=========================================================================== COMPARATIVE STUDY OF ELECTORAL SYSTEMS (CSES) INTEGRATED MODULE DATASET (IMD) CODEBOOK PART 1: CSES IMD - INTRODUCTION PHASE 2 RELEASE - OCTOBER 17, 2019 CSES Secretariat www.cses.org =========================================================================== HOW TO CITE THE STUDY: The Comparative Study of Electoral Systems (www.cses.org). CSES INTEGRATED MODULE DATASET PHASE 2 RELEASE [dataset and documentation]. October 17, 2019 version. doi:10.7804/cses.imd.2019-10-17. These materials are based on work supported by the American National Science Foundation (www.nsf.gov) under grant numbers SES-0817701, SES-1154687, SES-1420973, and SES-1760058, 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 numerous organizations that fund national 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. =========================================================================== =========================================================================== TABLE OF CONTENTS =========================================================================== ))) IMPORTANT NOTE REGARDING PHASE 2 RELEASE OF THE CSES INTEGRATED MODULE DATASET (IMD) ))) OVERVIEW OF "CODEBOOK PART 1: CSES IMD - INTRODUCTION" ))) LIST OF TABLES IN CODEBOOK PART 1 ))) THE COMPARATIVE STUDY OF ELECTORAL SYSTEMS (CSES) PROJECT OVERVIEW >>> CSES PROJECT PROFILE >>> CSES MODULE 5 PLANNING COMMITTEE >>> CSES MODULE 6 PLANNING COMMITTEE >>> CSES SECRETARIAT ))) CSES INTEGRATED MODULE DATASET (IMD): HOW TO ACCESS? >>> THE CSES IMD CODEBOOK >>> THE CSES IMD DATA FILES ))) CSES INTEGRATED MODULE DATASET (IMD): PRODUCT OVERVIEW >>> CSES INTEGRATED MODULE DATASET (IMD) STUDY DESCRIPTION - ABSTRACT >>> OVERVIEW OF CSES IMD DATA FILE PARTICULARS >>> LIST OF ELECTION STUDIES AND POLITIES INCLUDED IN CSES IMD >>> THE ROAD TO CSES IMD ))) CSES INTEGRATED MODULE DATASET (IMD) DOCUMENTATION: WHAT'S AVAILABLE AND HOW TO USE? >>> CSES IMD CODEBOOK OVERVIEW >>> CSES IMD CODEBOOK CONVENTIONS >>> CSES IMD CODEBOOK - VARIABLE NOTES, POLITY NOTES & ELECTION STUDY NOTES >>> CSES IMD - ADDITIONAL DOCUMENTATION >>> HOW TO NAVIGATE THE CSES IMD CODEBOOK ))) CSES INTEGRATED MODULE DATASET (IMD): CODING CONVENTIONS AND ADDITIONAL INFORMATION FOR USERS ABOUT DATASET >>> CSES CONVENTION OF PARTIES/COALITION CODING IN STANDALONE CSES MODULES >>> CSES CODING OF PARTY/COALITIONS AND LEADERS IN CSES IMD - A NEW DEPARTURE >>> PLACEMENT OF QUESTIONS IN CSES IMD AND STANDALONE CSES MODULES >>> RECODING OF VARIABLES FOR CSES IMD >>> DEVIATIONS FROM STANDALONE CSES MODULES >>> IDENTIFICATION VARIABLES >>> MISSING DATA >>> WEIGHTS >>> FREEDOM STATUS OF ELECTIONS >>> PROCESSING CHECKS OF INTEGRATED MODULE DATASET (IMD) BY THE CSES SECRETARIAT >>> HOW TO MERGE STANDALONE CSES MODULE DATA NOT INCLUDED IN CSES IMD TO THE CSES IMD FILE? >>> HOW TO SELECT DATA FROM ONE OR MULTIPLE STANDALONE CSES MODULES OR ELECTIONS HELD IN A PARTICULAR YEAR IN CSES IMD? ))) CSES IMD BIBLIOGRAPHY =========================================================================== ))) IMPORTANT NOTE REGARDING PHASE 2 RELEASE OF THE CSES INTEGRATED MODULE DATASET (IMD) =========================================================================== This dataset and all accompanying documentation comprises the Phase 2 Release of the CSES Integrated Module Dataset (IMD). The CSES Integrated Module Dataset is being released in a phased way. Thus, this release is an advance version of the dataset, and thus lacks some of the checking, documentation, and inclusion of certain variables that are expected with the Full Release of this dataset. This Phase 2 Release is provided as a service to the CSES user community, for those analysts who find it valuable to work with preliminary versions of the dataset. We would appreciate being notified of any errors in the dataset or documentation by email to "cses@umich.edu". Users should expect future changes and improvements to the data and documentation of variables. If users wish to re-use their programming code on a future release of the file, the code should be written in a way that is flexible and can be accommodating of these future changes. Users of the CSES Integrated Module Dataset (IMD) may wish to monitor the errata for CSES IMD on the CSES website, to check for known errors which may impact their analyses. To view errata for CSES IMD Phase 2, go to Data Download on the CSES website, navigate to the CSES IMD download page, and click on the Errata link in the white box to the right of the page. Users of the CSES Integrated Module Dataset (IMD) are also advised that while the CSES IMD Codebook is comprehensive, users are strongly advised to consult Standalone CSES Module Documentation including Standalone CSES Module Codebooks, Macro Reports, Design Reports, and Election Summaries. =========================================================================== ))) OVERVIEW OF "CODEBOOK PART 1: CSES IMD - INTRODUCTION" =========================================================================== Part 1 of the CSES IMD Codebook provides users with information about how to use the CSES IMD Codebook and Data, including the conventions adopted in the CSES IMD Codebook and how to load the data into various programs. Moreover, we provide an overview of all the national election studies included in CSES IMD, as well as CSES conventions concerning identification variables, missing data, weights, and coding of parties/coalitions and leaders is available here. Further, Part 1 of the CSES IMD Codebook provides users with an overview of the CSES project and details of the CSES governing council, namely the CSES Planning Committee and the Financial Support for the project. Details of the national collaborators for each election study are available in the Standalone CSES Module Codebooks. =========================================================================== ))) LIST OF TABLES IN CODEBOOK PART 1 =========================================================================== Below, we list the Tables located in CSES IMD Codebook Part 1. Tables can be accessed in the electronic version of the CSES IMD Codebook by searching for "+++". - OVERVIEW OF ELECTION STUDIES INCLUDED IN CSES IMD WITH NUMBER OF OBSERVATIONS, MODE OF DATA COLLECTION, AND FIELDWORK DATES - OVERVIEW OF POLITIES INCLUDED IN CSES IMD AND NUMBER OF ELECTION STUDIES AND YEAR OF ELECTION - OVERVIEW OF POLITIES INCLUDED IN CSES IMD AND STANDALONE CSES MODULES THEY APPEAR IN =========================================================================== ))) THE COMPARATIVE STUDY OF ELECTORAL SYSTEMS (CSES) PROJECT OVERVIEW =========================================================================== --------------------------------------------------------------------------- >>> 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 standard 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 CSES project focuses on respondents' behavior and attitudes during the time of a national election, with a particular emphasis on voting and turnout. Each CSES Module consists of a nationally-representative post-election survey and additional variables about the context of the overall country and electoral system within which the respondents find themselves. Every five years, a new CSES Module is designed with a different substantive theme selected to address essential questions in electoral studies and social science. As of 2016, four modules have been completely fielded. They are: - MODULE 1: THE IMPACT OF ELECTORAL INSTITUTIONS ON CITIZENS' POLITICAL COGNITION AND BEHAVIOR (1996-2002) - MODULE 2: THE LOGIC OF ELECTIONS AND CITIZEN ENGAGEMENT AND COGNITION - MODULE 3: "MEANINGFUL CHOICES" - CONDITIONS WHERE ELECTIONS PROVIDE A MEANINGFUL CHOICE SET TO VOTERS (2006-2011) - MODULE 4: DISTRIBUTIONAL POLITICS, SOCIAL PROTECTION, AND POLITICAL MOBILIZATION (2011-2016) An international committee of leading scholars of electoral politics and political science develop the research agenda, questionnaires, and study design. 