CSES Module 1 Data Set Errata
Posted: December 15, 2015
Updates made to the Full Release of December 15, 2015, since the Full Release of August 4, 2003
The files released on December 15, 2015 are an update to the prior release of August 4, 2003. Please do not use files from the previous release with files from the current release. Variable names and attributes, as well as the data, have sometimes been changed. However, no records were removed nor added. The respondent, country, and election study count remain the same.
With this release, the data and syntax file formats, and the codebook, have been improved to be more consistent with other CSES data releases. Furthermore, newly accompanying the release are datasets already prepared and ready to be immediately loaded into R, SAS, SPSS, and STATA. The dataset has also been made available in CSV format (albeit without metadata beyond variable names in the column headers).
Last, a number of corrections were made to known issues in the data and documentation. The CSES user community originally made us aware of many of these issues, and we thank you for that feedback. A list of some of the most important changes made is as follows, though this is not a comprehensive list of all changes:
- Some macro level features which refer to the United Kingdom were adjusted so that they refer only to Great Britain, and not the United Kingdom. The affected variables are: A5005_A - A5005_E (PERCENT VOTE - LOWER HOUSE - PARTY A to E), A5006_A - A5006_E (PERCENT SEATS - LOWER HOUSE - PARTY A to E), A5010 (ELECTORAL TURNOUT), A5026_1 (PRIMARY DISTRICTS - LOWER - 1ST SEGMENT), A5027_1 (NUMBER OF SEATS - LOWER - 1ST SEGMENT).
- In the United States (1996) portion of the CSES Module 1 datafile, the following macro variables had to be corrected due to an error during CSES processing: A5005_A-A5005_C (PERCENT VOTE - LOWER HOUSE - PARTY X), A5007_A-A5007_B (PERCENT VOTE - UPPER HOUSE - PARTY X), A5008_A (PERCENT SEATS - UPPER HOUSE - PARTY A).
- For the Switzerland (1999) data, corrections suggested by the Swiss collaborator have been implemented for the following seven macro variables: A5031, A5032_1, A5032_3, A5034_1, A5034_3, A5035_1, and A5042. Furthermore, new election study notes have been implemented for variables A5031, A5032_1, A5032_3 , A5034_1, A5034_3, A5035_1, A5042.
- In the Denmark 1998 portion of the CSES Module 1 data file, cases coded '9. MISSING' in A3010 were now coded '0. NOT APPLICABLE' in A3011.
- In the Spain (1996) and Spain (2000) election studies, variable A5051 (HEAD OF STATE - COMMANDER OF FORCES) was corrected to be coded '1. YES' instead of '5. NO'.
- The question wording used in the 1998 Dutch Election Study differs somewhat from the question wording in the CSES Module 1 questionnaire. To make users aware of this, we have added election study notes to the codebook for the following variables: A3004 - A3006, A3009 - A3012, and A3012.
- The election study note that appeared in the codebook for Belgium-Flanders (1999) for variables A3004, A3005_1, A3005_2, A3005_3, and A3006 has been corrected to indicate that it actually pertains to Belgium-Walloon (1999).
- An election study note concerning question wording of the Belgium-Flanders (1999) election study was added to the codebook for variables A3004-A3006 and A3009 (concerning closeness to political party).
- For New Zealand (1996), the data in variable A4004_F (PERCENT VOTE IN DISTRICT - PARTY F) has been corrected.
- Missing values for variables A5005_D, A5005_E, A5007_D, A5007_E, A5009_D, and A5009_E in the Mexico (2000) study, were corrected to be coded 997 instead of 97.
- For Thailand (2001), the values for variable A2018 (LANGUAGE USUALLY SPOKEN AT HOME) were corrected.
- For New Zealand (1996), the values for variable A2018 (LANGUAGE USUALLY SPOKEN AT HOME) were corrected.
- An election study note for variables A1010_1 to A1010_3 (ORIGINAL WEIGHTS) has been added to the codebook, clarifying why weights for some respondents were coded 0.
- The numbers for turnout (A5010) of 23 countries (see below) have been amended according to considerations presented in the CSES Technical Report 'Electoral Turnouts Reported in Modules 1 and 2' by Sebastian Netscher (October 2010).
- An election study note for Portugal (2001) for the variables A1026-A028 (DATE QUESTIONNAIRE ADMINISTERED) has been added to the codebook.
- For Germany (1998), the sample component specified in variable A1007 now differentiates between West and East Germany.
- For Peru (2000), variable A2022 (RURAL OR URBAN RESIDENCE), an election study note has been added to the codebook pertaining to the sampline in urban vs. rural areas for this election study.
- The election study note for Poland (1997), variable A2019 (REGION OF RESIDENCE), has been adjusted to not include the category '8. DON'T KNOW' anymore.
- In variable A2021 (ETHNICITY), 'refusal', 'don't know' and similar categories had been coded differently across countries. For the missing statement distributed by CSES to also apply to these cases, they were recoded to '999. MISSING'.
- In Module 1, there were no country labels associated with the variable A1003 (ID VARIABLE - ELECTION STUDY (NUMERIC POLITY)) as there are in Module 2, Module 3, and Module 4. They have now been added.
- Variable A2005 (UNION MEMBERSHIP) was coded missing for Denmark (1998) as the data distribution did not accurately reflect the reality at the time of the study. An election study note commenting this was added to the codebook.
- There was a lack of consistency in applying '999. MISSING' vs. '0. NOT APPLICABLE' to variables A4002 (NUMBER OF CANDIDATES IN DISTRICT) and A4003 (NUMBER OF PARTY LISTS IN DISTRICT), both within and across modules. This has been corrected such that for countries emplyoing party lists, A4002 (NUMBER OF CANDIDATES IN DISTRICT) is now coded '0. Not Applicable' and not '999. Missing'. There is one exception to this: A4002 also lists candidates from some systems with open lists, because there too voters can vote for candidates. This is in accordance with the variable description in the Module 1 codebook.
- For the elections in Israel (1996), Netherlands (1998), and Peru (2000), there was only one primary district, namely the whole country. This was indicated by coding these cases '0' on variable A2027 (PRIMARY ELECTORAL DISTRICT). However, in later modules, such cases have been coded '1'. To increase consistency across the modules, we adjusted Module 1 to now also take the value '1' for Israel (1996), Netherlands (1998), and Peru (2000).
- The election study note for Poland (1997) for variable A2021 (ETHNICITY) was removed from the codebook as it listed a category not present in the data.
- For variable A2021 (ETHNICITY), Israel has codes 93 and 99 in the data, but the documentation of this election study does not provide information on what these values refer to. They have thus been coded missing.
- For variables A3010 (DO YOU FEEL CLOSER TO ONE PARTY), we made the correction to recode 319 respondents from category '9. MISSING' into category '5. NO'.
If you have questions or concerns, e-mail us at firstname.lastname@example.org and we would be pleased to help.