing. Furthermore, the interviewers were trained according to basic skills of survey interviews and in how to conduct telephone surveys. Finally all of them had accumulated experiences by participating in other telephone interviews prior to the CSES study. (3) A sample weight is included to correct for skweness in the distribution age, sex, and the region of residence, according to the sampling process. As a benchmark, the "Age and Sex profile of registered electors by Legislative Council Constituencies in 2010" has been used, available at the website of the Hong Kong Registration and Electoral Office. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - ICELAND (2007) --------------------------------------------------------------------------- (1) The 12 May 2007 elections took place amid a heated debate on whether the country should prioritize economic growth or the environment, crystallized by a proposal to build new dams and aluminum smelters. The IP has dominated Icelandic politics since the country gained independence from Denmark in 1944. Since 1995, it has led a coalition government with the Progressive Party (PP). The PP pledged to continue corporate tax cuts in order to maintain a favorable environment for Icelandic business. The main opposition parties promised to stop the smelters project until appropriate studies on their environmental impact were carried out. The final results gave 25 seats to the IP. However, the PP took only seven (down from 12 seats in the 2003 elections), giving a majority of just one seat to the outgoing coalition. The Social Democratic Alliance won 18, losing two; while the Left-Green Alliance gained an extra four seats, winning nine in total. The Liberal Party kept its four seats, while the Icelandic Movement did not win any seats. On 17 May, Prime Minister Haarde and the PP leader Mr. Sigurdsson, announced that they would discontinue their coalition following the PP's poor election results. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The sample is meant to be representative of all voters in Iceland. Eligible voters are citizens from 18 years onwards. Institutionalized persons were excluded from the sample frame. In 2007 they were around 0.02 % of persons with a permanent residency in Iceland (60 / 307.672). The information about the number of prisoners that were Icelandic citizens comes from the Institute for Icelandic Prisons (i.icelandic; Fangelsismálastofnun Íslands). The total number, 307,672 is from Statistics Iceland and includes all Icelandic citizens and foreign citizens that had residency in Iceland on January 1, 2007. The simple random sample was drawn from the National registry, and the primary sampling unit was the individual. SPSS was used to draw a random sample from the population. 7.7% of the total eligible population were excluded from the sample. These are persons who had requested that their name would not be included in any sample drawn from the National registry according to information from Statistics Iceland. Interviews were conducted by telephone. It is estimated that less than 1% of the populations does not own a phone or mobile phone. According to the Post and Telecom Administration in Iceland around 96% have a landline. (3) The data set includes a weight variable (C1010_2) for age, as e.g. proportionally fewer in the younger age groups replied to the survey compared to the distribution of age in the population according to information from Statistics Iceland. The weight variable was calculated by dividing the proportion of each age according in the population with the proportion of age in the sample. An example for the sample of 18 years old respondents: The total number of persons in the population at the age of 18 to 80 were 219,875 on January 1, 2007. Of them 4,714 were 18 years old (year of birth=1999), or 2.1% of the total. In the sample, 1,8% were 18 years old (year of birth =1999, N=1.595). The weight for 18 years old is therefore 2.1/1.8=1.22 --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - ICELAND (2009) --------------------------------------------------------------------------- (1) The April 2009 elections were held against the backdrop of the global financial crisis, which brought down the Government in January 2009. They were constitutionally due by May 2011. The financial crisis that erupted in the United States in 2008 severely damaged Iceland's economy. In October 2008, the country's three main banks collapsed within the space of a week. The national debt soared to ten times the gross domestic product. Despite emergency aid of nearly US$ 1.4 billion under the International Monetary Fund (IMF) bailout plan a nd an additional loan of US$ 2.5 billion from its Nordic neighbours (Finland, Norway, Denmark and Sweden), the country's currency, the Krona, continued to plummet and unemployment and inflation spiralled. The Prime Minister tried in vain to save the situation by proposing early elections in May but ultimately announced that he would step down on 26 January. The main contenders were the SDA, the LGM, the IP, the PP and the Liberal Party (which took four seats in 2004). Two new parties formed after the financial crisis - the Civic Movement (CM) and the Democracy Movement (DC) - were also vying for seats. Although the SDA and the LGM said they would form a new coalition government, they disagreed on a key issue, the country's accession to the European Union (EU). The other parties were also divided over EU membership. The SDA came in first with 20 seats; its coalition partner, the LGM, took 14. The IP and the PP took 16 and nine seats respectively. The CM won four while the LP and the DC failed to win parliamentary representation. The elections also resulted in a high turnover of members, with 27 candidates being elected to parliament for the first time. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The sample is meant to be representative of all voters in Iceland. Eligible voters are citizens from 18 years onwards. Institutionalized persons were excluded from the sample frame. In 2009 they were around 0,003 % of persons with a permanent residency in Iceland (80 / 319.368). The information about the number of prisoners that were Icelandic citizens comes from the Institute for Icelandic Prisons (i.icelandic; Fangelsismálastofnun Íslands). The total number, 319,368 is from Statistics Iceland and includes all Icelandic citizens and foreign citizens that had residency in Iceland January 1, 2009. The simple random sample was drawn from the National registry, and the primary sampling unit was individuals. Excel was used to draw a random sample from the population. Interviews were conducted by telephone. It is estimated that less than 1% of the populations does not own a phone or mobile phone according to the Post and Telecom Administration 96% have access to a landline. Approximately 25.2% of the total eligible population were excluded from the sample. These were 7.5% (24,024 / 307,672) persons who had requested that their name would not be included in any sample drawn from the National registry according to information from Statistics Iceland. Another around 17.7/ (N=2,600; sample) were excluded because their phone numbers were not listed. Only some unlisted numbers were looked up using home address and surnames of respondents. (3) The data set includes a weight variable (C1010_2) for age, as e.g. proportionally fewer in the younger age groups replied to the survey compared to the distribution of age in the population according to information from Statistics Iceland. The weight variable was calculated by dividing the proportion of each age according in the population with the proportion of age in the sample. The total number of person in the population on the age from 18 to 80 were 229,659, January 1, 2009. Of them 4,943 were 18 years old (year of birth=2001), or 2.2% of the total. In the sample, 0.5% were 18 (year of birth=1999, N=1,385). The weight for 18 years old is therefore 2.2/0.5=4.26 --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - IRELAND (2007) --------------------------------------------------------------------------- (1) On 29 April 2007, Prime Minister Bertie Ahern asked the country's President, Mary McAleese, to dissolve the parliament, paving the way for elections to the House of Representatives on 24 May. A total of 466 candidates ran for the 2007 elections, in which Mr. Ahern sought a third term in office. The country enjoyed strong economic growth under the coalition government. Mr. Ahern, who had played a major role in political reconciliation in Northern Ireland, helped broker another power-sharing agreement in early May 2007. He pledged tax cuts, free health screening, and higher pensions. However, he suffered a setback when news leaked that he had received money from a friend to renovate a house in Dublin, which he subsequently bought. Fianna Fáil's partner, the Progressive Democrats, was led by Justice Minister Michael McDowell. During the election campaign, Mr. McDowell was faced a scandal involving the country's most secure prison, from which a convicted armed robber had phoned a call-in TV show using a mobile phone. Further investigation revealed the presence of illegal drugs and smuggled mobile phones within the prison, which raised public concern. Fine Gael, led by Mr. Enda Kenny, and the Labour Party of Mr. Pat Rabbitte, formed a centre-left alliance. Both parties criticized the government for failing to improve public services. Fine Gael presented a platform entitled, promising improved health services, more resources for tackling crime, better childcare services, and fairer taxes. The coalition government remained the largest group in the House of Representatives by winning a total of 80 seats (78 went to Fianna Fáil and two to the Progressive Democrats), but failed to win an absolute majority. The opposition Fine Gael gained 20 more seats, winning 51, with the Labour Party taking 20 seats. The Greens took six seats, while Sinn Fein took four. The five remaining seats went to independent candidates. The new House of Representatives had 41 first-time MPs. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) For the 2007 survey, 2135 respondents who had responded in 2002 were selected (excluding only those who had died or who had moved in the intervening years and no follow-up address was available). Interviews were obtained from 1013 people. A short mail questionnaire was sent to non-respondents and this succeeded in getting 130 replies. 43 percent of interviews were completed by the end of June, a month after the election, with 88 percent completed by the end of July and 95 percent by the end of August. In addition, a supplementary three-stage sample was selected from the Geo-Directory (a national listing of all addresses). At the first stage, addresses were clustered by electoral division, with ED’s aggregated to give a minimum population of 1,000 addresses. One cluster was selected at random within each of 39 Constituencies. At the second stage, addresses were selected at random from each cluster (total of 630 addresses). At the third stage, a respondent was selected within the household so as to over-represent adults who would not be included in the follow-up sample. The rule was as follows: If the household contains one or more person age 18-23 who is registered to vote, use the next birthday rule to select the respondent from this group. If the household contains no adult age 18 to 23, select the respondent with the next birthday from this group (must be eligible to vote in national election). If the household contains one or more person age 18-23 who is registered to vote, use the next birthday rule to select the respondent from this group. If the household contains no adult age 18 to 23, select the respondent with the next birthday from this group (must be eligible to vote in national election). From there 283 interviews were obtained. Nine responses to the mail questionnaire were received from other than the target respondents and these are also included. The PSUs were Electoral Divisions (ED)or aggregates of electoral divisions. EDs were aggregated so that each PSU had a minimum size of 1,000 electors (2002) or 1,000 addresses (2007). PSUs were selected using a systematic sample, having sorted by area and characteristics of area (socio-economic status, urban-rural location), with a random start. The sort means that an element of stratification is incorporated. (3) There are two sets of weight variables for 2007, one looking back to the 2002 sample and a second providing a representative cross section in 2007. For the follow-up (panel) sample, the weight from 2002 was carried forward to the 2007 sample. The resulting sample was then calibrated to population totals from 2002 using a minimum distance algorithm (Gross, developed by Johanna Gomulka). For the 2007 cross-sectional sample, a design weight was first applied (based on number of adults age 18 and over in the household) to the supplementary sample, the 2002 weight was carried forward for the panel sample, and the weighted (by design weight/2002 weight) sample totals were calibrated to the population totals using the Gross minimum distance algorithm. These population totals are from the 2006 Census and the Quarterly National Household Survey in terms of household size (number of adults age 18 and over), age, sex, education, economic status, marital status and region. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - ISRAEL (2006) --------------------------------------------------------------------------- (1) Parliamentary elections, originally scheduled for November 2006, were held on 28 March 2006, four months after the collapse of Prime Minister Ariel Sharon's government on 20 November 2005. The government had been in difficulty since Israel's withdrawal from the Gaza Strip began in August 2005, which caused rifts within Mr. Sharon's Likud Party. On 9 November 2005, Amir Peretz became leader of the Labour Party, and took Labour out of the coalition government. Mr. Sharon, who had helped found the centre-right Likud Party in 1973, subsequently established a new party called Kadima ("Forward" in Hebrew). Mr. Sharon said Kadima would seek to end the long-time domination of Israeli politics by Likud and Labour. He announced that his party would support the creation of a Palestinian State, while at the same time maintaining all of Jerusalem under Israeli control. A number of leading figures on the left and right, including former Labour leader and Nobel Peace Prize laureate Mr. Shimon Peres, and Likud Defence Minister Shaul Mofaz, joined the new party. Mr. Sharon suffered a major stroke in mid-December 2005, which left him incapacitated and unable to continue his political duties. Nevertheless, his legacy continued to cast a shadow over the election campaign. Kadima was led into the elections by interim Prime Minister Mr. Ehud Olmert. Mr. Peretz emphasized Labour's social platform, while Mr. Benjamin Netanyahu, who had been elected as Likud's new leader in December 2005, insisted that additional unilateral territorial pullbacks would merely increase Israel's vulnerability to terrorist attacks. Many observers viewed the elections as a referendum on future policy in the occupied territories of the West Bank. The final results gave Kadima 29 seats, followed by the Labour Party with 19 seats. Other possible coalition partners for Kadima won a total of 30 seats: the Sephardi ultra-orthodox Shas party obtained 12, followed by the Gil (Pensioners Party) with a surprising seven seats; the ultra-orthodox Torah Judaism and the left-wing Meretz respectively won six and five seats. On the right, Likud came in first with 12 seats, followed by the far-right parties Yisrael Beitenu and the National Union - National Religious Party (Ichud Leumi- Mafdal: NRP) which won respectively 11 and nine seats. The remaining ten seats went to three predominantly Arab parties: Hadash, National Democratic Alliance (Balad) and United Arab List (Ra'am). (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented between June 17 and July 28, 2006, after the March 18 election through telephonic interviews. The sample consists of 1200 persons aged 18 years or older who are citizens of Israel. The primary sampling unit is the household, randomly selected. Telephone listings were employed to determine the random selection of households (not through random digit dial (RDD), but listed). Final selection was made according to which person answered the phone and corresponded to the above criteria. The final sample consists of a stratified sample of Jews, Arabs and former Soviet Union immigrants by geographical areas, and random sampling in each of the strata. The sample excludes some parts of the population, for instance Arabs who live in east Jerusalem; because they are not citizens of Israel, (East Jerusalem Arabs are about 3% of the Israeli population). Military persons while at home were included, but not those who were in military camps at the time of the study. Institutionalized individuals were also excluded from the sample since there is no telephone in such institutions. All other households without access to a telephone, or are not listed in the telephone directory, were excluded by default: households without telephones represent about 13% of total households. A quota sampling method was used for the group of immigrants from the former USSR. However, no substitution of individuals permitted at any stage of the selection process or during fieldwork. Response rate was 10.02%, from 11977 valid households (13780 total), 1200 interviews were completed (6950 refused or broke-off, while 3827 were not successfully contacted). (3) There is no weight variable for this election study given that weights are not necessary to make the population under study representative of the general population of Israel. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - JAPAN (2007) --------------------------------------------------------------------------- (1) General elections were held on 29 July for half of the 242-member House of Councillors, the Upper Chamber of the Japanese Diet. A total of 377 candidates (218 for the prefectural constituencies, and 159 under the party list) contested the 2007 elections. Incumbent Prime Minister Shinzo Abe promised to build a "beautiful country" and pledged to resolve the issue of individuals believed to have been abducted by the Democratic People's Republic of Korea in the 1970s and 1980s. Former Peruvian President, Alberto Fujimori, who was under house arrest in Chile, filed his candidature under the proportional representation list of the People's New Party. Changes in the electoral law reduced the number of seats in the LDP stronghold in rural areas and attributed more seats to big cities like Tokyo. Both parties targeted the 29 constituencies where only one seat was being renewed. During the election campaign, a major earthquake hit the northern part of Japan on 16 July. The ruling LDP suffered a heavy defeat, taking only 37 seats, The total number of LDP seats was reduced to 83 in the 242-member House of Councillors, which meant that the LDP lost control of the chamber for the first time since the party's inception in 1955. The LDP's coalition partner, the New Komeito, took nine seats. Inversely, the opposition DPJ won a record 60 seats and became the largest party in the House of Councillors. Other seats went to small parties and independent candidates. The former Peruvian President, Mr. Fujimori, was not elected. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The post election study for CSES module 3 of Japan was carried out between July 31, 2007 and August 13, 2007 as face-to-face-interviews. The original sample was part of a panel study, the Asian Barometer, referring to all Japanese voters aged 20 years or older, who are Japanese citizens and registered as voters. The sample was filled up with additional addresses to enlarge the total sample size. This additional sample part was processed in a two-step manner: The primary sampling units had been the electoral districts, sampled with probability-proportional-to-size (PPS) selection. From each district, about 8-14 samples (based on the population in each district) were selected with an equal probability sampling mechanism (EPSEM). In the second step, individual respondents were randomly selected, based on the voting registry. The selected persons were informed by letter before the first interviewer contact. The interviewers themselves were mostly been females (84%) aged 50 years or older (70% irrespectively of the gender). 45% of the interviewers had experience of more than ten years and were continuously trained. Finally, an incentive of 1000 Yen (about US$9) was paid to respondents for any successfully completed interview. (3) As the implementation of the CSES Module 3 notes, Japan used two different samples for CSES module 3. Consequently weights are necessary to make the sample representative of the population being studied and to compensate bias and potential skewness in relevant demographic factors. The included weights in the original data file were constructed by strata of six layers of age (20s, 30s, 40s, 50s, 60s, and over 70) and gender, based on the frequencies of Population Census of Japan 2005. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - LATVIA (2010) --------------------------------------------------------------------------- (1) The 2010 elections were the first to be held following the 2008 global economic crisis that severely hit the country. In December 2008 Latvia was granted a 7.5-billion-euro bailout from by the International Monetary Fund (IMF) and another 1.3 billion euros from the European Union (EU). The country's economy shrank by 18 per cent in 2009 and unemployment soared to 20 per cent. Prime Minister Valdis Dombrovskis who took office in March 2009 implemented austerity measures including wage cuts of up to 50 per cent in the public sector. The 2010 elections were contested by 1 234 candidates from 13 political parties and coalitions of political parties. They saw a duel between the centre-right Unity coalition led by Prime Minister Dombrovskis and the left-wing opposition Centre for Harmony (SC) led by the Mayor of the capital Riga Mr. Nils Usakovs. Prime Minister Dombrovskis called on voters' support for the Unity coalition arguing that the austerity measures implemented under his government were the only way to restore the country's economy. The SC pledged to provide a social-democratic alternative and criticized the government for its budget cuts and tax increases. It pledged to renegotiate the reimbursement plan of the bailout granted by the IMF and the EU and withdraw Latvian troops from Afghanistan. Five parties/coalitions surpassed the 5-per cent threshold. The Unity coalition came in first with 33 seats winning four more seats than its rival SC. Prime Minister Dombrovskis received the highest number of preferential votes. The other parties in his outgoing government - the ZZS and the National Alliance - took 22 and eight seats respectively. For a Better Latvia (LPP-LC) won eight seats while the PCTVL failed to win parliamentary representation. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The study is a post-election study for the 2010 legislative election. The sample is meant to be representative of Citizens of the Republic of Latvia, age 18-74. It was proceeded in four stages: 1) selection of sampling points, 2) selection of addresses; 3) selection of households; 4) selection of a respondent. Primary sampling units are sampling points which were selected from the complete list of Latvia’s populated points, with the number of citizens as a size measure. Populated points are sorted by region, district and urbanization level to keep regional and urban proportions in the sample. The planned number of sample points is calculated with the assumption of 10 interviews in each sample point. The random number determines the point in the list of cities or parishes, from which the selection must begin with and beginning from this point, considering certain steps, cities or parishes, to be included in the sample, are selected. The random starting address for an interviewer’s route is selected within each sampling point; the selection within urban areas is based on lists of dwellings (addresses). Households are selected according to the route instructions, each second dwelling in apartment buildings, each second home in an area of individual houses; ascending or descending sequence varies on odd and even sizes of streets. Within the second sampling stage - selection of addresses - starting address for an interviewer’s route in each sampling point is chosen randomly. During the third sampling stage - selection of households - every second dwelling in apartment buildings, every second home in an area of individual houses; ascending or descending sequence varies on odd and even sizes of streets is chosen. But the last sampling stage - selection of respondent - includes "the principle of the youngest man in the household". (3)To make the survey data representative to the universe, the data weighting in this survey was performed with an aim to get the sample division as close as possible to the universe. Criteria that was used for the data weighting are: age, nationality, region, type of residence and gender. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - MEXICO (2006) --------------------------------------------------------------------------- (1) Mexicans went to the polls on 2 July 2006 to elect all members of the Chamber of Deputies and the Senate as well as a new President. The main issues in the 2006 elections were the high crime rate, poverty and undocumented Mexican workers in the United States. The PAN endorsed Mr. Felipe Calderon as its presidential candidate and presented tough policies on crime, including life sentences for kidnappers. The PRD, led by the former Mayor of Mexico City, Mr. Manuel Lopez Obrador, formed an alliance, called the "Alliance for the Good of All", with the Labor Party and the Convergence Party (CONV). The alliance fought the elections under the slogan "For the Good of All, the poor first", pledging to improve healthcare and education. The Alliance for Mexico coalition was composed of the PRI and the Green Party of Mexico (PVEM). The PRI, led by former senator Mr. Roberto Madrazo, pledged to create nine million jobs inside Mexico to prevent Mexican workers migrating to the United States. Other main parties in the 2006 elections were the New Alliance Party (NA), formed by former PRI member Mr. Roberto Campa, and the Social Democratic and Peasant Alternative Party (ALT), led by the prominent civil and women rights activist, Ms. Patricia Mercado. The final results for parliamentary elections gave a resounding victory to the ruling PAN in both chambers, winning 206 seats in the Chamber of Deputies, up from 158; and 52 seats in the Senate. The Alliance for the Good of All won respectively 158 and 36 seats. The Alliance for Mexico took 123 and 39 seats. In the presidential election, a narrow margin separated the leading candidates. After a partial recount, Mr. Calderon was declared elected on 6 July with 35.88 per cent of the valid votes, just 0.6 percentage points ahead of Mr. Lopez Obrador. On 8 July, 100,000 people responded to Mr. Obrador's call to protest the election results. He filed a legal challenge to the results on 10 July, alleging widespread fraud and calling for a manual recount of every ballot. On 6 September, the Federal Electoral Tribunal confirmed the victory of Mr. Calderon. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm (2) This is a face-to-face, post-election study focused on the Mexican presidential and legislative elections of 2006. The sample was selected through a multistage procedure in which the universe was stratified in four regions according to geography and partisanship of the state governor. An independent sample was drawn within each region. Within each region, the primary sampling units were electoral precincts clusters. The clusters were defined as groups of all of the precincts with similar vote for PRI and belonging to the same county (municipality). In the second stage the precincts within the clusters were selected with probability proportional to size (PPS). The third stage had been blocks randomly selected in the precinct area In each block, houses are selected following systematic methods of random start households within the precinct and the last stage was the respondent in the selected household. Respondents in the household were not selected randomly. Interviewers followed quotas of gender and age. The response rate was calculated to be 46.5%. (3) Weights are estimated and used to correct for unequal probability on the selection of sampling units. Additionally, post stratification weights were estimated to correct for non-response and to match known demographic characteristics of the population gender and age. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - MEXICO (2009) --------------------------------------------------------------------------- (1) At stake in the July 2009 elections were 500 seats in the Chamber of Deputies. In the 2009 elections, the PAN was once again challenged by the PRI. The latter had dominated the country's presidency from 1929 until the election of Mr. Vicente Fox (PAN) in 2000. PRI's new leader, Ms. Beatriz Elena Paredes Rangel, emphasized that the party had learned from its errors, referring to past corruption problems. The PRI used a slogan "proven experience, new attitude", recalling over seven decades of the party's leadership. It pledged to provide new economic policies to tackle the economic crisis. President Calderon pledged to continue to tackle crime. According to the government, crime-related deaths dropped by 26 per cent in the first trimester of 2009. He urged voters' to allow the PAN to continue to work for economic growth, job creation and public safety. The PRD led by Mr. Jesús Ortega also vowed to tackle organized crime. However, it was reportedly losing ground due to internal splits. Polling took place without major incidents in most areas, except in the state of Guerrero where 11 people, including a PRD town councilor and his family, were killed. The PRI won a decisive victory securing 237 seats, up from 104. The PAN came in second with 143 seats, losing 63. The PRD took 71 seats, losing 55. Of all the eight parties represented in the outgoing Chamber, only the Social Democratic and Peasant Alternative Party (ALT) failed to win seats. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This is a face-to-face, post-election study focused on the Mexican legislative elections of 2009 and is representative of the national population 18 years or older. The sample was selected by a multistage procedure in which the universe was stratified in three dominions: states with elections for governor, states with local legislatures and states without concurrent elections. An independent sample was drawn within each region. In each dominion four layers of similar electoral competition were formed. In the first stage electoral sections were selected with probability proportional to size of the nominal listing (PPS). In the second stage blocks of houses were selected following systematic methods of random start, and in a third stage houses within the blocks were chosen also following systematic methods of random start. In the households, respondents are selected according to more recent birthday. Response rate was 58.9%. (3) The Mexican data includes two different weights. A first sampling weight corrects for unequal probability of selection sampling units. A second post-stratification weight is available to correct for non-response and to match known demographic characteristics of the populations (gender and age). --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - NETHERLANDS (2006) --------------------------------------------------------------------------- (1) Early elections were held for all seats in the House of Representatives on 22 November 2006. The main electoral issues included economic reforms and immigration policies. Mr. Balkenende called on voters to support the strong economic recovery under his government and criticized the PvdA's plans to tax pensioners. The PvdA argued that the government was asking too much from citizens with low and medium income and not enough from the country's richest citizens. The SP leader Mr. Jan Marijnissen said he would put an end to excessive economic liberalism. Mr. Balkenende also pledged to continue his tough immigration policies insisting that illegal immigrants should be deported. The Immigration Minister Ms. Rita Verdonk (VVD) said that if reelected the government would ban the wearing of full-length veils (burqas) in public. The PvdA leader Mr. Wouter Bos insisted on the importance of the country's tradition of multi-culturalism and tolerance. The left-wing opposition argued in favour of a general pardon for an estimated 26000 asylum seekers whose applications had been previously turned down. Neither the right nor the left won a clear majority. The Prime Minister's centre-right CDA remained the largest party with 41 seats. The VVD won 22 the Party for Freedom took nine and the D66 three. The main left-wing opposition the PvdA remained the second largest party with 33 seats. The SP won 25 seats 16 more than in 2003 thus becoming the fourth largest party. The Green Left won seven seats while the centre-left Christian Union took six. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Dutch 2006 CSES study was included in The Dutch Parliamentary Election Study 2006 (DPES). The DPES 2006 consists of several parts: the first wave of interviews, the second wave of interviews, a self-completion questionnaire, a non-response project and a stimulus project. The first wave of interviews was held during the six weeks before the elections on November 22, 2006. The second wave of interviews was held shortly after the elections. The self-completion questionnaire was left with the respondents of the second wave, to fill in for themselves. A number of non-respondents of the first and second wave were approached with a telephone or written questionnaire. The second wave of interviews, containing the CSES questionnaire, was conducted between November 23, 2006 and January 4, 2007. The goal was to interview most people at the beginning of this period. In this second wave 2359 respondents of the first wave were re-interviewed. They were again visited at their homes and interviews were conducted with an electronic questionnaire (CAPI) with questions that were different from the first wave. Note that most of demographic variables have been added to the data-file by Statistics Netherlands on the basis of population records (i.e., they were not part of the administered questionnaire. The study is based on a two-stage sampling procedure. In the first stage the primary units are selected and for every selected unit the number of secondary units to be selected is set. In the DPES 2006 the primary units were municipalities. Persons are the secondary unit. Within every municipality an equal preset number of persons is selected. After that the municipalities are selected with unequal chances in proportion to the number of persons in the municipalities in question. The largest municipalities, including Amsterdam, Rotterdam, The Hague and Utrecht, are automatically included in the sample. Persons in these municipalities are selected with an equal chance as the rest of the sample. This way a self weighing sample comes about, with a preset sample size. In the DPES the number of persons to be selected in every municipality was 12. In order to achieve a high response rate, respondents were carefully contacted. First, the respondents received a letter of introduction. Enclosed with this letter was a small set of stamps. Every respondent was visited at least once during the first half of the observation period of three weeks (see above). A card was left behind if the first three house calls were unsuccessful and the respondent was not present. After that, the interviewer tried to contact the intended respondent by phone in order to make an appointment for an interview. Meanwhile the interviewer kept visiting the address. An address could only be labeled as a definite 'no contact’ after six idle visits. When the response was achieved in the first wave, an appointment was made for a second wave interview. These appointments were planned as shortly as possible after the election date, so the responses of the second wave were concentrated at the beginning of the fieldwork period. It was also preferred that the interviewer of the second wave was the same person as the one that conducted the first interview. Respondents of the second wave interviews were asked to fill in a self- completion questionnaire. Persons that had not send in their questionnaire after a week received a letter of reminder with a new questionnaire and a reply envelope. Response rate in the pre-election wave was 66.9%. In the post-election wave, 89.9% of 1st wave participants took par, which makes 60.2% of the initial first-wave sample. (Source: The Design Report DPES 2006 Codebook: Henk van der Kolk, Kees Aarts, Martin Rosema, Martha Brinkman (2007). Dutch Parliamentary Election Study 2006. Enschede: University of Twente. Persistent identifier: urn:nbn:nl:ui:13-4zd-x4e). (3) The Dutch data include two different weights. The demographic weight includes the characteristics age, gender, marital status, urbanization, region, ethnicity. The political weight includes voting turnout as an additional population characteristic. Both weights are based on all respondents participating in wave I and II (the original weight variables wgt3 and wgt4). --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - NETHERLANDS (2010) --------------------------------------------------------------------------- (1) Early elections were held for all seats in the House of Representatives on June 9, 2010. The elections followed the collapse of Prime Minister Jan Peter Balkenende's fourth coalition government (comprising the CDA the PvdA the Christian Union and the Democrats 66) in February 2010. Disagreement over continuing the Dutch military involvement in Afghanistan was the immediate cause of the collapse of the coalition. Prime Minister's CDA was in favour extending the deployment of Dutch troupes beyond 2010, but the PvdA was vehemently opposed. Shortly before the 2010 elections the country's Bureau for Economic Policy Analysis announced that 29 billion euros would have to be cut from public spending by 2015 prompting many parties to focus on the budget cuts during the election campaign. The parties on the left and the right presented conflicting austerity plans. The pre-election debate also focused on immigration. The VVD promised a reduction in benefits for immigrants while the PVV pledged to restrict immigration. The PvdA was reportedly drawing support from immigrants thanks to its new leader Mr.Job Cohen who had been popular among the immigrant community while serving as Amsterdam Mayor. As in the 2006 elections no party secured a majority in 2010. The final results gave 31 seats to the VVD and 30 to the PvdA. The PVV came in third with 24 seats while the CDA took 21 seats. After lengthy negotiations on 29 September the CDA and the VVD agreed to form a coalition government with parliamentary support from the PVV. (Source: Adjusted from PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) Dutch 2010 CSES study was conducted as part of The Dutch Parliamentary Election Study 2010 (DPES). The DPES 2010 consists of several parts: two waves of face-to-face interviews - before and after the elections, and a drop-off questionnaire distributed after the post-election wave. The first wave of interviews was was administered in May 2010. The post- election wave of interviews started on June 10, and ended on July 22, 2010. The CSES questions were predominantly, but not exclusively, asked in the second wave. The Dutch Electorate of the 2010 elections, as represented in the electoral registers, represented the sampling frame. The frame excluded institutionalized persons (about 1.5%), and two geographic regions. The Wadden isles in the North Sea have been excluded because of logistic problems. And some postal codes in South-Eastern Amsterdam have been excluded because these areas are not considered to be safe at every time of the day. This is less than 0,01% of the population. In total less than 3,5 percent from the eligible population was excluded. Respondents (individuals, not households) were selected on the basis of official population records. Selection was done in a two-step procedure. The first step selects primary sampling units (municipalities). Based on a set minimum number of successful interviews of 12, a desired net post- election sample size of at least 1,800, and an estimated response rate of 55 percent, municipalities were drawn with a weight proportional to the number of enfranchised voters. In the second step, a constant number of persons to be approached was drawn from the electoral registers in the selected municipalities. The product of these two steps ensured that the inclusion probability was the same for all enfranchised voters. In order to achieve a high response rate, some received a set of the most commonly used stamps, value 4.00 Eurpos, others a gift certificate. Maximum number of contact attempts made per respondent (person) was 6. The most conservative estimate of the response levels is 62,4 percent (first wave), and 53,8% for the first and second wave. A more positive view on response in the first wave (excluding errors in the sampling frame and including all CAPI, CATI and PAPI and aborted interviews) is 65,9%. Sources on DPES 2010: http://www.dpes.nl/documents/responsenumbersDPES2010.doc. CBS (2011) Verkiezingen: Participatie, Vertrouwen en Integratie, CBS: Voorburg (in Dutch only) [http://www.cbs.nl/NR/rdonlyres/920D38EF-DE95- 45C0-8BF3-F54721BB2622/0/2011g62pub.pdf] (3) The Dutch data includes two different weights. The demographic weight includes the characteristics age, gender, marital status, urbanization, region, ethnicity. The political weight includes voting turnout as an additional population characteristic. Both weights are based on all respondents participating in wave I and II (the original weight variables wgt3 and wgt4). --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - NEW ZEALAND (2008) --------------------------------------------------------------------------- (1) On 12 September, Prime Minister Helen Clark announced that the House of Representatives would be dissolved on 3 October and that elections would be held on 8 November. The 2008 elections once again saw a duel between the Labour Party and the National Party. Prior to the elections, scandals involving the Foreign Minister and the economic slowdown reportedly gave the opposition National Party the lead in opinion polls. In late September, the national statistics office announced that the country's economy had fallen into recession for the first time since 1998. The global financial crisis reportedly worsened the situation in the run-up to the elections. Prime Minister Clark of the Labour Party promised a deposit guarantee scheme to protect investments, expanded training opportunities for youth, and a universal student allowance by abolishing parental income testing. National Party leader John Key pledged to protect businesses from financial losses. His party's policies included income tax cuts and more spending on key infrastructure projects such as roads, schools and a broadband network. It promised to promptly execute dam projects on the West Coast, opposed by the Labour government. The final results gave 58 seats to the National Party and 43 to the Labour Party. The Greens won nine seats, followed by the ACT and the Maori Party, which took five seats each. The Progressive Party and the United Future won one seat each. New Zealand First failed to win any parliamentary representation. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The study is meant to be representative of New Zealanders on the electoral roll immediately prior to the 2008 election. A very small number who registered in the two weeks before the election were excluded. The sample is a random sample from the electronic electoral rolls. (A panel component was also present, but all data submitted to the CSES is from those freshly sampled for the 2008 election.) Sampling units were selected using SPSS select cases (random sample). Furthermore, the sample was partitioned into four groups. New Zealanders of Maori descent can choose to be registered on a separate Maori roll, and cast their electorate votes within seven constituencies that overlie the general electorates. For its new sample the 2008 New Zealand Election Study randomly sampled within four subsets of registered voters and oversampled subsets 1, 3, and 4 below, groups of particular interest to the researchers. 1. Those on Maori roll 27 and over 2. Those on General roll 27 and over 3. Those on Maori roll 18-26 4. Those on General Roll 18-26 For the CSES release, respondents from the three oversampled subsets were re-sampled to bring their numbers into the appropriate proportions among registered voters, removing any need for sampling weights. Those sampled were given the opportunity to fill out the questionnaire online but could only do so with the appropriate password and identifier. Without reducing the full sample totals by any means (deceased, inaccessible, or anything else) percentages of respondents out of the totals originally sampled were: General roll, 27 and over: 42.8% General roll, 18-26: 22.0% Maori roll, 27 and over: 27.3% Maori roll, 18-26: 12.7% Effective response rates would be somewhat higher if deceased and inaccessibles were removed from the denominator, but the New Zealand electoral roll is well maintained and therefore quite up to date. Population data for age and sex comes from the electronic version of the rolls used for sampling. Age groups are reported in five-year bands. The target sex ratio was estimated from a separate sample of 10,000 from the roll data. This data is only available under conditions of confidentiality for purposes of social science research from the Electoral Enrolment Centre. Education data comes from the 2006 census, available online at the Statistics New Zealand website. The large discrepancies reflect to some extent the propensity of respondents to be concentrated among the more educated. However, highest qualification data available from the New Zealand latest census in 2006 does not tie in particularly well to the codes requested by the CSES or those most respondents would recognize in a survey. The New Zealand qualifications framework has changed significantly over the last thirty years, making older respondents responses’ difficult to reconcile with current codes. I simply don’t believe that 45% of people over 15 in New Zealand in 2008 had no education or had not completed primary school - what this data apparently signifies, (New Zealand’s minimum school leaving age was set at 15 in 1944, with most children leaving primary school at age 12). Accumulating from lower qualifications upward, the percentage without post-secondary qualifications in the 2006 census data is 63 per cent: in the weighted sample it is 55 per cent. However, the most obvious discrepancy is the proportion of university-educated, twice what it should be. An alternative weight is available to adjust for this, along with sex and age, although not included in this version of the CSES deposit. (3) The weights were designed to match known sex and age distributions in the electoral roll from which the sample was taken. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - NORWAY (2005) --------------------------------------------------------------------------- (1) Norwegians went to the polls on 12 September 2005 to elect an enlarged parliament of 169 members, up from 165. Since the 2001 elections, when the Labour Party had suffered its worst result in 90 years, the country had been led by a center-right minority coalition under Prime Minister Kjell Magne Bondevik, which controlled 62 of 165 seats in parliament (Stortinget). The coalition was composed of three parties: the Prime Minister's Christian People's Party, the Conservatives and the Liberal Party. The coalition also had support from the far right Progress Party, known for its anti-immigration policy, which had won 26 seats in the 2001 elections, bringing the number of seats controlled by government to 88. The three main opposition parties formed the Red-Green alliance composed of former Prime Minister Jens Stoltenberg's Labour Party, the Socialist Left Party and the Centre Party (Agrarians), holding 76 seats in all. One MP represented the Coastal Party, defending the interests of fishing communities and opposing membership of the European Union. 77.4 per cent of Norway's 3.4 million eligible voters cast their ballots on election day. Final results gave an absolute majority of 87 seats to the Red-Green alliance, while the outgoing government and its ally, the Progress Party, won a total of 82 seats. Within the winning coalition, the Labour Party increased its seats to 61, up from 43 in the 2001 elections. The Centre Party (Agrarians) obtained 11 seats, one more than the previous election, while the Socialist Left Party was reduced to 15 seats, having lost eight. The Conservative Party, the biggest party in the outgoing government, suffered its worst results in its history, retaining 23 seats and losing 15. The Christian People's Party won 11 seats, down from 22. On the other hand, the Liberal Party won its largest number of seats ever: ten, up from two. The far-right Progress Party, led by Mr. Carl Hagen, increased its seats from 26 to 38, becoming the second largest party in the new parliament. The Coastal Party lost its only seat. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between September 13, 2005 and December 20, 2005. The original sample consists of 3000 persons aged 18-79 years which is part of an rolling panel (half of the sample replaced at each election). The sample is selected from registers covering all Norwegian inhabitants, and in accordance with Statistics Norway's sampling frame. Local municipalities are employed as primary sampling units (PSUS). In Norway there are 435 municipalities in 19 counties, but municipalities with very few inhabitants were pooled with adjacent municipalities to form one PSU. A total of 363 PSUS were then stratified into 109 strata. Stratification variables were industrial structure, number of inhabitants, centrality, communication structures, commuting patterns, trade areas and (local) media coverage. Stratification was done such that no PSU had less than 7 per cent of the total population in its stratum, and separately for each county. All PSUS with more than 30000 inhabitants, and some with a population number between 25000 and 30000, constitute separate strata. For the remaining PSUS, the most possible homogeneous strata was formed. In a second stage one PSU was selected from each stratum. PSUS which constituted separate strata were included with certainty. For the remaining strata, one unit was drawn with probability proportional to the number of inhabitants. A sample of survey units (persons) was drawn from the population register, selected from the 109 sampling areas, using systematic random sampling. The sampling fraction at the second stage is proportional with the inverse selection probability at the first stage. The final sample then is self-weighting when both stages are taken into consideration. If a sampled unit from the panel sample has moved out of their original psu, it is still included in the sample. Response rate was 68%. The total amount of households in the sample was 2965, out of which 2012 interviews were completed (655 refused or broke-off, 142 were not successfully contacted, 156 other non-response). (3) No weights are necessary to make the sample representative of the population, hence weights are not included in the data file. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - NORWAY (2009): --------------------------------------------------------------------------- (1) The September 2009 elections were the first to be held after constitutional amendments in February 2007 effectively transformed the Parliament (Stortinget) into a purely unicameral parliament as of 1 October 2009. Until now, the Stortinget, although unicameral, used to split into two bodies (the Odelsting and the Lagting) when handling legislation. These two bodies were abolished after the 2009 elections. The 2009 elections were held against the backdrop of the global economic crisis which had also hit Norway. The main issue in the 2009 elections was how much oil revenue should be used for public expenditure. Prime Minister Stoltenberg's Labour Party, known for its pro-oil industry stance, initially argued that the Arctic and Lofoten coastlines should be opened for oil drilling in addition to the current North Sea reserves, but then withdrew its proposals due to opposition from the Socialist Left Party. The Progress Party argued that more of the oil fund should be used for public works and taxes should be lowered. The final results gave 86 seats to Prime Minister Stoltenberg's coalition, including 64 to the Labour Party. It was the first time in 16 years that the governing coalition won the general elections. The Progressive Party remained the second largest party, winning 41 seats, followed by the Conservative Party with 30 seats. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The CSES module was implemented in a face-to-face post-election survey between September 15, 2009 and February 04, 2010. The original sample consists of 2946 households, whereat half of the sample is part of an rolling panel of the previous election in 2005 (replaced at each election). For the interviewing, respondents had to be between 18 and 80 years old, as well as national citizens of Norway. The sample was selected from registers covering all Norwegian inhabitants, and in accordance with Statistics Norway's sampling frame. Local municipalities are employed as primary sampling units (PSUS). In Norway there are 430 municipalities in 19 counties, but municipalities with very few inhabitants were pooled with adjacent municipalities to form one PSU. A total of 363 PSUS were then stratified into 109 strata. Stratification variables were industrial structure, number of inhabitants, centrality, communication structures, commuting patterns, trade areas and (local) media coverage. Stratification was done such that no PSU had less than 7 per cent of the total population in its stratum, and separately for each county. All PSUS with more than 30000 inhabitants, and some with a population number between 25000 and 30000, constitute separate strata. For the remaining PSUS, the most possible homogeneous strata was formed. In a second stage one PSU was selected from each stratum. PSUS which constituted separate strata were included with certainty. For the remaining strata, one unit was drawn with probability proportional to the number of inhabitants. A sample of survey units (persons) was drawn from the population register, selected from the 109 sampling areas, using systematic random sampling. The sampling fraction at the second stage is proportional with the inverse selection probability at the first stage. The final sample then is self-weighting when both stages are taken into consideration. If a sampled unit from the panel sample has moved out of their original psu, it is still included in the sample. The total amount of households in the sample was 2946, out of which 1782 interviews were completed (965 refused or broke-off, 197 were not successfully contacted, 2 other non-response). Response rate was 60.5%. (3) No weights are necessary to make the sample representative of the population, hence weights are not included in the data file. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - PERU (2011): --------------------------------------------------------------------------- (1) The 2011 parliamentary elections were held in parallel with the presidential polls and the elections for the country's representatives to the Andean Parliament. At stake were 130 seats in the newly enlarged Congress, up from 120. The media focused on the presidential race, pushing the parliamentary polls to the back burner. Mr. Humala's Peru Wins alliance pledged to review the Constitution on the basis of a "new economic model". It promised to work for a greater State role in the economy, and pledged to increase the tax on international mining companies and to prioritize natural gas produced at the Camisea fields for domestic use rather than for export. It also promised to provide free nursery and public education, and increase the minimum wage and pensions. Ms. Fujimori's Force 2011 Alliance promised to promote free trade and open markets and reduce the costs of doing business by 20 per cent over the next five years. It pledged to crack down on crime by applying harsher punishments, including the death penalty. It also promised to provide more housing with water and basic sanitation. The Perú Posible Party, led by Mr. Toledo - Peru's first elected indigenous president - focused on poverty reduction, promising to reduce the poverty rate from the current 35 per cent to 10 per cent. It promised to transform Peru into a maritime hub by improving logistics and infrastructure and attracting more foreign investment to develop the country's economy. Mr. Kuczynski's Alliance for Major Change pledged to eliminate extreme poverty within 10 years and reduce sales tax from the current 18 to 15 per cent. It promised to provide all Peruvians with clean drinking water. Mr. Kuczynski pledged to surrender his US nationality if elected. The final results gave 47 seats to the Peru Wins alliance and 38 to Force 2011 Alliance. Perú Posible Party came in third with 21 seats while the Alliance for Major Change took 12 seats. The remainder went to small parties. None of the presidential candidates secured an absolute majority. In the run-off presidential elections held on 5 June, Mr. Humala (Peru Wins alliance) won over Ms. Fujimori (Force 2011 Alliance). (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This study was conducted as a post-election, face-to-face survey over the period May 7th to May 08th, 2011. Multi-stage stratified sampling was used, with the primary sampling units being housing blocks in cities and towns in rural areas. (3) The sample was weighted according to the number of population in each region and urban-rural distribution. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - PHILIPPINES (2010): --------------------------------------------------------------------------- (1) The May 2010 parliamentary elections were held concurrently with the presidential polls. At stake were 280 seats in the House of Representatives (up from 269, see note) and half of the Senate seats (12). The media focused on the presidential polls, placing the parliamentary polls on the back burner. The Lakas-Kampi-CMD coalition was challenged by the NP-NPC coalition. The NP leader, former Senate President Manny Villar, was also running for the presidency. Mr. Villar, a property tycoon, focused on his rags-to-riches story, but his image was reportedly tarnished by corruption allegations. Former actor and president, Joseph Estrada, was also running. He had been jailed in 2001 for corruption but was pardoned by President Arroyo later the same year. He reportedly remained popular thanks to his heroic gangster movies. Senator Benigno Aquino III ran for the presidency on a Liberal Party (LP) ticket. The LP of former Speaker and Quezon City Mayor, Mr. Feliciano Belmonte Jr., was in coalition with the KKK (Kapayapaan, Katarungan, Kaunlaran). The LP coalition promised to lead a transparent administration. Mr. Aquino vowed to open an investigation into past allegations of corruption against President Arroyo. As in the past, the elections were marred by violence, resulting in the death of over 90 people. Four municipalities in Mindanao declared that polling could not take place after poll workers reported harassment and death threats. The election commission accepted the declaration and announced that it would organize special elections there. Technical problems with new automated voting machines that were being used for the first time were reported in several polling stations. No major international organizations sent observer missions. The People's International Observers' Mission 2010 (PIOM) noted several irregularities and criticized the lack of secrecy in certain precincts. The final results for the House of Representatives gave 119 seats to the LP, 45 to the Lakas-Kampi-CMD coalition, and 22 to the NP-NPC coalition. The remainder went to small parties. In the Senate, the LP and the NP-NPC coalition won three seats each. The Lakas-Kampi-CMD and the Pwersa ng Masang Pilipino (Force of the Filipino Masses) of outgoing Senate President Juan Ponce Enrile took two seats each while the People's Reform Party (PRP) took one seat. The remaining seat went to an independent candidate. In the presidential elections, Mr. Aquino was elected with over 40 per cent of the votes. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This post-election face-to-face survey was conducted between June 25 and 28, 2010, in five languages: Filipino, Cebuano, Ilocano, Maranao, and Hiligaynon. For sampling purposes, the Philippines was divided into four study areas: National Capital Region (NCR), Balance Luzon, Visayas, and Mindanao. The sample size for each of the four study areas was 300 voting-age adults. Multi-stage probability sampling was used in the selection of sample spots (60 spots per area). The three study areas, apart from NCR, were further divided into regions, and then provinces were selected from the regions according to the probability proportional to population size (PPS) of the region. Once the sample provinces were selected, 60 spots for each of the major areas were allocated among the sample provinces. Using the quota set for each spot in each region, the spots were distributed in such a way that each province was assigned a number of spots roughly proportional to its population size. For the third stage, within each sample spot, five to six households were established by systematic sampling (random route procedure). The individual respondents are identified using a probability selection table. Overall, 2,994 households were selected, out of which 2,238 had been valid. In sum, the Philippine sample includes 1,200 realized interviews, 300 in each of the study areas, resulting in a response rate of about 54%. (3) According to the sample process, a sample weight is necessary, to control for disproportional sampling of the overall population size of the study areas. The weight bases on data from the Philippine Census (2007), calculating the ratio the sample size of the an area and the projected population in the same area. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - POLAND (2005) --------------------------------------------------------------------------- (1) On 25 September 2005, parliamentary elections were held for both chambers of the Polish Parliament: the 460-seat lower house (Sejm) and the 100-seat Senate (Senat), in parallel with presidential elections. . The outgoing government had come to power promising to deal with unemployment, which had reached 16 per cent. Mr. Belka's policies included tax cuts and an increase in social spending, but these reforms were reportedly hampered by disagreements within his party. Long-running corruption scandals involving the SLD took the centre stage during the election campaign. The main rivals of the outgoing government were two centre-right parties, the Civic Platform (PO), led by Mr. Donald Tusk, a known free market advocate, and the more conservative Law and Justice (PiS) party, led by Mr. Jaroslaw Kaczynski. The two parties announced that they would seek to form a coalition government, although they proposed different tax policies during the election campaign. The final results showed major progress for the PiS and the PO, which won 155 and 133 seats respectively in the Sejm, and 49 and 34 seats in the Senate. Self-Defense, a populist grouping advocating State-funded agriculture and an increase in government social programmes, came in third in the Sejm with 56 seats. The SLD suffered a heavy defeat, retaining only 55 seats in the Sejm and none in the Senate. These elections marked the left's worst results since its establishment as reformed communists in 1990. In the run-off presidential election held on 23 October, Mr. Lech Kaczynski beat Mr. Tusk by securing 54.04 per cent of the votes. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Polish Election Study 2005 for the third module of CSES was a pilot study. At the time, the second release CSES-data had been published, neither information about the implementation of the module in the Polish post election survey, nor about the sampling process had been available. (3) The data include a demographic weight to make the sample comparable to the overall Polish population. Researchers note that the construction rules of the weight as well as the data used for the benchmark are unknown, to date. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - POLAND (2007) --------------------------------------------------------------------------- (1) Following the collapse of the ruling coalition in August, parliamentary elections were held on 21 October 2007, two years earlier than the official due date. In all, 6,187 candidates, including 1,428 women, ran for the Sejm, while 385 candidates vied for seats in the Senate. The 2007 elections again saw a duel between the PiS, led by Mr. Jaroslaw Kaczynski, and the PO of Mr. Tusk. The PiS promised to continue its policies, focusing on the domestic economy and promising tax cuts. It appealed to voters to support it in the interest of steady economic growth. The PO pledged to introduce a flat income tax of 15 per cent and to continue the privatization of State companies that stagnated under the PiS-led government. It promised to forge better ties with neighbouring countries, and pledged to introduce the European single currency - the Euro - as soon as possible, to further economic development. Mr. Tusk also promised to work towards the withdrawal of Polish troops from Iraq. The Left Democratic Alliance (SLD), which had won 55 seats in the Sejm in the 2005 elections, formed an election coalition - LiD - with the Social Democracy of Poland (SDPL), the Democratic Party (PD), and the Labour Union (UP) in September 2006. The coalition was led by former president Mr. Aleksander Kwasniewski. He criticized the PiS-led government, arguing that the country could not afford a radical reduction of taxation. A total of 53.88 per cent of the country's 30 million registered voters turned out at the polls, representing the highest turnout in the post- communist era. Due to the higher than expected turnout, some polling stations ran out of ballot papers, and others stayed open longer than scheduled, thus delaying the release of the first exit polls by nearly three hours. The Organization for Security and Co-operation in Europe (OSCE) observed the polls. Although it declared the elections to be pluralistic and democratic, it noted "sporadic partial interventions by State organs" in the election campaign. The final results gave the PO 209 seats in the Sejm and 60 seats in the Senate. The PiS took 166 and 39 seats respectively. The LiD and the Polish Peasant Party (PSL) took 53 and 31 seats in the Sejm, while the remaining one seat went to the German minority. One independent candidate was elected to the Senate. Outgoing Prime Minister, Mr. Jaroslaw Kaczynski, conceded defeat. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Polish interviews were held face-to-face as a post election survey. Sampling was based on a four stage process, drawn accordingly to the 60 territorial layers, defined on the basis of country region and size of the accommodation areas. Thereby the number of observations from each layer in the sample was proportional to the number of people from a given layer in the population. In a first stage, district districts were chosen randomly. Second, within each of the selected districts, a sample of streets or resorts was drawn. In the third step, based on the map of streets or resorts, addresses of houses and apartments had been selected. Finally, within each of the chosen houses or apartments one person was randomly sampled. (3) The data include a demographic weight to make the sample comparable to the overall Polish population. Researcher should be aware of the fact that the construction rules of the weight as well as the used data for benchmark are unknown, so far. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - PORTUGAL (2009) --------------------------------------------------------------------------- (1) The September 2009 elections were held against the backdrop of the global economic crisis, which severely affected the country of 10 million inhabitants. How to deal with the economic crisis remained the main issue in the 2009 elections, in which 15 political parties were vying for seats. Prime Minister Socrates (PS) promised to implement large-scale public works projects to create more jobs. These included a new Lisbon airport, a bullet train to Spain as well as a road and rail bridge across the River Tagus in Lisbon. He also promised to continue economic and social reforms. The largest opposition party, the PPD/PSD, was led by Ms. Manuela Ferreira Leite, who was aspiring to become Portugal's first elected woman prime minister. Ms. Ferreira Leite argued that the PS's public works projects were wasteful and a non-essential luxury. She claimed that the PS would be passing on debts to future generations. Ms. Ferreira Leite, a former finance minister, pledged to make huge cuts in government spending and boost private investment to revive the economy. The PS remained the largest party in the Assembly of the Republic with 97 seats but failed to retain an absolute majority. The PPD/PSD took 81, six more than in the 2005 elections. The CDS-PP secured 21 seats, up from 12. The BE doubled its share from eight to 16. The PCP-PEV took the remaining 15 seats. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The sample was stratified by five regions (according to the C2027), as well as by the population size of localities (according to C2030). All in all, 113 areas were randomly selected, based on the corresponding definition of the National Statistics Office. The sample excludes the islands of Azores and Madeira, as well as institutionalized and military persons. The further sampling process contains of three stages: First, the selection of a starting point within each of the localities, using lists of ZIP codes and tables of random numbers. Next, households had been sampled according to the given starting points, based on pre-determined instructions about the route to adopt and a systematic interval of households. Finally, within each of the selected households, respondents had been chosen on the basis of next person having his/her birthday. To be interviewed people must have been at least 18 years old at the day of contact and had to be registered as voters. In contrast, the Portuguese citizenship was no requirement for participating in the post election study. Selected Households were informed by letters. Incentives or payments had not been given to the respondents, whether before, nor after the interviews. The interviewers received advanced training for the interviews, procedures of contact, and the method of random route. Interviews took place between October 2nd, 2009 and February 8th, 2010. In total, 7151 households were selected, out of which 3532 had been invalid addresses due to several reasons. From the residual 3619 household, 1317 had been successfully contacted and interviewed, which corresponds to a recourse of 36.39%. After removing 13 ineligible interviews, 1303 cases remained. (3) The data include a demographic weight to make the sample comparable to the overall Portuguese population aged 18 years or older. The weight is based on the distribution of the target population in terms of gender, age, and education, according to the Portuguese National Statistics Office Census 2001. Thirteen respondents from the source file, who had system missing values in the weight variable, were removed from the CSES dataset, due to their ineligibility status. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - ROMANIA (2009): --------------------------------------------------------------------------- (1) The first round of Romanian presidential elections in 2009 took place at November 11th on schedule after a regular five-year term in office. For the first time in Romanian history, the presidential election was not held parallel to the one for the Romanian parliament, in order to "render the presidential elections less partisan and conflictual" (Muntean et al. 2010: 754). Electoral campaigning formed three major candidates for the electoral race: The current president Traian Basescu (Democratic Liberal Party), his major challenger Mircea Geoana (Social Democratic Party), and Crin Antonescu (National Liberal Party). Since Basescu won the Romanian presidential election in 2004, his term in office was mainly determined by conflicts about the division of power between the parliament and the president and the presidential right of nominating the prime minister. These conflicts dominated the presidential electoral campaigns, polarizing the Romania electorate: "Basescu’s presentation of himself as acting for the people and against big interests, specially those of the ‘moguls’ and the former communists, was pitted against the opposition’s view of him as a danger for democracy, a populist ould-be dictator, who was incapable of collaborating with the parliament and the parties" (Muntean et al. 2010: 755). In addition, topics about changes in the Romanian flat tax system, the reduction of social security, and private health care dominated the electoral campaigns. In the first round of the election, none of the favored candidates was able to gain an absolute majority. Basescu and Geoana had to re-run for the second round of elections, on December, 12th.|Traian Basescu managed to win an absolute majority (50.35) with a small advance over his challenger (49.7%). Thus, Basescu was re-elected as the Romanian president for another five years. (Source: Muntean et al. 2010) (2) Interviews in Romania took place between December 9th and 20th, 2009, as face-to-face interviews, after to the second round of presidential elections. The Romanian sample bases the third round of a three-wave panel, which covers un-institutionalized Romanian citizens aged 18 years and older, registered as eligible voters. About 0.25% of the Romanian population was thus excluded due to being institutionalized. The Romanian sampling process bases on two-step stratification. In the first stage voting districts (for parliament elections) were systematically sampled, according to their proportionality of registered adults. The voting districts were stratified with regard to 18 cultural areas and seven types of municipalities (poor communes, medium developed communes, developed communes, cities with less than 30,000 inhabitants, cities of 30- 100,000 inhabitants, cities of 100-200,000 inhabitants, cities with more than 200,000 inhabitants). In sum, this first stage yields 126 strata, whereat strata with less than seven cases were added to the closest neighbor category, inside the same cultural area and the same type of residence (urban or rural). The resulting number of strata was of 96.In a second step, eight respondents were selected by systematic sampling with equal probabilities from the electoral register of each voting district (the main sample). A reserve sample of eight respondents was selected in order to compensate for refusals etc. This reserve sample was used to substitute incomplete interviews per primary sampling unit. Romanian interviewers were mostly between 26 and 64 years old. About two thirds of them were females, mainly having a University undergraduate degree completed. Overall, 3157 households were sampled, out of which 2219 turned out to be valid. In the first round of the three-wave panel 1504 persons were interviewed. The current Romanian sample, used for the CSES, includes 1403 respondents. In sum, responds rate for the current sample is about 68%. (3) A demographic weight controls for differences in socio-demographic characteristics between the employed sample and the underlying Romanian population. The construction of the weight depends on the population’s distribution of gender, age, education, and the region of residence, based on data from the Romanian National Institute of Social and Economic Statistics (INS). --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SLOVAKIA (2010) --------------------------------------------------------------------------- (1) The June 2010 elections were the first to be held after the country's adoption of the Euro in January 2009. They followed elections in Hungary (April) and the Czech Republic (May) in which centre-right parties regained power after pledging to reduce their country's debt amid the Greek and euro crises. In May, the Slovak Government agreed to make some 800 million Euros available for the European Union's bailout package for Greece. In 2010, 18 parties and 2,397 candidates were vying for seats in the National Council. The parties in the outgoing government - the Smer-SD, the LS-HZDS and the SNS - were challenged by the SDKU - DS and the KDH. The SMK of Mr. Pal Csaky and the Most-Hid (Bridge) joined the parties on the right. Prime Minister Fico called on voters to support his Smer-SD and its allies so that he could continue to lead a "strong social State". The Smer-SD's coalition partner, the SNS, focused on patriotism and traditional values. statements. Although the SNS refrained from targeting the Hungarian minority during the election campaign, the party's election slogan was "Don't feed those who don't want to work". The SNS argued that the slogan referred to all those who abuse the welfare system. The LS-HZDS of Mr. Vladimir Meciar was reportedly weakened after certain prominent members, including Trnava regional governor Tibor Mikus, left the party. The SDKU - DS, leading the parties on the right, pledged to reduce the budget deficit, improve the business environment to create more jobs, and tackle corruption. It opposed Slovakia's pledge to provide some 800 million euros to Greece. The KDH and its leader, Mr. Jan Figel, focused on the party's traditional platform, underscoring the importance of Christian and family values. The SMK pledged to protect the interests of the country's 500,000 ethnic Hungarians in cooperation with the Hungarian Prime Minister, The Smer-SD increased its share to 62 seats. However, its ally, the SNS, took only nine seats, and the LS-HZDS failed to win parliamentary representation for the first time since its inception in 1991. The SDKU-DS came in a distant second with 28 seats, followed by the SaS with 22 seats. The KDH and the Most-Hid took 15 and 14 seats respectively. The SMK failed to win parliamentary representation for the first time since its inception in 1998. In all, 23 women were elected. Both Prime Minister Fico and the SDKU - DS election leader, Ms. Radicova, claimed election victory. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Slovakian post-election survey took place between June 18 and July 20, 2010, as face-to-face interviews. Respondents had to be Slovakian citizens, aged 18 and older to become part of the sample, excluding institutionalized persons. Sample selection based on a stratified multistage probability sampling method: The first step of the selection process was due to defining 240 sampling units, according to the census of Slovak population by the Statistical Office of the Slovak Republic. Second, household addresses were sampled. Finally, respondents were chosen, employing the method of next birthday, taking restriction of the sample, e.g. age and nationality, into account. A selected respondent was due to non-sample after four unsuccessful contacts at various day times. Before the first personal contact, respondents received a letter about the interviewing. About 75 per cent of the interviewers were females, most of them of higher education. TNS SK trained all interviewers before as well as during the field work. (3) The Slovakian data include a demographic weight the make the sample representative of the population under study. The weight compensates for disproportional probability of selection, taking gender, age, education, the size of settlement, and the region of residence into account. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SLOVENIA (2008) --------------------------------------------------------------------------- (1) On 16 June 2008 President Danilo Türk signed a decree setting the date of elections to the National Assembly for 21 September. Nearly 1 200 candidates from 19 parties were vying for 88 seats in the National Assembly two others being reserved for ethnic minorities. For the first time there were no independent candidates. The quota for women candidates introduced in 2006 was applied for the first time. A transitional provision set a 25 per cent quota for 2008 increasing to 35 per cent women candidates in 2012. Approximately one third of the registered candidates were women. Five candidates registered for the seat reserved for the Hungarian minority while four ran for the one reserved for the Italian minority. The SD and the SDS in coalition with the NSi were the major parties in the 2008 election campaign. The main election issues were taxes housing and border issues with Croatia. Seven parties surpassed the 4 per cent threshold to win parliamentary representation. The final results gave the opposition SD ZARES LDS coalition a total of 43 seats. The SDS came in second with 28 seats but its partner the NSi failed to win any parliamentary representation. Despite the quota the number of women members increased only by one bringing the total to 12. Only 41 outgoing members were re elected. Several veteran MPs who had held their seats since the country gained independence in 1991 were voted out. Among the members of the outgoing government only the Prime Minister and two ministers were returned to parliament. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This study was conducted as a face-to-face post-election survey. The study sampling frame corresponds to the Central Register of Population (CRP) and includes all residents with permanent address, including both citizens and non-citizens. The study employed a stratified two-stage probability sampling method in which the primary sampling units (clusters of enumeration areas, CEA) were selected using the following procedure: first, the entire country was divided into about 9000 CEA’s and stratified according to 12 regions x 6 types of settlements. After this, fixed numbers of CEA were selected inside each strata with probability proportional to size of CEA. In total 136 CEA were selected. Then 12 persons in each primary sampling unit were chosen by simple random sampling. Persons selected from Central Population Register were identified by name and address. Five contacts were made with the household before declaring it non-sample. The response rate was 64.0%. (3) There is no weight variable in this election study given that weights are not necessary to make the population under study representative of the general population of Slovenia. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SOUTH AFRICA (2009): --------------------------------------------------------------------------- (1) On 13 February 2009, President Motlanthe called elections for 22 April. At stake were 400 seats in the National Assembly (lower chamber of parliament). Parliament was due to elect the country's new president after the elections. Pursuant to the Government's 50:50 gender equality policy, the major parties endorsed more women candidates than in the previous elections. In all, 9,130 candidates, including 3,511 women, representing 40 parties ran in the 2009 elections. The ANC ran on its record, citing economic development and support for the poor in the form of housing, water and electricity. Fighting corruption became a key electoral issue for opposition parties during the campaign. The ANC came under heavy criticism over corruption charges against Mr. Zuma, who had been accused of accepting bribes from a French arms company. State prosecutors dropped the charges in April, two weeks before the elections. The DA, led by Ms. Helen Zille, pledged to run a "clean" government. It accused Mr. Zuma of having "bullied" the State prosecutors into withdrawing the corruption charges and filed a legal challenge against the decision. It promised to create more jobs, provide better education and tackle crime. The COPE pledged to push for the reinstatement of the corruption charges against Mr. Zuma. The COPE promised to fight crime by introducing a "three strikes" policy, whereby career criminals would face life imprisonment for repeat offences of rape, robbery or murder. The IFP, founded in 1975 by Mr. Mangosuthu Buthelezi, campaigned on a platform to fight corruption and crime. A total of 23,000 police officers and soldiers were deployed to ensure security. The run-up to the election was relatively peaceful, although both the ANC and the IFP alleged interference by the other party. The DA, the COPE and the IFP all accused the ANC of vote buying and claimed that it had distributed government food parcels during the campaign. The ANC dismissed the allegation. The polls were monitored by 355 international observers and hailed as free and fair by the African Union (AU) and the Southern African Development Community (SADC). The AU praised the turnout but recommended that the Independent Electoral Commission find a solution to the problem of long queues in future elections. Thirteen parties won parliamentary representation. The ANC won 264 seats, just short of the two-thirds majority. The DA and the COPE followed with 67 and 30 seats respectively. The IFP won 18 seats. Other parties took fewer than four seats each. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) Post election interviews were conducted between October, 2nd and 31st as face-to-face interviews. Specially trained translators therefore translated the original CSES questionnaire into ten languages (Afrikaans, IsiXhosa / Xhosa, SePedi / North Sotho, SeSotho / South Sotho, SetTswana / Tswana, Shangaan, Swazi, IsiZulu / Zulu, Venda, Ndebele). The underlying population of the current sample is South African adults aged 18 years and older, excluding institutionalized citizens, such as military personnel. The sample is a random selection of households, proportional to the underlying population size. The primary sample units were the Census enumerator areas, stratified by the nine provinces, the dominance of racial groups and the region of residence (urban vs. rural). In sum, this stratification yields 72 strata in which households were randomly selected. Within the households, all adult citizens were numbered, whereas the head of the household randomly chose a number and thus selected the person to be interviewed. The South African interviewers were between 25 and 60 years old, all of them having a high school matriculation. Interviewers were trained by their supervisors within the provinces. About 37% of the realized interviews were verified by call-back. Overall, the response rate was about 48%. (3) According to the slightly overrepresentation of citizens in the province of Northern Cape, as well as of the Indian race, a sample weight is included to adjust the underlying population. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SOUTH KOREA (2008) --------------------------------------------------------------------------- (1) The April 2008 elections were the first to be held under the country's new President Lee Myung-bak of the Grand National Party (GNP), who was elected in December 2007. Prior to the 2008 elections, many new parties were formed. In August 2007, the Uri Party disbanded itself. Most of its parliamentarians joined the newly-created United New Democratic Party (UNDP), along with five former Democratic Party (DP) members. In October, other former Uri Party members formed the Renewal of Korea Party. After the 2007 presidential elections, the UNDP merged with the DP and renamed itself the "United Democratic Party" (UDP), led by Mr. Son Hak-gyu, while the Liberty Forward Party (LFP) merged with the People First Party (PFP). In late March, Ms. Park Geun-hye - daughter of former president Park Chung- hee and former GNP chairwoman - complained of the elimination of many of her supporters in the GNP's selection of parliamentary candidates. Many of her loyalists left the GNP to form a new party, Park's Party, while Ms. Park herself remained in the GNP. The economy and employment figured prominently in the campaign. The UDP promised to create more jobs through economic growth. It pledged to work towards affording greater protection to the country's low- and middle-income families. The GNP leader, Kang Jae-sup, called on voters' to retrieve "the 10 years the country lost under incompetent liberal governments". Its manifesto included university reforms that would give more autonomy to universities. The GNP won 153 seats, securing a majority in the National Assembly for the first time since 1998. The UDP took 81. The LFP came in third with 18 seats, followed by the Park's Party, which took 14 seats. Around 30 elected GNP candidates were reportedly loyal to Ms. Park. The DLP took five seats while the NPP failed to win any. The Renewal of Korea Party won three seats. Only 46 per cent of the 37.8 million eligible voters turned out at the polls, down from 60 per cent recorded in the 2004 elections." (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This study is a post-electoral, face-to-face study, conducted between April 10 and 21, 2008, immediately after the Parliamentary election of April 9, 2008. A multi-stage sampling procedure was employed: a) In a first stage 80 electoral districts (PSU’s) were on randomly chosen; b) Voting districts within each electoral districts were randomly sampled, and c) Sample of voters, within each voting district, were drawn from the list of voters provided by the Central Election Management Committee. The sample frame excluded Cheju island and other small islands, since people in these regions, who comprise a very small percentage of the total population (less than 0.5%), are difficult and costly to contact. (3) The original data, provided by South Korea, did not include weights; due to the sampling procedure, weights are not necessary to make the sample representative for the whole South Korean population. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SPAIN (2008) --------------------------------------------------------------------------- (1) On 20 December 2007 Prime Minister José Luis Rodríguez Zapatero announced that parliamentary elections would be held on 9 March. The bicameral parliament was subsequently dissolved on 14 January 2008. At stake were all 350 seats in the Congress of Deputies and the 208 directly elected seats in the 264-member Senate. As a result of a gender equality law all parties had a 50-50 gender balance for the candidates contesting the 2008 elections. Once again there was a duel between the PSOE and the PP. Due to the stagnating economy and a 12-year high inflation rate Mr. Rajoy (PP) accused the socialist government of not taking enough measures to boost the economy. Prime Minister Zapatero rebutted that his government was willing to use a budget surplus to revive the economy. Both PSOE and the PP promised tax cuts. The PP also called for a "contract" for immigrants. It argued that immigrants should respect Spanish customs and values and should be expelled if they failed to find work within a year of their arrival. Two days before the polls a former town councillor Mr. Isaias Carrasco of the PSOE was shot dead in the Basque region. Police blamed ETA for the shooting and the election campaign was briefly halted. The PSOE remained the largest party in the Congress of Deputies with 169 seats followed by the PP with 153. The PP took 101 seats of the 208 directly elected seats in the Senate 12 more than the PSOE. In all 127 women were elected to the Congress of Deputies while 67 were elected to the Senate. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) Interviews for the CSES survey took place in the period of April 11-21, 2008. The survey based on face-to-face interviews, on a random sample of national citizens, aged 18 years and older, registered as voters. The sample frame covered approximately 99% of the total population, although it excluded the two North African cities, Ceuta and Melilla, as well as institutionalized persons, at all. The interviews were distributed among the 17 Autonomous Regions in proportion to their population and to community size within each region. Municipalities with more than 500,000 inhabitants were all included, while the remaining units were randomly selected. Once the number of interviews to be done was established (by size of community and autonomous region), a computerized system to randomly extract municipalities and electoral sections was applied. The number of electoral sections was related to the total number of interviews to carry out in the municipality. Household selections based on a random route system. In case of refusal or non-contact, the interviewer went to the next household. Age and sex quotas were used for respondent selection. These quotas were established in each sampling point according to the ration between size of community and age and sex at the national level and regional level in the regions of Cataluńa, Andalucía, and País Valenciano. In the other regions, the distribution was proportional. Overall, the Spanish sample included 3954 households, out of which 1640 were not contactable. The responds rate was about 52.03%. (3) The Spanish data include a demographic weight, which compensates for disproportionate probability of the sample selection. The construction of the weight bases on age and gender distribution of the studied population. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SWEDEN (2006) --------------------------------------------------------------------------- (1) Sweden voted on 17 September 2006 to elect a new parliament. The country's membership of NATO and its generous welfare system were the main electoral issues in 2006. While the SAP argued for maintenance of neutrality, the Alliance for Sweden called for Swedish membership of NATO. The Alliance pledged to reform Sweden's cradle-to-grave welfare policy by cutting employers' taxes and unemployment benefits. The SAP accused the opposition of wanting to destroy the welfare state. The SAP lost 14 seats, although it remained the largest party with 130 seats. Conversely, the Moderate Party increased its seats to 97, winning 42 more seats than in the 2002 elections. The FP, badly damaged by the hacking scandal, lost 20 seats, winning 28. The CP and the Kd won 29 and 24 seats respectively, thus giving the Alliance for Sweden a narrow majority in the new parliament with 178 seats. Prime Minister Göran Persson accepted defeat and resigned on 17 September 2006. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The survey is part of a panel study of the Swedish National Elections Study, which has a two-step ongoing design in which half of the sample is interviewed in connection with the previous election, and the other half in connection with the succeeding election. However, for the CSES data, only the post-election part of the panel is being used. The sample is meant to be representative of all Swedish citizens living in Sweden and who are entitled to vote. Respondents were randomly selected at all stages of the sampling procedures. Excluded from the sample frame were the estimated 2 percent of the total eligible population, who live abroad. 1547 respondents were interviewed of which 870 answered the full questionnaire and 269 a shortened version. 408 respondents answered a very shortened extract. Which version a respondent answered was captured in variable C1007. Also see the election study note on C1007. (3) There are no post-survey adjustment weights, thus there are no weights included in the data set. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - SWITZERLAND (2007) --------------------------------------------------------------------------- (1) On 21 October 2007, elections were held for all 200 seats in the National Council and 41 of the 46 seats in the Council of States. The SVP/UDC has moreover taken up its usual discourse of cutting taxes, keeping Switzerland out of the European Union and addressing the issue of exploitation of the country's social and asylum policies. The election campaign was marked by an SVP/UDC poster showing a black sheep being kicked out of Switzerland by three white sheep. Other parties also used their traditional election platforms, with the Socialists (SP/PS) calling for social security and solidarity, the Radical Democrats (FDP/PRD) urging further economic growth based on creativity and efficiency and the Christian Democrats (CVP/PDC) pledging to ensure job opportunities. The Greens (GPS/PES) campaigned for combating climate change, calling for a ban on air shows in the country and the introduction of a jet fuel surcharge. In the elections to the National Council, the SVP/UDC and the GPS/PES made the biggest gains (seven seats each), bringing their seats to 62 and 20 respectively. The SVP/UDC obtained the largest share of seats in this chamber since the country's proportional voting system started in 1919. The SP/PS, which took 43 seats, suffered a major defeat, losing nine seats. The FDP/PRD also lost five, winning 31 seats. The remainder went to small parties. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) This is a post-election study, focused on the Swiss parliamentary election of October 21, 2007. The CSES survey module was conducted in three languages (German, French and Italian) and as a combination of a telephone and follow-up self completion survey (which could be completed either by as mail-back or online questionnaire). The follow-up survey was completed by all respondents from the telephone sample willing to answer a supplementary self-administered questionnaire. See variable C1007 to identify each sample component. Total panel attrition was 27.59%. Sample for the telephone survey was drawn from the telephone register of households selected by the federal office of statistics. The sample had three components: a) A national sample of households with 2000 target respondents; b) An over- sample in cantons (which correspond to the electoral districts) used to guarantee at least 100 respondents in each; c) An over-sample in 3 cantons used to guarantee at least 600 respondents in each (corresponding to Tessin, Zurich and Geneva). Within each household, respondents were randomly selected. Notice that the CSES module was only applied in sample (a) and (b), thus respondents from sample (c) in not included in the data set. Additionally, one canton (containing approximately 0.6% of the total population) was excluded from the sample since there was only one candidate, and thus, no election took place. (3) The Swiss study includes two different weights. The first one is a design weight that adjusts for oversampling of certain cantons (see point 2 for details of the sample design). The second weight is a cumulative political weight that adjusts for sample bias in turnout and party choice variables, in addition to the cantonal oversampling. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - TAIWAN (2008) --------------------------------------------------------------------------- (1) Unlike the past four presidential elections, the presidency in 2008 was contested by candidates from just the two major parties. With incumbent President Chen Shui-bian prevented by the constitutional term limit - two consecutive four-year terms - from running again, the DPP fielded two veteran heavyweight politicians, Frank Hsieh and Su Tseng-chang, as their presidential and vice-presidential candidates. Hsieh won the DPP’s primary, triumphing over several other senior DPP politicians. For the KMT, the charismatic former mayor of Taipei Ma Ying-jeou and former economic affairs minister Vincent Siew were chosen to represent the party. Two months after the KMT’s landslide victory in the legislative elections, its candidate Ma Ying-jeou won a decisive victory in the presidential election. The KMT’s resounding victory has been attributed to Ma Ying-jeou’s unsullied image, the unpopularity of the incumbent President Chen, and voters’ weariness with the political stalemate and deadlock of the past eight years. (Source: Tan. Alexander C. (2009): The 2008 Taiwan elections: Forward to the past? Electoral Studies 28, 492-517. (quoted are: 503-504; 505-506)). (2) The sample is meant to be representative of eligible voters in Taiwan's 2008 presidential election. Taiwanese citizens are eligible from 20 years of age onwards. Institutionalized persons were excluded from the sample frame. For the sampling, the country was divided into six geographic regions, then the proportions of samples for each region was estimated. 