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. The work of the CSES Secretariat is funded by the American National Science Foundation, the GESIS - Leibniz Institute for the Social Sciences, and the University of Michigan's Center for Political Studies. The project also receives in-kind support from participating election studies, additional organizations that sponsor planning committee meetings and conferences, and the many organizations that fund national election studies that participate in CSES. This product, the CSES Integrated Module Dataset (IMD), brings together all four completed modules into one unified and harmonized data product. More specific information about this data product is available in the section: "CSES INTEGRATED MODULE DATASET (IMD): PRODUCT OVERVIEW" below. --------------------------------------------------------------------------- >>> CSES MODULE 5 PLANNING COMMITTEE --------------------------------------------------------------------------- The CSES Module 5 Planning Committee was responsible for overseeing the development and early implementation of the CSES Integrated Module Dataset (IMD). Besides the Chair, Planning Committee Members are listed alphabetically by surname. The following persons were members of the CSES Module 5 Planning Committee: JOHN ALDRICH (Module 5 Planning Committee Chair) Duke University, United States EVA ANDUIZA Universitat Autonoma de Barcelona, Spain ALI CARKOGLU Koc University, Turkey GORAN CULAR University of Zagreb, Croatia RACHEL GIBSON University of Manchester, United Kingdom ELISABETH GIDENGIL McGill University, Canada SARA B. HOBOLT London School of Economics and Political Science, United Kingdom DAVID A. HOWELL (CSES Director of Studies, Ex-officio member) University of Michigan, United States CHI HUANG National Chengchi University, Taiwan AIDA JUST Bilkent University, Turkey ORIT KEDAR The Hebrew University of Jerusalem, Israel GEORG LUTZ University of Lausanne, Switzerland PEDRO MAGALHAES University of Lisbon, Portugal RACHEL MENEGUELLO Universidade Estadual de Campinas, Brazil HENRIK OSCARSSON University of Gothenburg, Sweden STEPHEN QUINLAN (CSES Project Manager, Ex-officio member) GESIS - Leibniz Institute for the Social Sciences Mannheim, Germany NICOLAS SAUGER Sciences Po, France RUDIGER SCHMITT-BECK University of Mannheim, Germany CARLOS SHENGA Higher Institute of Public Administration, Mozambique ALBERTO SIMPSER Instituto Tecnologico Autonomo de Mexico (ITAM), Mexico WOUTER VAN DER BRUG University of Amsterdam, The Netherlands MARKUS WAGNER University of Vienna, Austria MASAHIRO YAMADA Kwansei Gakuin University, Japan ELIZABETH ZECHMEISTER Vanderbilt University, United States --------------------------------------------------------------------------- >>> CSES MODULE 6 PLANNING COMMITTEE --------------------------------------------------------------------------- The CSES Module 6 Planning Committee is responsible for the design and implementation of Module 6 but oversaw the implementation of the CSES Integrated Module Dataset (IMD). Besides the Chair, Planning Committee Members are listed alphabetically by surname. The following persons were members of the CSES Module 6 Planning Committee: ELIZABETH ZECHMEISTER (Module 6 Planning Committee Chair) Vanderbilt University, United States EVA ANDUIZA Universitat Autonoma de Barcelona, Spain ALI CARKOGLU Koc University, Turkey MARINA COSTA-LOBO Institute of Social Sciences, University of Lisbon CATHERINE DE VRIES Vrije Universiteit Amsterdam, Netherlands DAVID A. HOWELL (CSES Director of Studies, Ex-officio member) University of Michigan, United States AIDA JUST Bilkent University, Turkey ORIT KEDAR The Hebrew University of Jerusalem, Israel GEORG LUTZ University of Lausanne, Switzerland IRFAN NOORUDIN, Georgetown University, United States HENRIK OSCARSSON University of Gothenburg, Sweden STEPHEN QUINLAN (CSES Project Manager, Ex-officio member) GESIS - Leibniz Institute for the Social Sciences Mannheim, Germany RUDIGER SCHMITT-BECK University of Mannheim, Germany JILL SHEPPARD Australian National University, Canberra, Australia CARLOS SHENGA Higher Institute of Public Administration, Mozambique ALBERTO SIMPSER Instituto Tecnologico Autonomo de Mexico (ITAM), Mexico LAURA STEPHENSON Western University of Ontario, Canada DAVID SULMONT Pontifical University Catholic of Peru, Peru JOSH TUCKER New York University, United States WOUTER VAN DER BRUG University of Amsterdam, The Netherlands MARKUS WAGNER University of Vienna, Austria MASAHIRO YAMADA Kwansei Gakuin University, Japan ERIC YU National Chengchi University, Taiwan --------------------------------------------------------------------------- >>> CSES SECRETARIAT --------------------------------------------------------------------------- The CSES Secretariat comprises the central staffing and operations for the CSES project, under the leadership of the chair of the CSES Planning Committees (listed above). Since June 2011, the Secretariat has been a collaboration between the GESIS Leibniz Institute for the Social Sciences, Germany, and the University of Michigan's Center for Political Studies in the United States. Professor John Aldrich, Chair of the CSES Module 5 Planning Committee and Professor Elizabeth Zechmeister, Chair of the CSES Module 6 Planning Committee have overseen the operations of the CSES Secretariat during the development of the CSES Integrated Module Dataset (IMD). The CSES Secretariat for the Integrated Module Dataset (IMD) was comprised of the following persons: David Howell served as the Director of Studies, and Dr. Stephen Quinlan served as the Project Manager. Katharina Blinzler, Klara Dentler, Christian Schimpf, Slaven Zivkovic, and Dr. Bojan Todosijevic were responsible for research support, documentation, data preparation, data checking, and cleaning. Support was received from various sources for the activities of the CSES Secretariat during the development of the CSES Integrated Module Dataset: (1) American National Science Foundation (NSF) grant SES-1420973, "The Fifth Module of the Comparative Study of Electoral Systems (CSES)" with Principal Investigators Nancy Burns (University of Michigan), Andre Blais (University of Montreal), and John Aldrich (Duke University) supported CSES Secretariat activities at the University of Michigan beginning in 2014. (2) American National Science Foundation (NSF) grant SES-1760058, "The Sixth Module of the Comparative Study of Electoral Systems (CSES)" with Principal Investigators Ken Kollman (University of Michigan) and John Aldrich (Duke University) supported CSES Secretariat activities at the University of Michigan beginning in 2018. (3) The CSES Secretariat activities at the GESIS Leibniz Institute for the Social Sciences, Germany are funded by GESIS. (4) The Center for Political Studies at the University of Michigan provides additional financial support. =========================================================================== ))) CSES INTEGRATED MODULE DATASET (IMD): HOW TO ACCESS? =========================================================================== --------------------------------------------------------------------------- >>> THE CSES IMD CODEBOOK --------------------------------------------------------------------------- Users are advised to first download the CSES IMD Codebook file: csesIMD_codebook.zip Contains the four CSES IMD Codebook files, including this one, in text format. The CSES IMD Codebook can also be navigated online at the CSES Integrated Module Dataset study page. --------------------------------------------------------------------------- >>> THE CSES IMD DATA FILES --------------------------------------------------------------------------- The following ZIP files, which contain the CSES data are available to download from the CSES IMD study page at: https://cses.org/data-download/cses-integrated-module-dataset-imd/ Users can download the data in a variety of formats depending on which statistical packages they intend to use with the data: cses_imd_csv.zip Contains a CSV file with variables names as column headers but no additional metadata (for instance, no code labels are included). cses_imd_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: csesimd.sas for SAS, csesimd.sps for SPSS, and csesimd.do for STATA. cses_imd_r.zip Contains an R Workplace system file (.rdata), with the dataset already prepared and ready to be loaded into R. Missing data statements are not applied. cses_imd_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. cses_imd_spss.zip Contains an SPSS system file (.sav), with the dataset already prepared and ready to be loaded into SPSS. Missing data statements are not applied. cses_imd_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. Please note that all of the above packages will need a File Extractor program downloaded to their computer to be able to Unzip and open the above files. We recommend that PC users create the following directory on their hard drive, and to download their files from this IMD Phase 2 release to that location: "c:/cses/imdphase2/20191017/" The subdirectory value "20191017" represents the version (release date) of the dataset - this being 2019, and the October 17 version of CSES IMD. This file structure is compatible with how the "csesimd_syntax.zip" file (detailed above) is organized. The 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. =========================================================================== ))) CSES