Thank you for your interest in the "Party
Systems and Electoral Alignments in East Central Europe" project of the
Political Science Department of the Central European University. This project
was part of the department's 1992-97 project on "Transition to What?"
and sponsored by the Central European University Foundation. The data are in the
public domain and no fee has to be paid by the prospective users.
In order to receive the
CEU data files, please sign the users'
declaration form and return it to me together with a short abstract of the
work (paper, book, etc.) for which you liked to use these data. Please let me
know an email or FTP address where I can send you the data.
Due to the limited
availability of qualified manpower, the documentation of the surveys may leave
something to be desired. I myself worked on this project in part time without
any assistants to help me, and occasionally forgot to squeeze out original
language questionnaires and detailed technical reports from the polling
organizations. For the time being, an incomplete edition of the files is
distributed directly by the original data collector (i.e. the Department of
Political Science). This includes all the data sets, except some special
election surveys in Poland in 1993, in the Czech Republic in 1996, and in
Hungary in 1998, all of which had a very different questionnaire than the rest
of our studies. The available data files have an SPSS file format and are
labeled in English. The variable names are standardized across the files.
The surveys were carried
out with the assistance of one of the major commercial political polling
institutes in each country: CBOS in Poland, STEM in the Czech Republic and
Slovakia, and Median in Hungary. The samples were drawn from the
non-institutionalized adult population of the respective countries. In Poland we
used clustered random sampling. In the Czech Republic and Slovakia we started
with clustered random sampling but switched to quota from 1993. In the Hungarian
studies the method of sampling always was clustered random route. As of 1998, a
final version of the 1994 Hungarian election panel data was released through the
Hungarian, German and Slovenian social science data archives. This contains the
September 1992, December 1993, and April 1994 pre-election survey data, the data
set from the May 1994 post-election survey, and documentation and
back-translation of the questionnaires.
Survey (country, date)
File name
Poland, October 1992
p92
Poland, January 1993
p93jan
Poland, August 1993
p93aug
Poland, April 1994
p94apr
Poland, December 1994
p94dec
Poland, June 1995
p95apr
Poland, December 1995
p95dec
Czech R., September 1992
cr92
Czech R., April 1993
cr93apr
Czech R., November 1993
cr93nov
Czech R., April 1994
cr94apr
Czech R., November 1994
cr94nov
Czech R., June 1995
cr95jun
Czech R., January 1996
cr96jan
Slovakia, September 1992
sr92
Slovakia, April 1993
sr93apr
Slovakia, November 1993
sr93nov
Slovakia, April 1994
sr94apr
Slovakia, Nov. 1994-Jan. 1995
sr94nov
Slovakia, July 1995
sr95jul
Slovakia, April 1996
sr96apr
Hungary, September 1992
h92
Hungary, January 1993
h93jan
Hungary, December 1993
h93dec
Hungary, April 1994
h94apr
Hungary, June 1995
h95jun
These data files are
contained by the POLAND.ZIP, CZECH.ZIP, SLOVAKIA.ZIP, HUNGARY.ZIP and
CEU94HUN.ZIP files, which you can open with PKUNZIP.EXE.
In addition to the data
files, an English language version of the CEU-modules (these typically
constituted the larger part, and in a few instances the entire questionnaire of
the individual surveys) is distributed. These document files have a Rich Text
format (extension *.rtf). Any Windows-based word processor should be able to
read them once you unzipped them. Also CONVERT.EXE of the Word Perfect 5.1
package can be used to convert them into WP for DOS format. The first file that
you should look into is CEU_DATA.RTF, which gives an overview of the data sets
(including the files that are not yet distributed).
Generally, the standard
demographic variables have self-explanatory names, such as SEX, YBIRTH, URBRUR,
etc. Only in the Hungarian files you will find a weighting variable (WEIGHT).
The core CEU-items (starting with interest in politics and normally ending,
within individual questionnaire, with left-right and/or liberal-conservative
self-placement) have the same name in every file. Q1 always stands for interest
in politics, Q2 always is satisfaction with democracy (irrespectively of whether
or not that item was the second question in the questionnaire of the given
survey). The document files distributed together with the data files are in the
QUESTION.ZIP file (again, you have to use PKUNZIP.EXE to open it).
There are some variables
in the data-files that have names starting either with P (these are
Poland-specific items), or CS (items which were only asked in Slovakia and the
Czech Rep), or H (Hungary-specific variables). Usually, these variables were
provided by the local polling institute, although I did not request them. In
some other cases (most notably in the 1993 Polish, 1994 Slovak, and 1994
Hungarian post-election study) they were requested by me but were only asked in
one country at a time.
I have in my possession
copies of most of the original (Polish etc.) language questionnaires, and
photocopies of documents received from the polling organizations about the
sampling procedures that they followed. These documents are not normally sent to
the prospective users but the information on sampling will certainly be
incorporated into the standard documentation at some stage.
If you have comments and
suggestions concerning any aspect of the documentation, the questionnaire
design, this letter, the users' declaration form or anything related to the CEU
project, I would be very grateful if you could let me know. It would also be appreciated if you
can send any bibliographic as well as other information about publications and
conference papers making use of these data sets.
Back to previous page Back to main page
Send your comments to tokag@ceu.hu.