73 single-member districts of the 2008 legislative election served as primary sampling units (PSU). The secondary sampling units are villages or "li" and then, respondents were systematically drawn from household registration list provided by the Ministry of Interior. Based on the principle of probabilities proportional to the size (PPS), a total of 1,905 interviews were completed. Around 16 respondents were sampled from every selected village or li. (3) To generalize to the population of eligible voters in the 2008 presidential election, the survey data should be weighted by ranking with gender, age (5 groups), education (5 groups), and area (6 regions). The demographic characteristics are based on "2007 Taiwan-Fuchien Demographic Fact Book Republic of China", published by the Ministry of Interior Republic of China. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - THAILAND (2007) --------------------------------------------------------------------------- (1) The December 2007 elections were the first to be held following a military takeover in September 2006 which ousted the then Prime Minister, Shinawatra Thaksin. The adoption of a new Constitution on 19 August 2007 opened the way to a return to democracy. At stake in the 2007 elections were the 480 seats in the new House of Representatives. The TRT had been dissolved by the Constitutional Tribunal in May 2006, and exiled former prime minister Thaksin was barred from running in the 2007 elections, along with 111 former TRT executives. Mr. Thaksin's supporters formed the People's Power Party (PPP) prior to the elections, whose candidates included children of barred former TRT members. The PPP leader Mr. Samak Sundaravej pledged to allow Mr. Thaksin's return to the country so he could "prove his innocence". The PPP promised to give priority to economic recovery. The Democratic Party (DP, which took 96 seats in the 2005 elections), was led by Mr. Abhisit Vejjajiva. It pledged to fight corruption and improve the country's economic situation within 99 days. Its leader said his party would not join a PPP-led government. Preliminary results gave the PPP 233 seats and the DP 165 seats, although the results for several seats were invalidated following allegations of vote-buying. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The post election interviewing began at January 9, 2008, and ran until January 31, 2008, as face-to-face interviews. The current sample population was meant to be representative of all eligible Thai voters, aged 18 years and older and having the Thai citizenship. However, Thailand's electoral law excludes monks, novices, Brahmin priests and clergies from the right to vote. The sample was selected in a three-stage probability sample process. Based upon clusters of legislative districts, voting units have been selected in which voters were chosen by systematic sampling based on voter lists. Overall, out of the 145 legislative districts of Thailand, 50 have been selected, including 100 voting units in total. The respondents had not been informed before the first contact for an interview. Prior to the interview, a small gift (a pen) was given to respondents. Detailed information about the interviewer, their gender, age, training or any instruction were not available. (3) The original data, provided by Thailand, did not include any kind of weights. Furthermore, due to the sampling procedure, weights are not necessary to make the sample representative for the whole Thai population. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - TURKEY (2011): --------------------------------------------------------------------------- (1) On 3 March 2011, the Grand National Assembly of Turkey adopted a proposal by the Justice and Development Party (AKP) for elections to be held on 12 June, two months earlier than constitutionally due. In 2011, Prime Minister Erdogan was seeking to win a third term in office. His party's main rivals were the CHP, led by Mr. Kemal Kilicdaroglu, and the MHP, led by Mr. Devlet Bahceli. In all, 15 parties and 200 independent candidates were vying for seats. Political parties need to win over 10 per cent of the valid votes nationwide, while independent candidates need to win over 10 per cent of the valid votes in the province where they stand for election. On 14 May, the Prime Minister announced that drafting a new constitution would be his key priority after the 2011 elections. The country's current Constitution was written shortly after the September 1980 military coup. Erdogan urged voters to give the AKP the two-thirds majority (367 seats) required to modify the Constitution. The opposition CHP criticized the Prime Minister for what it termed despotic ambitions. It argued that the new constitution should be drafted in consultation with all parties. It urged voters to support the CHP to prevent the AKP from drafting the new constitution single-handedly. The CHP presented its own draft constitution, promising to ensure more rights for minorities. The MHP promised to work for national unity, which, in its view, had been threatened by the AKP government's overtures towards the Kurds. In the run-up to polling day, many Syrian refugees fleeing a crackdown crossed Turkey's southern border. On 8 June, Prime Minister Erdogan said that Turkey would not close its borders to refugees coming from Syria. Transparent plastic ballot boxes were used for the first time instead of the wooden boxes used previously, in a move to prevent fraud. Three parties secured the required 10 per cent of the valid votes nationwide to win parliamentary representation. The AKP won 326 seats, 41 short of the two-thirds majority. The CHP increased its share from 112 to 135 seats. Inversely, the MHP saw its share reduced from 71 to 53 seats. Thirty-six independent candidates, including 35 backed by the BDP, were elected. In all, 78 women were elected. On 22 June, the Supreme Election Board cancelled the parliamentary mandate of nine MPs-elect. They had campaigned from prison: six independent candidates backed by the BDP, two CHP members and one MHP member. The Constitution stipulates that persons sentenced to prison terms of one year or more cannot be elected to parliament. The said MPs-elect were all serving prison terms exceeding one year. Many opposition members announced that they would boycott the swearing-in of the newly elected parliament unless all the jailed deputies were allowed to join parliament. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The Turkish survey for the current CSES module was administrated between July, 21st and August 26th 2011, as face to face interviews. The target population of sample was non-institutionalized Turkish citizens, aged 18 years and older, excluding about 2.4% of the overall population. The sampling procedure adopted starts with the use of Turkish Statistical Institute’s (TUIK) NUT-1 regions. The target sample was distributed according to each region’s share of urban and rural population in accordance with the Address Based Population Registration System (ADNKS) records as of the end of 2010. Next, TUIK’s block data were used with block size set at 400 residents. Twenty voters were targeted to be reached from each block and no substitution was used. The probability proportionate to population size (PPPS) principle was used in distributing the blocks to NUT1 regions. For each of the 20 addresses up to three visits were carried out with the expectation that approximately 50% of the addresses will have a completed interview. In some rural areas where TUIK was unable to provide addresses we contacted the headman (Muhtar) of the village and obtained 20 addresses in a systematic random sample selection from the list of households in the village. Selection of individuals in households is done on the basis of reported target population of 18 years or older in each household according to a lottery method. Names of all reported individuals of 18 years or older are written on cards and a person in the household then is asked to randomly select one card with the name of the individual to be interviewed. Interviewers were between 20 to 45 years old with higher levels of education. They were trained in house by the Turkish researchers. About one third of the realized interviews were verified by phone calls, In addition, another 20% of the interviews were supervised during the fieldwork. In total 2440 households were sampled, out of which 2102 addresses remained as valid households. In 1109 of them interviews were successfully completed, i.e. the response rate was about 52.76%. (3) The Turkish subsample of the CSES includes a demographic weight to control for a selection bias. It bases on the total population’s distribution according to three age groups (18-34, 35-54 and 55+), gender, and six educational levels (illiterate, literate but no school completed, primary 5 years, primary 8 years, high school, and university graduates). --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - UNITED STATES (2008): --------------------------------------------------------------------------- (1) The November 2008 congressional elections were held jointly with the presidential elections. At stake were all 435 seats in the House of Representatives and 35 seats in the Senate. The 2008 elections were held against the backdrop of a financial crisis, the worst since the 1929 Great Depression. The budget deficit hit a record US$ 455 billion in the fiscal year ending in September 2008. Both the Democrats and the Republicans focused on stabilizing the US economy. Other main issues included the deployment of US troops overseas and health care reform. In the elections to the House of Representatives the Democrats won 257 seats while Republicans took 178. In the Senate elections the Democrats won 20 seats to hold 59 seats in all (including support from two independents) while Republicans took 15 (41 in all). In the presidential elections Mr. Obama secured 53 per cent of the popular vote (or 365 votes in the electoral college) becoming the first African American person to accede to the highest office in the USA. (Source: PARLINE database: http://www.ipu.org/parline-e/parline.htm) (2) The 2008 American National Election Study is a pre-, post-election survey. It was conducted on a multi- stage area probability sample intended to be representative of the United States. The entire target population constitutes all English- and Spanish-speaking citizens of voting age. The ample was conducted using a five stage sampling process: a primary stage sampling of U.S. Metropolitan Statistical Areas (MSAs) and non-MSA counties, followed by a second stage sampling of census tracts (CTs). A third stage sampling referred to Census block groups (CBGs), containing 550 household on average. In a fourth stage, these blocks were controlled for valid residential mailing addresses. Finally, respondents were selected within the sampled household, based on a stratification of Latinos, African American and others. Pre-election interviews were conducted September 2 through November 3, 2008. The post-election interviews were held between November, 5th and December, 30th. On average a period of 54 days lied in between both interviewing dates, with maximum of 116 days. Variables, included in the CSES surveyed in the pre- (socio-demographics) and post-election (survey variables) interviews. Overall, 116 interviews were hired for the post- election survey, out of which 24 persons were trained bilingual. 100% of the completed questionnaires were verified, based on audio records of the interviews. (3) The US data include a sample weight to make the post-election survey representative for the US target population. Based on age and education, it controls for a sampling bias, caused by non-response. --------------------------------------------------------------------------- >>> ELECTION SUMMARIES AND GENERAL NOTES - URUGUAY (2009) --------------------------------------------------------------------------- (1) At stake in the 2009 elections were all 99 seats in the House of Representatives and the 30 seats in the Senate. They were held in parallel with the presidential elections. In March 2009, the General Assembly (Senate plus House of Representatives) approved a law to increase the percentage of women on the candidates' lists for elections at all levels. However, it will only apply as of the parliamentary elections due in 2014. In the 2009 elections, Senator José Mujica was elected in Preliminary Elections as FA presidential candidate. A former member of the rebel Tupamaros National Liberation Movement (MLN), he was challenged by former president Luis Alberto Lacalle (PN) and Mr. Pedro Bordaberry (PC). The FA pledged to solve housing problems, create jobs and fight poverty. The PN pledged to work for security, employment and independence. It also promised to remove the income tax introduced by the FA government. PC leader Bordaberry promised a lower tax burden for people and promised to establish a merit-based pay system. The FA retained the majority in both chambers, taking 50 seats in the House of Representatives and 16 in the Senate. The PN took 30 and nine seats respectively. The PC followed with 17 and five. The PI, led by Mr. Pablo Mieres, obtained two seats in the House of Representatives. Fourteen women were elected to the House of Representatives and four to the Senate. No candidate won the required majority in the presidential elections to be elected in the first round. The run-off elections between Mr. Mujica (FA) and Mr. Lacalle (PN), backed by the PC, were held on 29 November. Mr. Mujica (FA) was elected with over 52 per cent of the votes. Source: http://www.ipu.org/parline-e/reports/2341_B.htm (Parline), which has been revised by the Factum Institute (Uruguayan collaborators) (2) Face-to-face interviews took place between August 13th and August 31st, 2010. The sampling process required the Uruguayan citizenship, as well as a minimum age of 18 years. The sample relies on three primary units, the department of Montevideo (400 cases), the metropolitan area of Montevideo (97 cases) and the rest of the country, including 18 further departments (471 cases), reflecting the overall distribution of Uruguayan citizens. Based these three areas, household sampling was randomly, whereat respondent’s selection with a chosen household rested up on the Kish grid method. Interviewers contacted each household only once and declared it as non-sampled after an unsuccessful effort. In sum, 30 interviewers participated throughout Uruguay. All of them were university students, trained for the CSES interviews in personal sessions. Survey verifications included about 15% of the interviews, by telephone checking. A total of 968 interviews were completed. (3) A political weight relies on corrections according to the official election results. It is based on the electoral outcome of the parliamentary and presidential elections and controls for the distribution of C3023_LH_PL. (For further details, see CSES Design Report for Uruguay 2009.) =========================================================================== ))) BIBLIOGRAPHY =========================================================================== --------------------------------------------------------------------------- >>> WEBSITES --------------------------------------------------------------------------- http://aceproject.org/ace-en/topics/es/esy/esy_ch ACE Electoral Network. Switzerland. http://africanelections.tripod.com/za.html African Election Database http://www.electionstudies.org/studypages/download/datacenter_all.htm American National Election Studies (ANES) Data Center. http://www.andmevara.ee/pr AS Andmevara (Estonian Population Register). http://www.aec.gov.au/ Australian Election Commission. http://wahl08.bmi.gv.at/ Austrian Ministry of Interior http://www.bundeswahlleiter.de/ Bundeswahleiter (The Federal Returning Officer). http://web.cvk.lv/pub/public/29776.html The Central Election Commission of Latvia http://www.rec.gov.by/en/ENG-Central_Commission Central Commission of the Republic of Belarus. http://vote.nccu.edu.tw/engcec/A200800E.PDF Central Level Public Officials, presidential election Taiwan 2008 http://www.elections.gov.hk/legco2008/eng/ Chief Electoral Officer of Hong Kong 2008. https://www.cia.gov/library/publications/the-world-factbook/geos/ks.html The CIA World Factbook. http://www.comelec.gov.ph/ Commission on Election of the Republic of the Philippines. http://www.cvk.lv/cvkserv/constitution_2007.pdf The Constitution of the Republic of Latvia http://www.sejm.gov.pl/prawo/konst/konst.htm The Constitution of the Republic of Poland (Konstytucja Rzeczypospolitej Polskiej). http://www.vjesnik.hr/Izbori_2007/Default.asp Croatian Political