INTEGRATED MODULE DATASET (IMD): PRODUCT OVERVIEW =========================================================================== --------------------------------------------------------------------------- >>> CSES INTEGRATED MODULE DATASET (IMD) STUDY DESCRIPTION - ABSTRACT --------------------------------------------------------------------------- The CSES Integrated Module Dataset (IMD) brings together the existing Standalone CSES Modules (CSES Modules 1-4 inclusive, all cross-section cases and variables for select questions) into one longitudinal and harmonized dataset. This data product has been created by the CSES Secretariat. The dataset includes over 281,000 individual-level observations, 174 elections across 55-polities and allows exploration of electoral behavior across 20-years, three decades, and two centuries. All data included in the CSES Integrated Module Dataset (IMD) was collected between 1996 and 2016 and is available in the Standalone CSES Modules. Variables that appear in at least three Standalone CSES Modules, up to and including CSES Module 5, are eligible for inclusion in IMD with all polities ever having run a CSES Module also included in the dataset. All variables in CSES IMD are harmonized, allowing for longitudinal analysis. The CSES IMD offers several unique innovations to CSES users including: - Harmonization of party codes within countries across modules - Harmonization of administrative variables - Pre-coded incumbent vote choice variable - In-data labels for parties and candidates - Extensive checking of all CSES data More information regarding the CSES IMD can be found in the Comparative Study of Electoral Systems (CSES) Integrated Module Dataset (IMD) Stimulus Paper available on the CSES website. --------------------------------------------------------------------------- >>> OVERVIEW OF CSES IMD DATA FILE PARTICULARS --------------------------------------------------------------------------- Type of study: CROSS-SECTIONAL Kind of data: SURVEY DATA FUSED WITH CONTEXTUAL MACRO DATA Primary Unit of Analysis: INDIVIDUALS Universe: ALL PERSONS OF ELIGIBLE VOTING AGE AND ELIGIBLE TO VOTE IN THE NATIONAL ELECTION Geographic Coverage: GLOBAL (Europe, North America, parts of Africa, Asia, and South America, Australia & New Zealand) File Structure: RECTANGULAR Total Case Count: 281,083 Total Election Studies: 174 Total Variable Count: 276 Total Polities: 55 Time Period: 1996-2016 Modules: 4 (Modules 1-4 inclusive) --------------------------------------------------------------------------- >>> LIST OF ELECTION STUDIES AND POLITIES INCLUDED IN CSES IMD --------------------------------------------------------------------------- Phase 2 of the CSES Integrated Module Dataset (IMD) contains data from Standalone CSES Modules 1-4. CSES IMD contains data from 174 election studies in 55 polities fielded between 1996 and 2016. They are listed below in alphabetic order with an overview of some particulars of each election study. | +++ TABLE: OVERVIEW OF ELECTION STUDIES INCLUDED IN CSES IMD WITH | NUMBER OF OBSERVATIONS, MODE OF DATA COLLECTION, AND | FIELDWORK DATES | | POLITY (ELEC YEAR) No of Mode of Dates of Fieldwork | Observations Interview (Start-End date) | --------------------------------------------------------------------- | ALBANIA (2005) 1,116 F2F Jul 5, 2005-Jul 20, 2005 | ARGENTINA (2015) 1,406 F2F Nov 21, 2015-Dec 30, 2015 | AUSTRALIA (1996) 1,798 MB Mar 2, 1996-Jun 24, 1996 | AUSTRALIA (2004) 1,769 MB Oct 10, 2005-Feb 15, 2005 | AUSTRALIA (2007) 1,873 MB Nov 23, 2007-Mar 14, 2008 | AUSTRALIA (2013) 3,953 MX Sep 6, 2013-Jan 6, 2014 | AUSTRIA (2008) 1,165 F2F May 5, 2009-Jul 2, 2009 | AUSTRIA (2013) 1,000 TP Oct 1, 2013-Oct 29, 2013 | BELGIUM - FLAN (1999) 2,179 TP Sep 1999-Jan 2000 | BELGIUM - WALL (1999) 1,960 F2F Sep 1999-Jan 2000 | BELGIUM (2003) 2,225 F2F Dec 15, 2003-Apr 31, 2004 | BELARUS (2001) 1,000 F2F Sep 24, 2001-Oct 6, 2001 | BELARUS (2008) 1,000 F2F Feb 2, 2009-Feb 24, 2009 | BRAZIL (2002) 2,514 F2F Oct 31, 2002-Dec 28, 2002 | BRAZIL (2006) 1,000 F2F Dec 17, 2006-Dec 27, 2006 | BRAZIL (2010) 2,000 F2F Nov 4, 2010-Nov 20, 2010 | BRAZIL (2014) 3,136 F2F Nov 1, 2014-Nov 19, 2014 | BULGARIA (2001) 1,482 F2F Jul 5, 2001-Jul 15, 2001 | BULGARIA (2014) 999 F2F Jan 23, 2015-Jan 31, 2015 | CANADA (1997) 1,851 TP Multi-wave study*** | CANADA (2004) 1,674 TP Multi-wave study*** | CANADA (2008) 4,495 TP Multi-wave study*** | CANADA (2011) 3,458 MX Multi-wave study*** | CANADA (2015) 4,202 MX Multi-wave study*** | CHILE (1999) 1,173 TP Jan 4, 2000-Jan 18, 2000 | CHILE (2005) 1,200 TP Jan 1, 2006-Jan 8, 2006 | CHILE (2009) 1,200 TP Jan 1, 2010-Jan 9, 2010 | CROATIA (2007) 1,004 F2F Mar 10, 2008-Apr 3, 2008 | CZECH REPUBLIC (1996) 1,229 F2F Jun 10, 1996-Jun 18, 1996 | CZECH REPUBLIC (2002) 948 F2F Jun 24, 2002-Jul 1, 2002 | CZECH REPUBLIC (2006) 2,002 F2F Jun 9, 2006-Jun 21, 2006 | CZECH REPUBLIC (2010) 1,857 F2F May 31, 2010-Jun 18, 2010 | CZECH REPUBLIC (2013) 1,653 F2F Oct 28, 2013-Nov 14, 2013 | DENMARK (1998) 2,001 F2F Apr 6, 1998-Jun 16, 1998 | DENMARK (2001) 2,026 F2F Dec 7, 2001-Mar 14, 2002 | DENMARK (2007) 1,442 F2F Nov 21, 2007-Jun 12, 2008 | ESTONIA (2011) 1,000 F2F Mar 7, 2011-Mar 25, 2011 | FINLAND (2003) 1,196 F2F Mar 17, 2003-Apr 30, 2003 | FINLAND (2007) 1,283 F2F Mar 24, 2007-May 20, 2007 | FINLAND (2011) 1,298 F2F Apr 18, 2011-May 25, 2011 | FINLAND (2015) 1,587 MX Apr 24, 2015-Jul 7, 2015 | FRANCE (2002) 1,000 TP May 23, 2003-May 24, 2003 | FRANCE (2007) 2,000 TP Jun 18, 2007-Jul 7, 2007 | FRANCE (2012) 2,014 F2F May 9, 2012-Jun 9, 2015 | GERMANY (1998) 2,019 TP Sep 28, 1998-Oct 17, 1998 | GERMANY (2002) 2,000 TP Oct 31, 2002-Nov 12, 2002 | GERMANY (2002) 1,023 MB Sep 23, 2002-Oct 31, 2002 | GERMANY (2005) 2,018 TP Sep 21, 2005-Oct 5, 2005 | GERMANY (2009) 2,095 TP Sep 28, 2009-Nov 23, 2009 | GERMANY (2013) 1,889 F2F Sep 23, 2013-Dec 23, 2013 | GREAT BRITAIN (1997) 2,897 MB May 1, 1997-Jul 29, 1997 | GREAT BRITAIN (2005) 860 F2F Jun 20, 2005-Nov 24, 2005 | GREAT BRITAIN (2015) 1,567 F2F May 9, 2015-Sep 13, 2015 | GREECE (2009) 1,022 TP Dec 10, 2009-Dec 18, 2009 | GREECE (2012) 1,029 MX Oct 19, 2012-Jan 5, 2013 | GREECE (2015) 1,008 MX Jun 12, 2015-Sep 8, 2015 | HONG KONG (1998) 1,000 F2F Jun 1, 1998-Aug 31, 1998 | HONG KONG (2000) 674 F2F Sep 17, 2000-Jan 31, 2001 | HONG KONG (2004) 582 F2F Oct 5, 2004-Dec 19, 2004 | HONG KONG (2008) 815 TP Nov 3, 2008-Nov 30, 2008 | HONG KONG (2012) 1,044 TP Sep 13, 2012-Sep 22, 2013 | HUNGARY (1998) 1,525 F2F May 14, 1998-May 21, 1998 | HUNGARY (2002) 1,200 F2F Apr 11, 2002-Apr 19, 2002 | ICELAND (1999) 1,631 TP May 24, 1999-Jun 23, 1999 | ICELAND (2003) 1,446 TP May 15, 2003-Jun 29, 2003 | ICELAND (2007) 1,595 TP Jun 2, 2007-Aug 22, 2007 | ICELAND (2009) 1,385 TP May 7, 2009-Aug 31, 2009 | ICELAND (2013) 1,479 TP May 4, 2013-Sep 30, 2013 | IRELAND (2002) 2,367 F2F May 22, 2002-Nov 29, 2002 | IRELAND (2007) 1,435 MX May 31, 2007-Nov 2, 2007 | IRELAND (2011) 1,853 F2F Mar 6, 2011-Apr 10, 2011 | ISRAEL (1996) 1,091 TP Feb 2, 2003-Feb 7, 2003 | ISRAEL (2003) 1,212 TP Feb 2, 2003-Feb 7, 2003 | ISRAEL (2006) 1,200 TP Jul 13, 1996-Aug 7, 1996 | ISRAEL (2013) 1,017 TP Feb 8, 2013-Mar 13, 2013 | ITALY (2006) 1,439 F2F May 10, 2006-May 16, 2006 | JAPAN (1996) 1,327 F2F Oct 22, 1996-Nov 3, 1996 | JAPAN (2004) 1,977 F2F Jul 15, 2004-Jul 26, 2004 | JAPAN (2007) 1,373 F2F Jul 31, 2007-Aug 13, 2007 | JAPAN (2013) 1,937 F2F Jul 22, 2013-Aug 25, 2013 | KENYA (2013) 1,200 F2F Oct 21, 2013-Nov 28, 2013 | KYRGZSTAN (2005) 2,000 F2F Feb 1, 2006-Mar 7, 2006 | LATVIA (2010) 1,005 F2F Oct 16, 2010-Nov 7, 2010 | LATVIA (2011) 1,004 F2F Oct 16, 2011-Nov 11, 2011 | LATVIA (2014) 1,036 F2F Nov 7, 2014-Nov 20, 2014 | LITHUANIA (1997) 1,009 F2F Jan 13, 1998-Jan 20, 1998 | MEXICO (1997) 2,033 F2F Jul 14, 1997-Jul 29, 1997 | MEXICO (2000) 1,766 F2F Jul 15, 2000-Jul 20, 2000 | MEXICO (2003) 1,991 F2F Jul 19, 2003-Jul 26, 2003 | MEXICO (2006) 1,591 F2F Jul 23, 2006-Aug 2, 2006 | MEXICO (2009) 2,400 F2F Jul 12, 2009-Jul 23, 2009 | MEXICO (2012) 2,400 F2F Jul 13, 2012-Jul 19, 2012 | MEXICO (2015) 1,197 F2F Jun 20, 2015-Jun 28, 2015 | MONTENEGRO (2012) 967 F2F Feb 2013-Mar 2013 | NETHERLANDS (1998) 2,101 F2F May 11, 1998-Jul 4, 1998 | NETHERLANDS (2002) 1,574 F2F May 16, 2002-Jun 27, 2002 | NETHERLANDS (2006) 2,359 F2F Nov 23, 2006-Jan 4, 2007 | NETHERLANDS (2010) 2,153 F2F Jun 10, 2010-Jul 22, 2010 | NEW ZEALAND (1996) 4,080 MB Oct 13, 1996-Dec 24, 1996 | NEW ZEALAND (2002) 1,741 MB Specific dates unavailable | NEW ZEALAND (2008) 1,149 MB Nov 10, 2008-Feb 27, 2008 | NEW ZEALAND (2011) 1,374 MX Nov 30, 2011-Apr 4, 2012 | NEW ZEALAND (2014) 1,224 MX Sep 23, 2014-Feb 6, 2015 | NORWAY (1997) 2,055 MX Sep 16, 1997-Nov 