News and Errata

Errata of Version 2.0

2009-09-07:

• In case of Slovenia the values of the variables RINCOME (Respondent's income) and INCOME (Family income) are completely mixed up in the integrated file due to an error in the original data of Slovenia. Please note that for Slovenia RINCOME should be replaced in fact by INCOME and vice versa. 

2008-12-04:

• For Czech Republic the coding of V56 'Are you citizen of' is completely reversed in the version 2.0 due to a coding error in the original Czech coding scheme. Please use the SPSS syntax statement as follows before analyzing the integrated data file:

Do if V3=14. Recode V56 (1=2) (2=1) (else=copy). End if. Execute.

• For Japan and Norway the values of Repondent's and Family income are based on yearly income and NOT on monthly income. Additional the continuous values for Japan are declared in thousand Yen. 

Version 2.0 - 2007-08-30

As of 2007-08-30, a revised version (2.0, version date Aug 27, 2007) of the integrated file is available through the ZACAT online platform. Change in this version:

• data in variables ATTEND for a number of countries and DEGREE for Ireland corrected, see below for details.
• USA have provided a corrected Study Description.
• Other changes:
  - Variable "ZA Study Number" now holds the ZA No. (3910) instead of the module year (2003).
  - Variable "Respondent Number": now standardised for all countries, 8-digit number, begins with Country ID from V3, last digits are the country specific ID numbers.
  - Variable "Country": Code 11 "Netherlands" added, Code for "Uruguay" changed to '43' (for consistency with ISSP2004, ZA3950).
  - Variable (AGE recoded to classes) has been removed.
  - A number of missing value definitions has been set, in particular in country specific variables.

Errata of Version 1.1

2007-09-03:

• The Study Description contains incorrect figures about the gross sample and non-respondents.
• Variable ATTEND: A filter condition has erroneously been enforced where respondents with no religion/denomination (code 0 in RELIG) were recoded to a missing value of 0 "NAP" on ATTEND if they previously had a valid response of NO church attendance (code 8 on ATTEND). This cannot be reliably undone by recoding data of the integrated file, please use version 2.0 if you want to analyse ATTEND.
• Variable DEGREE, data for Rep. of Ireland: The derivation of DEGREE from the national variable IE_DEGR is obviously incorrect, having far too many cases under "no formal schooling" of DEGREE. Please do not use DEGREE for Ireland in version 1.1, but fall back to IE_DEGR and recode that according to your own needs.

Version 1.1 - 2007-02-20

• As of 2007-02-16, a revised version (1.1, release date Jan 29, 2007) is available through the ZACAT online platform. This version fixes the coding errors of the CD-ROM version reported below. (The data set version is stored in variable 'version'. If your data set has no variable named 'version', it is version 1.0.)
  

Errata of the CD-ROM version (1.0)

• The questionnaires for (both Latvian and Russian language versions) as originally offered here and on the data CD-ROM are incomplete. Latvia has now provided complete questionnaires. Since 2006-02-06, the links above lead to the complete versions!
• We have identified a coding error in the SPSS system file <ISSP2003_V3Final_rev2.sav> contained on the CD-ROM. Only for data from , responses from Q.24b (relating to EU members) are incorrectly stored in , otherwise holding responses from Q.24 (relating to EU members). In V74 (meant to hold Q.24b), where Spain should be, all Spanish cases are incorrectly set to '0'.
• If you are working with the <ISSP2003_V3Final_rev2.sav> file, you can simply copy the data for the Spanish cases from V73 to V74. In SPSS, you would use a syntax statement like this to do that:

IF (V3=25) V74=V73. /* V3=25: Country=Spain.
EXECUTE.
Please note that this error has already been corrected in the dataset version that is available online through the ZACAT system!

• Questionnaires for Russia, Rep. of Ireland, and Venezuela were not available when the CD-ROM was produced, but have been provided later on and are now available here.
• For Austria, the coding of variable "v291 : R:Top Bottom self-placement 10 pt scale" is completely reversed in version 1, date March, 2006 of the data set.
  In SPSS, you can fix this as follows:

do if v3=7 and v362=1.
recode v291 (1=10) (2=9) (3=8) (4=7) (5=6) (6=5) (7=4) (8=3) (9=2) (10=1) (else=copy).
end if.
execute.