Diary - Croatian Electoral District Results. http://www.volby.cz/index_en.htm Czech Statistical Office http://www.verkiezingsuitslagen.nl/Na1918/ Verkiezingsuitslagen.aspx?VerkiezingsTypeId=1 De Kiesraad (contains election results) http://www.dpes.nl/ The Dutch Parliamentary Electoral Studies website. http://www.elections.ca/home.aspx Election Canada http://www.elections.org.za/content/ Electoral Commission of South Africa (IEC) http://www.ect.go.th/newweb/en/home Electoral Commission of Thailand http://www.ect.go.th/newweb/Files/law_eng/ The%20Annoucement%20of%20ECT%20%28cut%29.pdf Electoral Commission of Thailand (2007): The Announcement of the Election Commission of Thailand Subject: The Number of Members of the House of Representatives (MPs) and Constituencies of Each Province on the Constituency-basis. http://www.kiesraad.nl/English-Home.html The Electoral Council of the Netherlands (The Dutch Electoral Act and explanation). http://www.corteelectoral.gub.uy Electoral Court of Uruguay (Corte Electoral). http://www.electoralgeography.com/new/en/countries/c/croatia/croatia- legislative-election-2007.html Electoral Geography 2.0. http://www.eisa.org.za/WEP/sousystem.htm Electoral Institute for Sustainable Democracy in Africa. http://electionresources.org/ Election Resources on the Internet. http://www.bundeswahlleiter.de/de/bundestagswahlen/rechtsgrundlagen/ Essential legal bases of Bundestag elections. http://www.siseministeerium.ee/11577/index.php Estonian Ministry of Interior. http://www.vvk.ee/ Estonian National Electoral Committee. http://www.nsd.uib.no/european_election_database/country/iceland/ European Election Database (NSD), Icelandic elections http://www.factum.edu.uy Factum Digital (de Instituo Factum, Uruguay). http://www.fec.gov/ Federal Election Commission of the USA. http://www.bfs.admin.ch/bfs/portal/de/index/themen/17.html Federal Statistical Office of Switzerland (Statistik Schweiz). http://www.vaalit.fi/53561.htm Finnish Ministry of Justice - Electoral Website. http://www.interieur.gouv.fr French Electoral Commission (Minister of Interior). http://www.oireachtas.ie/viewdoc.asp?DocID=8153 General Election for the National Parliament of the Republic of Ireland 2007. http://www.government.gov.sk/ Government Office of the Slovak Republic. http://ekloges-prev.singularlogic.eu/ Hellenic Ministry of Interior: Singular Logic. http://www.voterregistration.gov.hk/eng/statistic.html Hong Kong Registration and Electoral Office. http://www.idea.int/ IDEA website. http://www.electionguide.org IFES Election Guide. http://www.ibge.gov.br Instituto Brasileiro de Geografia e Estatística. http://www.uis.unesco.org/TEMPLATE/pdf/isced/ISCED_A.pdf International Standard Classification of Education - ISCED-1997. http://knesset.gov.il/main/eng/home.asp The Knesset. http://www.ife.org.mx Mexican Institute of Federal Elections. http://clerk.house.gov/member_info/electionInfo/2008election.pdf. Miller, Lorraine C.: Statistics of the Presidential and Congressional Election of November 4, 2008. http://www.minv.sk/?nr10-uvk Ministry of Interior of the Slovak Republic. http://www.nrsr.sk/ National Council of the Slovak Republic. http://www.nec.go.kr National Election Commission of Republic of Korea (NEC). http://www.wybory2005.pkw.gov.pl/index_EN.html National Election Commission of Poland 2005. http://www.wybory2007.pkw.gov.pl/SJM/EN/WYN/M/index.htm National Election Commission of Poland 2007. http://www.dvk-rs.si/index.php/si/ National Electoral Commission, Republic of Slovenia. http://www.inegi.org.mx National Institute of Statistics and Geography of Mexico. http://www.nist.gov/index.html National Institution of Standards and Technology (NIST). http://www.oireachtas.ie/parliament/ National Parliament of the Republic of Ireland (The Houses of the Oireachtas). http://www.electionresults.govt.nz/electionresults_2008/e9/html /statistics.html New Zealand elections website. http://www.electionresults.govt.nz/electionresults_2008/e9/html/ e9_part1.html New Zealand Electoral Commission (election results). http://www.osce.org/odihr/elections/latvia/74785 Latvia Parliamentary Elections 2 October 2010, OSCE/ODIHR Limited Election Observation Mission Final Report. OSCE, Warsaw, 2010. http://www.osce.org/odihr/elections/71251 OECD Election assessment mission; Observation of Parliamentary Elections in the Netherlands. http://www.eleicoes.mj.pt/legislativas2009/territorio-nacional.html Official election results 2009 by the Portuguese Minister of Interior. http://www.fec.gov/pubrec/fe2008/2008presgeresults.pdf 2008 Official Presidential General Election Results. http://aceproject.org/regions-en/countries-and-territories/HR/reports/ croatia-report-of-the-state-election-commission-on/at_download/file Official report of the Croatian State Election Commission 2007. http://www.sabor.hr/Default.aspx Official Website of the Croatian Parliament. http://www.eac.gov.hk/en/about/chairman.htm Official Website of the Chairman of the Electoral Affairs Commission of Hong Kong 2008. http://www.diputados.gob.mx Official Website of the Mexican House of Deputies. http://www.senado.gob.mx Official Website of the Mexican Senate. http://www.web.onpe.gob.pe/ Oficina Nacional de Procesos Electorales (National Office of Electoral Processes - ONPE) http://www.osce.org Organization for Security and Co-operation in Europe (OSCE). http://www.ft.dk/Dokumenter/Publikationer/Engelsk/Parliamentary_System.aspx The Parliamentary System of Denmark (English). http://www.ipu.org/parline/parlinesearch.asp PARLINE database on national parliaments. http://www.verkiezingsprogramma.info/ Party platforms of Ditch political parties: http://www.census.gov.ph/ Philippine National Statistical Office. http://www.pkw.gov.pl Polish Electoral Commission (Panstwowa Komisja Wyborcza). http://www.sejm.gov.pl/english/sejm/pos.htm Polish Electoral law. http://www.dgai.mai.gov.pt/ Portuguese Minister of Interior. http://www.ine.pt Portuguese National Statistics Office. http://www.archives.gov/federal-register/electoral-college/2008 /popular-vote.html 2008 Presidential Election USA, Popular Vote Totals. http://psephos.adam-carr.net Psephos - Adam Carr's Election Archive. http://www.qog.pol.gu.se/ QoG-Institute (Quality of Governemt Institute). http://www.eac.gov.hk/pdf/legco/2008/en/report/2008lce_full_lc_report_e.pdf Report on the 2008 Legislative Council Election of Hong Kong. http://www.roaep.ro Romanian Electoral Authority. http://www.insse.ro Romanian National Institute of Social and Economic Statistics (INS). http://www.saarpoll.ee/ Saar Poll Company (Estonia). http://web.cvk.lv/pub/public/28126.html The Saeima Election Law. http://www.servel.cl/controls/neochannels/neo_ch1/neochn1.aspx Servicio Electoral Republica de Chile. http://volitve.gov.si/dz2008/en/ Slovenian Ministry of Public Administration. http://archive.fairvote.org/reports/1995/spot4/lijphart.html The South African Electoral System. http://www.infoelectoral.mir.es/min/ Spanish Minister of Interior. http://www.statcan.gc.ca/subjects-sujets/standard-norme/noc-cnp/2011/cnp- noc2011-citp-isco2008-eng.htm Statistics Canada: Concordance: National Occupational Classification (NOC) 2011 to International Standard Classification of Occupations (ISCO) 2008. http://www.ssb.no/stortingsvalg_en/ Statistics Norway (Statistik entralbyra). http://www.regjeringen.no/krd/html/valg2009/bs5.html Statistics Norway, Election results 2009. http://www.dzs.hr/default_e.htm Statistical Office of Croatia. http://www.dst.dk/ Statistical Office of Denmark (Danmarks Statistik). http://clerk.house.gov/member_info/electionInfo/2008election.pdf Statistics of the Presidential and Congressional election of November 4, 2008. Compiled from official sources by Lorraine C. Miller. Washington: 2009. http://app.statistics.sk/nrsr_2010/sr/tab1.jsp?lang=en Statistical Office of the Slovak Republic. http://www.tse.jus.br/eleicoes/eleicoes-anteriores Superior Electoral Court - Brazil http://www.tse.gov.br Supreme Electoral Tribunal of Brazil. http://www.val.se/val/val2006/slutlig/R/rike/delar.html Swedish electoral results, 2006 Riksdagsvalkretsar http://www.ysk.gov.tr/ysk/index.html Turkish Electoral Board Website http://www.turkstat.gov.tr/Start.do;jsessionid= L8h1Q0wG4bGMTdgvWc75g3Tzn4y57h1lKcf9M0HTZN7t0JkhpNjJ!-31035612 Turkish Statistical Institute (TUIK) http://www.census.gov/hhes/www/socdemo/voting/publications/p20/2008 /tables.html United States Census Bureau. Voting and Registration in the Election of November 2008 - Detailed Tables. http://elections.gmu.edu/Turnout_2008G.html The United States Elections Project. 2008 General Election Turnout Rates. http://www.archives.gov/federal-register/electoral-college/links.html U.S. Electoral College Home. U.S. Voting & Election Resources. http://www.nsd.uib.no/macrodataguide/set.html?id=11&sub=1 World Bank Dataset. --------------------------------------------------------------------------- >>> ARTICLES AND BOOKS --------------------------------------------------------------------------- Aalberg, Toril and Tove Brekken (2006). Norway. European Journal of Political Research. 45, 1221 - 1230. Aardal, B. (2011). "The Norwegian Electoral System and its Political Consequences." World Political Science Review 7, 1-31. Aarts, Kees, and Henk van der Kolk. "The Parliamentary Election in the Netherlands, 22 November 2006". Electoral Studies 26, 832-837. ACEEEO (ed) (2010). Elections in Europe. Budapest: ACEEEO. Altman, David (2010). "The 2009 elections in Uruguay". Electoral Studies, 29, 521-540. Antic, Miljenko and Maja Dodic Gruicic (2008). "The Parliamentary Election in Croatia, November 2007". Electoral Studies 27, 740-773. Antic, Miljenko (2006). "The Presidential Election in Croatia, January 2005". Electoral Studies 25, 147-191. Aurelian Muntean, Grigore Pop-Eleches and Marina Popescu (2010). "The 2009 Romanian presidential election". Electoral Studies 29, 753-757. Beltrán, Ulises (2009). "Ideología y polarización en la elección de 2006" ["Ideology and Polarization in the 2006 Election"]. Política y Gobierno, Volumen Temático 2009, Elecciones en México. Bottinelli, Oscar A. (2008). El Múltiple Voto Simultáneo y el Voto Conjunto en Uruguay [The Multiple Simultaneous Vote and the Join Vote]. Bottinelli, Oscar A., W. Giménez and J. L. Marius (2008). Enciclopedia Electoral del Uruguay - Digesto Electoral del Uruguay [Electoral Encyclopedia of Uruguay - Uruguay Election Digest]. Brunell, Thomas L. (2009). "The presidential and congressional elections in the USA, November 2008". Electoral Studies 28, 322-325. Burns, Nancy and Ian McAllister (2008). "The Dynamics of Political Choice: the Third Module of the Comparative Study of Electoral Systems (CSES)". Proposal to the American National Science Foundation, January 15, 2008. Bútora M., Kollár M. and G. Meseznikov (eds) (2010). Slovakia 2009. Trends in Quality of Democracy. Institute for Public Affairs, Bratislava. Chambers, Paul (2009). "Superfluous, Mischievous or Emancipating? Thailand’s Evolving Senate Today". Journal of Current Southeast Asian Affairs, 28, 3-80. Cheibub, Jose. "Reforming Presidential and Semi-Presidential Democracies". Available at: http://www.bibliojuridica.org/libros/6/2748/5.pdf. Deloy, Corinne (2011). "General Elections in Estonia, 6th March 2011". Fondation Robert Schuman, European Elections Monitor. Available at: http://www.robert-schuman.eu/doc/oee/oee-679c-en.pdf. Denmark Statistik (2007). "Befolkning og Valg. Folketingsvalget den 13. November 2007". Statistiske Efterretninger. Availible at: http://www.dst.dk/valg/SE_FV2007.pdf. Diskin, Abraham and Reuven Y. Hazan (2007). "The Knesset election in Israel, March 2006". Electoral Studies. 26, 699-724. Elklit, Jřrgen, Anne Birte Pade and Nicoline Nyholm Miller (2011). "The Parliamentary Electoral System in Denmark". Ministry of the Interior and Health and The Danish Parliament, Copenhagen. Available at: http://www.ft. dk/Dokumenter/Publikationer/Engelsk/Electorial_System_Denmark.aspx. Espirito Santo, Paula do (2010). "The 2009 Portuguese parliamentary elections". Electoral Studies 29, 276-296. Fallend, Franz (2009). "Austria". European Journal of Political Research, 48, 884-902. Fish, M., Steven Kroenig and Matthew Kroenig. "The Handbook of National Legislatures". Cambridge University Press. 2009. Gallagher, Michael and Michael Marsh (eds) (2008). How Ireland Voted 2007: The Full Story of Ireland's General Election. Bassingstoke, Palgrave Macmillan. Gherghina, Sergiu, and George Jiglau (2012). "Where does the Mechanism Collapse? Understanding the 2008 Romanian Electoral System". Representation 48, 445-459. Gwiazda, Anna (2008). "The parliamentary election in Poland, October 2007". Electoral Studies 27, 740-773. Gyárfásová, Olga et al (2011). Visegrad Elections 2010: Domestic Impact and European Consequences. Bratislava: IVO. Hardarson, Ólafur Th. and Gunnar Helgi Kristinsson (2008). "Iceland". European Journal of Political Research 47, 1005-1011. Hardarson, Ólafur Th. and Gunnar Helgi Kristinsson (2010). "Iceland", in Dieter Nohlen and Philip Stöver (eds.), Elections in Europe. Baden-Baden: Nomos. Helms, Ludger (2007). "The German federal election, September 2005". Electoral Studies 26, 223-227. Het Grote Partijprogramma Boek 2010. Alle Originele en Complete Programma’s op een Rijtje. Amsterdam: Rainbow, 219-244. Hix, Simon, Ron Johnston and Iain Mclean (2010). Choosing an Electoral System. London: The British Academy. Jasiewicz, Krzysztof and Agneszka Jasiewicz-Betkiewicz (2006). "Poland" European Journal of Political Research 45, 1231-1246. Linek, Lukas (2007). "Czech Republic". European Journal of Political Research 46, 929-937. Linek, Lukas (2008). "Czech Republic". European Journal of Political Research 47, 947-951. Lupu, Noam (2012). "The 2011 general elections in Peru". Electoral Studies 31, 621-624. Macherras, Malcom (2008). "Australia". European Journal of Political Research 47, 892-901. Mainwaring, Scott, Rachel Meneguello and Power, Timothy (2000). "Conservative Parties, Democracy, and Economic Reform in Contemporary Brazil". In Kevin J. Middlebrook (ed.), Conservative Parties, the Right, and Democracy in Latin America. Baltimore, The Johns Hopkins University Press. Marian, Cosmin Gabriel and Ronald F. King (2010). "Plus ça change: Electorallaw reform and the 2008 Romanian parliamentary elections". Communist and Post-Communist Studies 43, 7-18. Milic, Thomas. (2008). "Switzerland". European Journal of Political Research 47, 1149-1155. Millard, Frances (2007). "The 2005 parliamentary and presidential elections in Poland". Electoral Studies 26, 196-231. Mueller, Wolfgang C. (2005). "Austria: A Complex Electoral System with Subtle Effects". In Michael Gallagher and Paul Mitchell (eds.), The Politics of Electoral Systems. Oxford University Press. National Statistical Committee of the Republic of Belarus (2009). Socio-economic conditions of households in the Republic of Belarus. Nicolau, Jairo (2004) "Brazil: Democratizing with Majority Runoff". In Josep M. Colomer (ed.), Handbook of Electoral System Choice, London, Palgravre, MacMillan. Nicolau, Jairo (2007). The open-list electoral system in Brazil. Dados 3, Selected Edition 2007. Available from http://socialsciences.scielo.org/ scielo.php?script=sci_arttext&pid=S0011-52582007000100003&lng=en&nrm=iso>. Nicolau, Jairo (2008). "The presidential and congressional elections in Brazil, October 2006". Electoral Studies 27, 170-175. OSCE/OHIHR (2009). Election Assessment Mission Report, Iceland early parliamentary elections 25 April 2009. Warsaw. Pedersen, Susannah (2011). My Constitutional Act with explanations. The Communications Section, Danish Parliament, Copenhagen. Raunio, Tapio. "Finland: One Hundred Years of Quitude" in Michael Gallagher and Paul Mitchell (eds.) The Politics of Electoral Systems. Oxford University Press, 473-489. Schafferer, Christian (2009). "The Parliamentary Election in Thailand, December 2007". Electoral Studies 28, 141-173. Solsten, Eric ed. (1994). "Austria: A Country Study." Washington: GPO for the Library of Congress. http://countrystudies.us/austria/116.htm]. Spoon, Jae-Jae (2008). "Presidential and legislative elections in France, April - June 2007". Electoral Studies 27, 151-190. Sundberg, Jan (2008). "Finland". European Journal of Political Research 47, 1221-1230. Tan, Alexander C. (2009). "The 2008 Taiwan elections: Forward to the past?" Electoral Studies 28, 492-517. Toro, Sergio Y. and Luna, Juan Pablo (2011). "The Chilean elections of December 2009 and January 2010". Electoral Studies 30, 226-230. Tweede Kamerverkiezing 22 november 2006. Kiesraad, Den Haag, 2007. Van der Kolk, Henk, Kees Aarts, Martin Rosema and Martha Brinkman (2007). Dutch Parliamentary Election Study 2006. Enschede: University of Twente. Volkens, Andrea et al (2009). The Manifesto Data Collection. Manifesto Project (MRG/CMP/MARPOR). Berlin: Wissenschaftszentrum Berlin für Sozialforschung. Zsuffa, István (2010). Developing accurate voter lists in transitional democracies. Budapest: ACEEEO. //END OF FILE