30, 1997 | NORWAY (2001) 2,052 MX Sep 11, 2001-Jan 31, 2002 | NORWAY (2005) 2,012 MX Sep 13, 2005-Dec 20, 2005 | NORWAY (2009) 1,782 MX Sep 15, 2009-Feb 4, 2010 | NORWAY (2013) 1,727 MX Sep 12, 2013-Jan 6, 2014 | PERU (2000) 1,102 F2F May 5, 2000-May 8, 2000 | PERU (2001) 1,118 F2F Specific dates unavailable | PERU (2006) 2,032 F2F May 18, 2006-May 21, 2006 | PERU (2011) 1,570 F2F May 7, 2011-May 8, 2011 | PERU (2016) 1,572 F2F May 7, 2016-May 17, 2016 | PHILIPPINES (2004) 1,200 F2F Aug 5, 2004-Aug 22, 2004 | PHILIPPINES (2010) 1,200 F2F Jun 25, 2010-Jun 28, 2010 | PHILIPPINES (2016) 1,200 F2F Jun 24, 2016-Jun 27, 2016 | POLAND (1997) 2,003 F2F Sep 29, 1997-Oct 6, 1997 | POLAND (2001) 1,794 F2F Sep 29, 2001-Oct 11, 2001 | POLAND (2005) 2,402 F2F Sep 27, 2005-Oct 8, 2005 | POLAND (2007) 1,817 F2F Nov 8, 2007-Dec 1, 2007 | POLAND (2011) 1,919 F2F Oct 20, 2011-Nov 13, 2011 | PORTUGAL (2002) 1,303 F2F Mar 23, 2002-Apr 8, 2002 | PORTUGAL (2005) 2,801 F2F Mar 5, 2005-May 8, 2005 | PORTUGAL (2009) 1,316 F2F Oct 2, 2009-Feb 8, 2010 | PORTUGAL (2015) 1,499 F2F Oct 7, 2015-Dec 9, 2015 | ROMANIA (1996) 1,175 F2F Nov 26, 1996-Dec 3, 1996 | ROMANIA (2004) 1,913 F2F Dec 14, 2004-Jan 7, 2005 | ROMANIA (2009) 1,403 F2F Dec 7, 2009-Dec 20, 2009 | ROMANIA (2012) 2,283 F2F Dec 15, 2012-Jan 30, 2013 | ROMANIA (2014) 1,112 F2F Nov 20, 2014-Dec 8, 2014 | RUSSIA (1999) 1,842 F2F Dec 25, 1999-Jan 31, 2000 | RUSSIA (2000) 1,748 F2F Dec 25, 1999-Jan 31, 2000 | RUSSIA (2004) 1,496 F2F Apr 4, 2004-May 11, 2004 | SERBIA (2012) 1,568 F2F Dec 6, 2012-Feb 13, 2013 | SLOVAKIA (2010) 1,203 F2F Jun 18, 2010-Jul 20, 2010 | SLOVAKIA (2016) 1,150 F2F Oct 13, 2016-Nov 28, 2016 | SLOVENIA (1996) 2,031 F2F Oct 4, 1997-Nov 7, 1997 | SLOVENIA (2004) 1,002 F2F Mar 17, 2005-Apr 25, 2005 | SLOVENIA (2008) 1,055 F2F Mar 24, 2009-Jun 4, 2009 | SLOVENIA (2011) 1,031 F2F Mar 29, 2012-May 28, 2012 | SOUTH AFRICA (2009) 1,200 F2F Oct 2, 2009-Oct 31, 2009 | SOUTH AFRICA (2014) 1,300 F2F Feb 2, 2015-Feb 26, 2015 | SOUTH KOREA (2000) 1,100 F2F May 12, 2000-May 23, 2000 | SOUTH KOREA (2004) 1,500 F2F Apr 15, 2004-Apr 20, 2004 | SOUTH KOREA (2008) 1,000 F2F Apr 10, 2008-Apr 21, 2008 | SOUTH KOREA (2012) 1,000 F2F Apr 12, 2012-Apr 29, 2012 | SPAIN (1996) 1,212 F2F Mar 11, 1996-Mar 16, 1996 | SPAIN (2000) 1,208 F2F Mar 13, 2000-Mar 20, 2000 | SPAIN (2004) 1,212 F2F Mar 15, 2004-Mar 21, 2004 | SPAIN (2008) 1,204 F2F Apr 11, 2008-Apr 21, 2008 | SWEDEN (1998) 1,157 MX Sep 11, 1998-Nov 21, 1998 | SWEDEN (2002) 1,060 F2F Sep 16, 2002-Oct 31, 2002 | SWEDEN (2006) 1,547 F2F Sep 18, 2006-Oct 31, 2006 | SWEDEN (2014) 832 F2F Sep 15, 2014-Nov 17, 2014 | SWITZERLAND (1999) 2,048 TP Oct 25, 1999-Nov 17, 1999 | SWITZERLAND (2003) 1,418 TP Oct 20, 2003-Nov 2, 2003 | SWITZERLAND (2007) 3,164 TP Oct 22, 2007-Dec 31, 2007 | SWITZERLAND (2011) 4,391 MX Oct 24, 2011-Dec 12, 2011 | TAIWAN (1996) 1,200 F2F Jun 29, 1996-Sep 11, 1996 | TAIWAN (2001) 2,022 F2F Jan 20, 2002-Apr 15, 2002 | TAIWAN (2004) 1,823 F2F Jun 20, 2004-Sep 29, 2004 | TAIWAN (2008) 1,905 F2F Jun 28, 2008-Aug 19, 2008 | TAIWAN (2012) 1,826 F2F Jan 15, 2012-Mar 6, 2012 | THAILAND (2001) 1,081 F2F Specific dates unavailable | THAILAND (2007) 1,990 F2F Jan 9, 2008-Jan 31, 2008 | THAILAND (2011) 1,500 F2F Jul 15, 2011-Aug 11, 2011 | TURKEY (2011) 1,109 F2F Jul 21, 2011-Aug 26, 2011 | TURKEY (2015) 1,086 F2F Jul 18, 2015-Sep 10, 2015 | UKRAINE (1998) 1,148 F2F Specific dates unavailable | UNITED STATES (1996) 1,534 MX Nov 6, 1996-Jan 2, 1997 | UNITED STATES (2004) 1,066 F2F Nov 3, 2004-Dec 20, 2004 | UNITED STATES (2008) 2,102 F2F Nov 5, 2008-Dec 21, 2008 | UNITED STATES (2012) 1,929 F2F Nov 7, 2012-Jan 12, 2013 | URUGUAY (2009) 968 F2F Aug 13, 2010-Aug 31, 2010 | -------------------------------------------------------------------- | TOTAL 281,083 | | Key: F2F=Face to Face; TP=Telephone, MB=Mailback; MX=Mixed. | | *** = Study run across multiple waves. Please see variables in | Standalone CSES Module Codebooks for more comprehensive details | for dates of fieldwork. For election studies that are embedded in multi-wave panel studies, only those respondents who participated in the CSES component of the study are included in the CSES Integrated Module Dataset (IMD). Phase 2 of the CSES Integrated Module Dataset (IMD) contains data from 55 polities. They are listed below in descending order with an overview of how many times Standalone CSES Modules were run in each state and the year in which they were fielded. | +++ TABLE: OVERVIEW OF POLITIES INCLUDED IN CSES IMD AND NUMBER | OF ELECTION STUDIES AND YEAR OF ELECTION | | POLITY No of Election Years | Election Studies | --------------------------------------------------------------------- | MEXICO 7 1997; 2000; 2003; 2006; 2009; 2012; 2015; | GERMANY 6* 1996; 2002; 2005; 2009; 2013; | CANADA 5 1997; 2004; 2008; 2011; 2015; | CZECH REPUBLIC 5 1996; 2002; 2006; 2010; 2013; | HONG KONG 5 1998; 2000; 2004; 2008; 2012; | ICELAND 5 1999; 2003; 2007; 2009; 2013; | NEW ZEALAND 5 1996; 2002; 2008; 2011; 2014; | NORWAY 5 1997; 2001; 2005; 2009; 2013; | PERU 5 2000; 2001; 2006; 2011; 2016; | POLAND 5 1997; 2001; 2005; 2007; 2011; | PORTUGAL 5** 2002; 2005; 2009; 2015: | ROMANIA 5 1996; 2004; 2009; 2012; 2014; | TAIWAN 5 1996; 2001; 2004; 2008; 2012; | AUSTRALIA 4 1996; 2004; 2007; 2013; | BRAZIL 4 2002; 2006; 2010; 2014; | FINLAND 4 2003; 2007; 2011; 2015; | ISRAEL 4 1996; 2003; 2006; 2013; | JAPAN 4 1996; 2004; 2007; 2013; | NETHERLANDS 4 1998; 2002; 2006; 2010; | SLOVENIA 4 1996; 2004; 2008; 2011; | SOUTH KOREA 4 2000; 2004; 2008; 2012; | SPAIN 4 1996; 2000; 2004; 2008; | SWEDEN 4 1998; 2002; 2006; 2014; | SWITZERLAND 4 1999; 2003; 2007; 2011; | U STATES OF AMERICA 4 1996; 2004; 2008; 2012; | CHILE 3 1999; 2005; 2009; | DENMARK 3 1998; 2001; 2007; | FRANCE 3 2002; 2007; 2012; | GREAT BRITAIN 3 1997; 2005; 2015; | GREECE 3 2009; 2012; 2015; | IRELAND 3 2002; 2007; 2011; | LATVIA 3 2010; 2011; 2014; | PHILIPPINES 3 2004; 2010; 2016; | RUSSIA 3 1999; 2000; 2004; | THAILAND 3 2001; 2007; 2013; | AUSTRIA 2 2008; 2013; | BELARUS 2 2001; 2008; | BELGIUM 2*** 1999; 2003; | BULGARIA 2 2001; 2014; | HUNGARY 2 1998; 2002; | SLOVAKIA 2 2010; 2016; | SOUTH AFRICA 2 2009; 2014; | TURKEY 2 2011; 2015; | ALBANIA 1 2005 | ARGENTINA 1 2015 | CROATIA 1 2007 | ESTONIA 1 2011 | ITALY 1 2006 | KENYA 1 2013 | KYRGYZSTAN 1 2005 | LITHUANIA 1 1997 | MONTENEGRO 1 2012 | SERBIA 1 2012 | UKRAINE 1 1998 | URUGUAY 1 2009 | -------------------------------------------------------------------- | TOTAL 174 | | * = In 2002, two separate studies were run in Germany. | ** = In 2002, Portugal ran CSES Module 1 and 2 simultaneously. | *** = In 1999, two separate studies were run in Belgium, one in | the Flanders region and the other in the Wallonia region. Phase 2 of the CSES Integrated Module Dataset (IMD) contains data from polities across four completed Standalone CSES Modules. Below, the polities included are listed by their participation in Standalone CSES Modules. Polities are listed in alphabetical order. | +++ TABLE: OVERVIEW OF POLITIES INCLUDED IN CSES IMD AND STANDALONE | CSES MODULES THEY APPEAR IN | | POLITY M1 M2 M3 M4 | --------------------------------------------------------------------- | ALBANIA - X - - | ARGENTINA - - - X | AUSTRALIA X X X X | AUSTRIA - - X X | BELARUS X - X - | BELGIUM X X - - | BRAZIL - X X X | BULGARIA - X - X | CANADA X X X X | CHILE X X X - | CROATIA - - X - | CZECH REPUBLIC X X X X | DENMARK X X X - | ESTONIA - - X - | FINLAND - X X X | FRANCE - X X X | GERMANY X X X X | GREAT BRITAIN X X - X | GREECE - - X X | HONG KONG X X X X | HUNGARY X X - - | ICELAND X X X X | IRELAND - X X X | ISRAEL X X X X | ITALY - X - - | JAPAN X X X X | KENYA - - - X | KYRGYZSTAN - X - - | LATVIA - - X X | LITHUANIA X - - - | MEXICO X X X X | MONTENEGRO - - - X | NETHERLANDS X X X - | NEW ZEALAND X X X X | NORWAY X X X X | PERU X X X X | PHILIPPINES - X X X | POLAND X X X X | PORTUGAL X X X X | ROMANIA X X X X | RUSSIA X X - - | SERBIA - - - X | SLOVAKIA - - X X | SLOVENIA X X X X | SOUTH AFRICA - - X X | SOUTH KOREA X X X X | SPAIN X X X - | SWEDEN X X X X | SWITZERLAND X X X X | TAIWAN X X X X | THAILAND X - X X | TURKEY - - X X | UKRAINE X - - - | U STATES OF AMERICA X X X X | URUGUAY - - X - | -------------------------------------------------------------------- | | KEY: X=Participated in CSES Module; - =Did not participate in CSES | Module. --------------------------------------------------------------------------- >>> THE ROAD TO CSES IMD --------------------------------------------------------------------------- At the CSES Planning Committee (PC) Meeting in Seattle in October 2015, the PC mandated the CSES Secretariat to devise a Strategic Plan for the creation of a unified CSES data product including harmonization of party code variables. In November 2017, upon delivery by the Secretariat of the Strategic Plan to the Module 5 PC Meeting in Mannheim, the PC mandated the Secretariat to create an integrated harmonized data product encompassing CSES data from all modules to be known as the CSES Integrative Module Dataset (IMD). Previously, Giebler, Lichteblau, May, Melcher, Wagner & Wessels (2016) from the WZB made a user contribution to the CSES Website by creating the first publicly available cumulative CSES dataset which brought together CSES Modules 1-3 in one dataset and connected several variables of interest to scholars over time in a unified way for the first time. =========================================================================== ))) CSES INTEGRATED MODULE DATASET (IMD) DOCUMENTATION: WHAT'S AVAILABLE AND HOW TO USE? =========================================================================== There are several components to the CSES IMD documentation. We detail each in turn below: --------------------------------------------------------------------------- >>> CSES IMD CODEBOOK OVERVIEW --------------------------------------------------------------------------- The primary component of documentation for this data product is the CSES IMD Codebook. The CSES IMD Codebook consists of four components, namely: 1) PART 1: CSES IMD - INTRODUCTION (file name: csesIMD_codebook_part1_introduction.txt) Part 1 (This file) overview of the CSES study and data, information about how to use the files, election study descriptions, information on the CSES IMD datafile, the checks the CSES Secretariat conducts on the datafile and information on the CSES Planning Committees that oversaw the development of CSES IMD. 2) PART 2: CSES IMD - VARIABLES DESCRIPTION (file name: csesIMD_codebook_part2_variables.txt) Part 2 is the variable description file and includes the survey questions, code frames, general notes, election study notes, and details about sources for macro data. 3) PART 3: CSES IMD - PARTIES/COALITIONS - NUMERICAL CODES HARMONIZED BY POLITY (file name: csesIMD_codebook_part3_parties_coalitions_numerical_codes.txt) Part 3 details the harmonized party/coalition numerical codes for each polity across all modules included in the CSES IMD dataset. 4) PART 4: CSES IMD - PARTIES/COALITIONS & LEADERS - ALPHABETICAL (RELATIONAL) CODES BY POLITY BY MODULE (file name: csesIMD_codebook_part4_parties_coalitions_leaders_alphabetical_codes.txt) Part 4 details the party/coalition and leader alphabetical (relational) codes assigned for each polity in each module in the CSES IMD dataset. Users are advised that alphabetical codes for parties/coalitions and leaders can change between modules. The Standalone CSES Module Questionnaires are available from the CSES website - please consult the original study webpages. For all election studies included in CSES, collaborators have provided documentation to accompany their election studies. These documents, where available, can be found on the Standalone CSES Module study pages. Analysts will also want to become familiar with the CSES IMD errata page. It is accessible from the CSES IMD download page under "Data Center" on the CSES website. Information, updates, and error notifications and corrections are posted there, in real-time, as they become available. Please regularly check for errata notifications to keep up to date. --------------------------------------------------------------------------- >>> CSES IMD CODEBOOK CONVENTIONS --------------------------------------------------------------------------- The CSES project uses American English language and date standards (MM-DD-YYYY). In the CSES Integrated Module dataset, all variables begin with the letters "IMD" (IMD standing for the Integrated Module Dataset). Variables are presented in four groupings: 1) IMD1001-IMD1999 Identification, weight, and election study variables 2) IMD2001-IMD2999 Demographic variables 3) IMD3001-IMD3999 Micro-level (survey) data 4) IMD5001-IMD5999 Macro-level data In the Variable Descriptions section of the CSES IMD Codebook (Part 2), the headers for individual variables are surrounded by two lines of dashes. --------------------------------------------------------------------------- >>> CSES IMD CODEBOOK - VARIABLE NOTES, POLITY NOTES, & ELECTION STUDY NOTES --------------------------------------------------------------------------- <<>> VARIABLE NOTES Variable notes provide information on the rationale of a variable as well as source information for that variable. It also details the polities for which no data for that particular variable are available. VARIABLE NOTES are listed below the descriptive information for the said variable and can be navigated in the CSES IMD Codebook by searching for "VARIABLE NOTES" in Part 2 of the CSES IMD Codebook. <<>> POLITY NOTES AND ELECTION STUDY NOTES A unique dimension of the CSES are the inclusion of POLITY NOTES and ELECTION STUDY NOTES. They are notes which are attached to each variable included in the dataset and refer to case-specific information regarding a particular variable. This information may apply to a polity consistently in the dataset (i.e.: a POLITY NOTE) or one election study in the dataset (i.e.: an ELECTION STUDY NOTE). Their purpose is to provide users with more detailed information on the case or explain essential deviations specific to cases from CSES conventions. They can be navigated in the Codebook by searching for "POLITY NOTES" or "ELECTION STUDY NOTES" in Parts 2-4 of the CSES IMD Codebook. --------------------------------------------------------------------------- >>> CSES IMD - ADDITIONAL DOCUMENTATION --------------------------------------------------------------------------- All election studies included in CSES provide numerous source material. These documents include the following: - Macro reports - Design reports - Original questionnaires including the language of origin We describe each in turn below. <<>> MACRO REPORT Collaborators submit a macro report to the CSES Secretariat when depositing their national data. Its purpose is to provide a coherent link between national level specialists and data specific to the election and polity in question. It provides information on the election, the composition of cabinet before and after the election, expert assessments of the parties, information on electoral rules operating in the polity, as well as original sources for the country level data. It aids the CSES Secretariat in collating some of the macro level data for each polity included in the country. Where available, macro reports can be found on the Standalone CSES Module download pages. <<>> DESIGN REPORT Collaborators also submit a design report to the CSES Secretariat when depositing their national data. It provides all information on the implementation of each individual election study, including details regarding fieldwork dates, mode of interview, sampling procedures, sampling frame, response and refusal rates, information on translation procedures, and weights. Some of this data is included directly in the CSES data in variables IMD1001-IMD1999. Where available, design reports can be found on the Standalone CSES Module download pages. <<>> ORIGINAL QUESTIONNAIRES Where available, CSES provides the original language questionnaires from each polity's national election study. Further, CSES requests that all studies included provide the English language questionnaire used as the basis for translation of the CSES questionnaire into a polity's native tongue(s). Where available, the questionnaires can be found on the Standalone CSES Module download pages. --------------------------------------------------------------------------- >>> HOW TO NAVIGATE THE CSES IMD CODEBOOK --------------------------------------------------------------------------- CSES IMD Codebook is produced in .txt format to allow for easy accessibility and as such the Codebook can be read into a variety of programs. The CSES Codebook can be navigated quickly in the electronic files, with the following commands allowing for quick searching: ))) = Section Header >>> = Sub-section Header 1 <<>> = Sub-section Header 2 +++ = Tables VARIABLES NOTES = Notes for particular variables POLITY NOTES = Notes for a particular polity ELECTION STUDY NOTES = Notes for a particular election study [ELECTION STUDY NOTES] = Notes for a particular election study where data in CSES IMD deviates from published data in Standalone CSES Modules. =========================================================================== ))) CSES INTEGRATED MODULE DATASET (IMD): CODING CONVENTIONS AND ADDITIONAL INFORMATION FOR USERS ABOUT DATASET =========================================================================== --------------------------------------------------------------------------- >>> CSES CONVENTION OF PARTIES/COALITION CODING IN STANDALONE CSES MODULES --------------------------------------------------------------------------- The CSES coding convention for parties and coalitions in Standalone CSES Modules involves a dual coding scheme for these data, namely a numerical classification and an alphabetical classification. All parties/coalitions in the data received a numerical code (allocation of which was in part random, part deliberate depending on the parties/coalitions performance in the election). Meanwhile, the six most popular parties/coalitions received an alphabetical code (A-F), along with some supplementary parties/ coalitions (G-I), which are linked to relational data for that party/ coalition (e.g., party likability). Consequently, allocation of numerical codes for parties/coalitions are module-specific and thus within the Standalone CSES Modules are not consistent within countries across time. --------------------------------------------------------------------------- >>> CSES CODING OF PARTY/COALITIONS AND LEADERS IN CSES IMD - A NEW DEPARTURE --------------------------------------------------------------------------- <<>> CSES IMD NUMERICAL PARTY/COALITION CODING - HARMONIZATION CSES IMD marks a radical new departure in the CSES approach to the coding of parties and coalitions. All parties/coalitions or presidential candidates, where applicable, participating in the election or the previous election receive a NUMERICAL code. These codes are used to identify the following in the micro-component of the CSES dataset: - who the respondent voted for in the current election (IMD3002). - who the respondent voted for in the previous election (IMD3004). - the respondent's party identification (IMD3005_3). These codes are used to identify the following in the macro-component of the CSES dataset: - Party/coalition of the prime minister before and after the election (IMD5008_) - Party/coalition of the president before and after the election (IMD5009_) In CSES IMD, each party/coalition receives a unique numerical identifier that is consistent across modules. This seven-digit numerical identifier contains information on the polity and a unique numerical value to distinguish the party/coalition. Hence, numerical party/coalition codes are harmonized across Modules within CSES IMD. The first three digits of the identifier consist of the three-digit UN Polity Identifier Code. The remaining four digits consist of numerical codes ranging from 0001 to 9999, with each party/coalition assigned a value that remains consistent across Modules. In assigning of the last four digits, macro data specialists have assigned codes with consistent leading vote-getters in a polity being assigned lower values. Beyond this, the allocation of the final four digits is random. The harmonized and consistent codes for parties/coalitions are detailed in Part 3 of the CSES IMD Codebook. Users can search for the following term: "CSES IMD HARMONIZED PARTY/COALITION NUMERICAL CODES" <<>> CSES IMD NUMERICAL PARTY/COALITION CODING - IMPORTANT NOTES In cases of a merger between two parties, the newly created party/coalition receives a different numerical code from its previous incarnations. When parties compete in electoral alliances/coalitions, the alliance/coalition receives a different numerical code from the constituent parties that make it up. Parties/coalitions that merely undergo a name change do not receive a new unique code. Instead, a POLITY/ELECTION STUDY NOTE will note the name change. <<>> CSES IMD ALPHABETICAL PARTY/COALITION CODING - RELATIONAL DATA CSES IMD includes relational data for parties/coalitions. In CSES IMD, as in Standalone CSES Modules, this data is coded using an ALPHABETICAL classification with parties/coalitions where data is available receiving an alphabetical code (A-I). Parties/coalition A through F are the six most popular parties/coalitions, ordered in descending order of their share of the popular vote in the said polity's election (unless otherwise stated). Thus Party A is the party/coalition that received the most votes in the election, party B the second most votes, etc... Parties/coalitions who achieve at least 1% of the vote nationally are eligible for an alphabetical A-F assignment. In countries with multiple electoral tiers and where one vote is cast, parties are ordered according to their vote share in tier 1 (the lowest tier), unless otherwise stated. In countries where voters have two votes (i.e., a constituency and a list vote) simultaneously, for example Germany, parties are ordered by the national share of the party list vote (tier 2). Parties G, H, and I are supplemental parties. They may, but do not have to, accord with how parties A-F are ordered, that is ordered on the popular share of the vote in a country. More often, they are codified in no particular order. These parties are voluntarily provided by each polity's election study and often reflect important or notable parties within a polity. They may also include data about individual parties within a coalition, where data about the coalition and the individual parties, or some of these parties that make it up, are provided. These codes are used to identify the following in the micro-component of the CSES dataset: - Respondent's left-right placement of the party/coalition (variable IMD3007). - Respondent's likability of the party/coalition (variable IMD3008). These alphabetical codes are used to identify the distinct and macro level information about these said parties/coalitions, namely: - Election Results: percentage of vote for each party/coalition in lower house (variable IMD5001_) - Election Results: percentage of seats for each party/coalition in lower house (variable IMD5002_) - Election Results: percentage of vote for each party/coalition in upper house (variable IMD5003_) - Election Results: percentage of seats for each party/coalition in upper house (variable IMD5004_) - Election Results: percentage of vote for each party/coalition in presidential election (variable IMD5005_) - Expert judgments by the national collaborators of the said party/ coalition's ideological family placement (variable IMD5011_). - Expert judgments by the national collaborators of the said party/ coalition's left-right placement (variable IMD5012_). - Number of cabinet portfolios held by each party/coalition before the election (variable IMD5029_). - Number of cabinet portfolios held by each party/coalition after the election (variable IMD5031_). To allow users to see what party/coalition A-I refers to in a particular election within a polity, variable IMD5000 provides identifiers within the dataset with detailed labels. In addition, Part 4 of the CSES IMD Codebook details the alphabetical classifications for each polity by Standalone CSES Module. Users can search in Part 4 of the CSES IMD Codebook with the following term: "ALPHABETICAL PARTY CODES BY CSES MODULE" <<>> CSES ALPHABETICAL LEADER CODING CSES IMD includes relational data for leaders of parties/coalitions. In CSES IMD, as in Standalone CSES Modules, this data is coded using an ALPHABETICAL classification with leaders where data is available receiving an alphabetical code (A-I). Leaders A through F tends to be the leaders of the six most popular parties/coalitions or the presidential candidates of these parties. They correspond to parties A-F (i.e., Leader A will be related to Party A in some way, Leader B will be related to Party B, etc...) Leaders G, H, and I are supplemental leaders. They may be related to parties G, H, I, but they do not have to be. These leaders are voluntarily provided by each country's election study and often include data about additional personalities of interest. For example, in a parliamentary system, data about a President might be provided, even if the Presidency is not being contested. On many occasions, slots for Leader G, H, and I will include additional data for parties/coalitions that have multiple leaders. These codes are used to identify the following in the micro and macro components of the CSES dataset: - Respondent's likability of the leader/personality in question (variable IMD3009). To allow users to see what Leader A-I refers to in a particular election within a polity, Part 4 of the CSES IMD Codebook details the alphabetical classifications of leaders for each polity by Standalone CSES Module. Users can search in Part 4 of the CSES IMD Codebook with the following term: "ALPHABETICAL LEADER CODES BY CSES MODULE" --------------------------------------------------------------------------- >>> PLACEMENT OF QUESTIONS IN CSES IMD AND STANDALONE CSES MODULES --------------------------------------------------------------------------- When using variables in the CSES IMD, users are advised that even when a question is worded identically in successive surveys, its placement in the survey questionnaire may be different in each Standalone CSES Module, with unknown effects. Thus, even when a question is worded identically in successive surveys, analysts may wish to examine the placement of the question in each questionnaire to ensure that changes in its placement do not contaminate analyses. --------------------------------------------------------------------------- >>> RECODING OF VARIABLES FOR CSES IMD --------------------------------------------------------------------------- Variables for CSES IMD have been recoded to be consistent over time. Questions (and variables) are not necessarily coded the same way in CSES IMD as they are in the Standalone CSES Modules. Users are advised to consult the CSES IMD Codebook Part 2, where possible, details have been provided in VARIABLE NOTES and ELECTION STUDY NOTES regarding harmonization schemes and specific election study deviations. Users are also advised to have to consult Standalone CSES Module Codebook for specific information on Module-specific studies. --------------------------------------------------------------------------- >>> DEVIATIONS FROM STANDALONE CSES MODULES --------------------------------------------------------------------------- The CSES goal is to keep deviations between Standalone CSES Modules and CSES IMD to a minimum. Nonetheless, despite question wordings being virtually identical across CSES modules, CSES IMD harmonizes variables to be consistent over time, thus necessitating some deviations to occur. These deviations may occur because of coding errors found in the Standalone CSES dataset or Standalone Codebooks. In these circumstances, these errors may result in changes being implemented in CSES IMD and the issues listed in errata on the Standalone CSES Module pages. These issues will be corrected in the Standalone CSES Modules upon re-release of these respective Standalone CSES Modules at some future date. Deviations may also legitimately arise because of different coding schemes applied in Standalone CSES Modules compared with CSES IMD, lack of data at the time of processing of Standalone CSES Modules which has since become available, or that data for certain variables were not collected in particular Standalone CSES Modules but have become eligible for inclusion in CSES IMD as it meets the “3 and 1” eligibility criteria. In line with the policy of minimizing differences between CSES IMD and Standalone CSES Modules, CSES only applies deviations between CSES IMD and Standalone CSES Modules when the above circumstances are met. In circumstances where CSES is unable to explain why coding differs between studies from the same polity over time, no changes are made to the data. Concerning deviations, Election Study Notes and Polity Notes are included under the applicable variables noting the deviations. Irregular codes discovered in Standalone CSES Modules and which could not be identified were set to missing. These are also noted in Variable Notes and Election Study Notes. Users can identify these by using the following unique search term in Part 2 of the CSES IMD Codebook: - VARIABLE NOTES - [POLITY NOTES] - [ELECTION STUDY NOTES] --------------------------------------------------------------------------- >>> IDENTIFICATION VARIABLES --------------------------------------------------------------------------- There are several identification variables in CSES IMD which allow users to not only identify an individual respondent, but election studies, and polities. <<>> ELECTION STUDY IDENTIFIERS Each Election Study in CSES IMD is uniquely identified by two variables, namely: - variable IMD1003 ID VARIABLE - ELECTION STUDY (NUMERIC POLITY) This variable is an eight-digit numerical code constructed from two components: the CSES polity code (variable IMD1006) and the year in which the election took place (IMD1008). The first three digits represent the country codes assigned by the United Nations Statistics Division. The fourth digit distinguishes between multiple election studies within a single country for the same election. The final four digits represent the year of the election. E.g., 03602013. AUSTRALIA (2013) - variable IMD1004 ID VARIABLE - ELECTION STUDY (ALPHABETIC POLITY) This variable is an alphanumerical code constructed from two components: the alpha-3 country codes assigned by the United Nations Statistics Division. The remaining characters correspond to the year of the election. E.g., USA_2012 <<>> POLITY IDENTIFIERS Each Polity in CSES IMD is uniquely identified by two variables, namely: - variable IMD1006_UN ID COMPONENT - POLITY UN CODE This variable consists of the numeric country codes assigned by the United Nations Statistics Division to polities. E.g., 372. IRELAND (2011) - variable IMD1006_NAM ID COMPONENT - POLITY NAME This variable consists of polity names based on those used by the United Nations Statistics Division. E.g., New Zealand These polity identifiers allow for easy data bridging with other macro data sources such as the World Bank. <<>> RESPONDENT IDENTIFIER Respondents can be uniquely identified in the dataset by variable IMD1005. It is an 18-character identifier. The first three characters are the numeric version of the country codes assigned by the United Nations Statistics Division. The next four characters denote the year of the election. The remaining characters constitute the respondent ID within that particular election study (which is also equivalent to variable IMD1008_RES). --------------------------------------------------------------------------- >>> MISSING DATA --------------------------------------------------------------------------- Multiple response categories can relate to missing data relating from not applicable to a respondent refusing to answer or failing to answer a question. Users should consult individual variables for the specific missing designations assigned to each variable. 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. Moreover, 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, users should refer to the original questionnaires of each country. These are available on the Standalone CSES Module pages. While there is no consistent CSES convention regarding the application of missing values, some commonalities exist, namely: - Not applicable values are commonly designated as 7, 97, 997, 9997, etc... - Don't know values are commonly designated as 8, 98, 998, 9998, etc... - Missing values are commonly designated as 9, 99, 999, 9999, etc... However, users are advised that the commonalities do not always hold, and they are advised to consult individual variables for the specific missing designations assigned to the variable in question. --------------------------------------------------------------------------- >>> WEIGHTS --------------------------------------------------------------------------- CSES IMD provides users with up to three original weights from each national election study (see variable IMD1010_) namely: - ELECTION STUDY SAMPLE WEIGHT (variable IMD1010_1): intended to correct for unequal selection probabilities resulting from booster samples procedures for selection within the household, non-response, or other sample design features - ELECTION STUDY DEMOGRAPHIC WEIGHT (variable IMD1010_2): intended to adjust sample distributions of socio-demographic characteristics to more closely resemble the characteristics of the population - ELECTION STUDY POLITICAL WEIGHT (variable IMD1010_3): intended to reconcile discrepancies in the reported electoral behavior of respondents vis-a-vis official electoral counts. For more information on polity weights, users are advised to consult the Standalone CSES Module Codebooks. Analysts are advised to read the weight documentation carefully to ensure that their analyses are weighted appropriately (if applicable). The CSES project does not provide advice as to which weights are appropriate to use in particular circumstances. This is best left to analysts to decide based on their detailed knowledge of the research question under investigation. --------------------------------------------------------------------------- >>> 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 polity 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 regarding inclusion of particular polities in an analysis is thus left to users. To assist users in making appropriate decisions concerning their analysis, CSES IMD includes two measures about the freedom and liberty of a polity in the year the election was held (and indeed the two preceding years), namely: - FREEDOM HOUSE RATING (variables IMD5050) Freedom House assigns a numerical rating of a polity on a scale of 1 to 7 providing an indication of freedom. - POLITY IV DEMOCRACY-AUTOCRACY RATING (variables IMD5051) Polity IV assigns a numerical rating to a polity on a scale of -10 to 10 indicating whether the country is strongly democratic or strongly autocratic. Freedom House and Polity IV are not affiliated with the CSES project. --------------------------------------------------------------------------- >>> PROCESSING CHECKS OF INTEGRATED MODULE DATASET (IMD) BY THE CSES SECRETARIAT --------------------------------------------------------------------------- Besides combining and harmonizing individual studies from each module for IMD to ensure they are fit for comparative analysis, a key role of the CSES Secretariat is to perform several checks on the CSES IMD before it is released. These checks include (but are not confined to): - CHECK OF DUPLICATE IDs Identification of respondents with corresponding answers to all questions or respondent identification data that are similar - PARTY/COALITION HARMONIZATION CHECKS To ensure the accuracy of the harmonization of party/coalition codes developed specifically for CSES IMD (see variables IMD3002, IMD3004, and IMD3005_3). - IRREGULAR AND EXTRAORDINARY CODE CHECKS Identification of irregular and extraordinary codes in the CSES IMD Dataset. Sometimes these irregular or extraordinary codes are legitimate in the sense that they may be accounted for by a polity deviation on a particular variable. Other times, these data may have been miscoded in the Standalone CSES Modules. In circumstances where the CSES Secretariat could verify the miscode and a potential correction, these irregular and extraordinary codes were altered. Where this was impossible, the codes were left as is but documented in CSES IMD. - THEORETICAL CHECKS These series of checks explore expected relationships between variables that we might expect to occur (e.g., correlation between Political Efficacy and Satisfaction with Democracy). We do this by exploring distributions, correlation analysis, and regression analysis across CSES IMD but also extending the checks to the Standalone CSES Modules. - DATA OVERVIEW CHECKS These series of checks if the data particulars (such as total number of observations, number of observations per polity, and the data collection period are consistent with documentation. - MACRO DATA CONSISTENCY CHECKS These series of checks explore the internal consistency of macro data over-time. - VARIABLE AND VALUE LABEL CHECKS To ensure all variables in CSES IMD are appropriately assigned labels and documented in the CSES IMD Codebook. If you identify any potential issue with the CSES IMD data, please contact the CSES Secretariat by e-mail at: cses@umich.edu --------------------------------------------------------------------------- >>> HOW TO MERGE STANDALONE CSES MODULE DATA NOT INCLUDED IN CSES IMD TO THE CSES IMD FILE? --------------------------------------------------------------------------- There are several ways to merge administrative, demographic, or survey level data from the Standalone CSES Modules (and which have not been included in CSES IMD due to ineligibility or phased development of the product) to the CSES IMD dataset. One option is to use Variable IMD1005 (ID Variable – Respondent), which uniquely identifies each respondent within the IMD dataset and which also identifies a respondent within the Standalone CSES Module. Regarding merging macro data from the Standalone CSES Modules (and which have not been included in CSES IMD due to ineligibility or phased development of the product) to the CSES IMD dataset, several polity and election study level identifiers within CSES IMD allow for amalgamation. They include: - IMD1003 which is an Election Study Identifier detailing the Numerical Polity Code and Election Year. - IMD1004 which is an Election Study Identifier detailing the Alphabetical Polity Code and Election Year. - IMD 1006_UN which is a Polity Identifier detailing the UN Numerical Polity Code assigned to each polity. - IMD1006_NAM which is a Polity Identifier detailing the Alphabetical verbatim name of each polity Regarding merging of relational data for parties/leaders assigned alphabetical codes, as parties/leaders in CSES IMD retain the alphabetical coding assigned to parties/leaders in Standalone CSES Modules, these data can be merged using the alphabetical codes along with the appropriate election study, polity, or Standalone CSES Module specification. --------------------------------------------------------------------------- >>> HOW TO SELECT DATA FROM ONE OR MULTIPLE STANDALONE CSES MODULES OR ELECTIONS HELD IN A PARTICULAR YEAR IN CSES IMD? --------------------------------------------------------------------------- The CSES IMD contains two variables that allow selection of data from particular Standalone CSES Modules or elections held in a particular year. IMD1008_Year allows users to select particular election years for analysis. IMD1008_MOD_1, IMD1008_MOD_2, IMD1008_MOD_3, IMD1008_MOD_4 allows users to select particular Standalone CSES Modules. =========================================================================== ))) CSES IMD BIBLIOGRAPHY =========================================================================== To construct the CSES Standalone Module Codebooks, the CSES Secretariat has consulted various sources that are listed in Part 1 in each of the CSES Standalone dataset Codebooks. The below list constitutes a list of additional sources that the CSES Secretariat has consulted in the development of the CSES IMD Codebook. The list, while thorough, is not comprehensive. Instead, all sources not listed here are listed in the appropriate section of the CSES IMD Codebook that the information refers to. Bormann, N.-C. & Golder, M. (2013). Democratic Electoral Systems Around the World, 1946-2011. Electoral Studies, 32, 360-369. Carey, J.M. (2017). Electoral Formula and Fragmentation in Hong Kong. Journal of East Asian Studies, 17, 215-231. GMA News Online (2012). 62 party-list groups delisted from May 2013 polls. Available at: https://www.gmanetwork.com/news/news/nation/256768/ 62-party-list-groups-delisted-from-may-2013-polls/story/ (Date accessed: January 27, 2018) Hesli, Vicki L. & Reisinger, W.M. (2003). The 1999-2000 Elections in Russia. Their Impact and Legacy, Cambridge University Press Hsieh, J.F. & Niou, E. (1996). Taiwan's March 1996 election. Electoral Studies, 15(4), 545-550 Iglesias, J.D. (2013). New Nation-states and National Minorities. ECPR Press. International Institute for Democracy and Electoral Assistance (IDEA) http://www.idea.int/data-tools/data/voter-turnout (Date accessed: May 17, 2018) Lansford, T. (2014). Political Handbook of the World 2014. Sage. LSM.LV (2014). Political parties get their numbers. Available at: https://eng.lsm.lv/article/politics/politics/political-parties-get-their- numbers.a94067/ (Date accessed: September 4, 2018) Maiz, R. & William, S. (2000). Identity and Territorial Autonomy in Plural Societies. Routledge. Moon, E.P. (1997). Single Non-Transferable Vote Methods in Taiwan in 1996: Effects of an Electoral System. Asian Survey, 37(7), 652-668. Nohlen, D. (2005). Elections in the Americas: A Data Handbook: Volume 2: South America. Oxford University Press. Nohlen, D. & Grotz, F. (2002). Elections in Asia and the Pacific: A Data Handbook: South East Asia, East Asia and the South Pacific. Oxford University Press. Parliaments and Government Database (ParlGov) http://www.parlgov.org/#data (Date accessed: April 4, 2018) POLITY IV Project: Political Regime Characteristics and Transitions, 1800-2007, Monty G. Marshall and Keith Jaggers, George Mason University and Colorado State University. Available at: http://www.systemicpeace.org/polity/polity4.htm (Date accessed: May 17, 2018) The Diplomant (2011). Thailand's 'Vote No' Campaign. Available at: https://thediplomat.com/2011/06/thailands-vote-no-campaign/ (Date accessed: September 4, 2018) The Polity IV Dataset Users' Manual. Available at: http://www.systemicpeace.org/inscr/p4manualv2012.pdf (Date accessed: November 25, 2010) The Polity IV annual time-series dataset. Available at: http://www.systemicpeace.org/inscrdata.html (Date accessed: May 17, 2018) Toro, Sergio. Y. & Luna, J. P. (2011). The Chilean elections of December 2009 and January 2010. Electoral Studies 31(1), 226-230. doi: 10.1016/j.electstud.2010.08.005 World Bank (n.d.). Worldwide Governance Indicators. Available at: https:// databank.worldbank.org/data/source/worldwide-governance-indicators (Date accessed: June 28, 2019) //END OF FILE