 |
Monday, Sep 8 |
|
| ||||
Study Name: Sozio-oekonomisches Panel (SOEP)/German Socio Economic Panel (GSOEP)
Study Director: Gert G. Wagner
Host Organization: DIW Berlin (German Institute for Economic Research)
Year Initiated: 1984 west Germany, 1990 east Germany
- Funding sources
The SOEP was originally conducted by a project of the Special Research Unit 3 „Micro-analytical Foundations of Social Policy“ (Sfb3) , which was financed by the Deutsche Forschungsgemeinschaft (German National Science Foundation) at the universities in Frankfurt, Mannheim and Berlin.When the activities of the Special Research Unit came to their scheduled conclusion in 1990, the entire responsibility for the SOEP project was transferred to the DIW Berlin (German Institute for Economic Research Berlin).Since than continuing DFG funding is dispersed through the Federal Committee for Educational Planning and Research Sponsorship (BLK) on a matching basis with all the Federal States of Germany.
The most recent subsample (subsample F) which almost double the size of the overall sample is financed exclusivlyby the Federal Government which is intetrested in more detailed policy analysis and thus needs more cases.
- Host organization
-Governing body and role of external research advisors
The members of the GSOEP Advisory Board of Overseers (up to 13 scientists plus two representatives of the Federal Government and the State Governments) serve a maximum of three three-year terms. The scientific members of the board are appointed by DIW Berlin in collaboration with DFG (German National Science Foundation). The members are representatives of stakeholders of GSOEP in various fields (economics, sociology, methodology, demography-public health may come in next term). An officer of DFG is an observer of the annual meetings of the Board.
Since May 2000 Daniel S. Hamermesh, University of Texasat Austin, is head of the advisory board.
- Sample selection
The
original GSOEP sample, starting in 1984, consisted of the two German random
samples. Every individual 16 years of age or older in the household is surveyed
for all partial samples (institutionalized households in both West and East
Germany do not belong to the population of the originalsamples of the SOEP):
Sample A: Individuals in private households not headed by someone with Turkish, Greek, Yugoslavian, Spanish or Italian citizenship (about 4,500 households) and
Sample B: Individuals
in private households headed by someone with Turkish, Greek, Yugoslavian,
Spanish or Italian citizenship (about 1,400 households). In order to allow separate
conclusions about these five groups of labor migrants most strongly represented
in the Federal Republic of Germany at that time, they were over-represented in
the study.
To allow the best observation of the great changes to society in the territory of the former GDR, along with their individual repercussions, the first wave of the sample for East Germany was collected in June 1990, before the currency, economic, and social union, and was defined as follows:
Sample C: Individuals
in private households of the GDR headed by someone with German citizenship
(citizen of the GDR) (almost 2,200 households).
Since the time the original sample was drawn, Central Europe, and especially Germany, have experienced immigration on a large scale. Such a significant change in the population was included in the sample by supplement explicitly the data for immigrants:
Sample D („Immigrant
sample„): Private households in West
Germany in 1994/95 in which individuals live who immigrated in the years from
1984 through 1994/95 (about 500 households). Data were collected
retrospectively in 1994/95 for immigrants who had arrived since 1984.
1998
a „supplementary sample“ was begun to stabilize the number of cases in the SOEP
for cross-sectional and longitudinal data:
Sample E („Supplementary
Sample“): Individuals in private households in East and West Germany in 1998
(about 1,100 households).
Positive
experience from this sample lead to the idea to significantly enlarge the
sample size of GSOEP in order to increase the value of the data for political
analysis by allowing the changes for relatively small groups of the population
to be analyzed on the basis of sufficient numbers of cases:
Sample F („Supplementary
and Innovative Sample“): Individuals in private households in East and West
Germany in the year 2000 (about 5,000 households).
Sample F will be used to introduce some new fields of questions, especially on wealth and health, and for testing a link to register data as well as using the internet/e-mail for contucting data.
In
year 19999 about 15,000 adults participated in the GSOEP survey, who lived in
nearly 7,700 households with around 4,000 children under 16 years of age.
After the end of the fieldwork of sample F in year 2000 GSOEP will consist of about 25,000 adults and 7,000 children living in about 12,000 households.
-Follow rules
The families of individuals that have lived in the household interviewed in the first wave of each sample.
-Oversampling
Oversampling is realized for foreigners (in 1984) by means of sample B and east Germans (in 1990) by means of sample C.
-Weights and attrition bias
The sample is reweighted every wave.
-Sample „refreshing“
The 1995 „Immigrant Sample“ and the 1998 „Supplementary and Innovative Sample F“ are discussed above.
- Demograhics (all issues of demographic development like births, fertility, deaths, life expectancy, migrations, as well as the processes of starting and dissolving families.Detailed information is also available for foreign individuals surveyed, like desire to be naturalized, intentions to re-migrate, indicators of integration).
- Employment/ Income and Wealth / Consumption / Housing
- Health / Social Security
- Education / Culture / Leisure Time / Recreation
- Mode and instrument design (paper and pencil, CAPI)
As a rule, data are collected in face-to-face interviews using completely standardized survey instruments, with the interviewer conducting an oral interview or the respondents filling out questionnaires themselves.In older households which have participated in the SOEP for a long time, centrally administrated surveys without an interviewer are performed on a limited scale. Since 1998, „Computer Assisted Personal Interview“ (CAPI) methods have been introduced gradually to complement the conventional „paper and pencil“ questioning technique.
-Dependent interviewing
nothing
-Calendar year, survey year and point of survey measures
Some of each, but mostly calendar year
-Data and documentation standards
At the beginning of the study paper (what else?) was the medium of documentation. Since 1995 most of the documents are in files of which almost all are accessable via the internet.
A major piece of the documentation ist the „Desktop Companion“ which comes with the data and is on the homepage of the GSOEP as well.
-Dissemination techniques (CD)
Active data users number about 250 domestic and over 100 foreign contractors.In many cases, however, for each data release contract a considerably larger number of scientists work with the SOEP data.
GSOEP data are disseminated in several formats on CD-ROM.The formats include ASCII, SPSS, SAS, STATA and TDA/RZOO.
For reasons of data protection, a special „International Use File“ has been created for users who do not perform research in the areas covered by the German Data Protection Law (BDSG) (95%-random sample of the total cases).
Signing a contract on data distribution with the DIW Berlin is necessary for working with SOEP data. This is not only necessary due to German data protection law, but in the age of internet public use files which are downloadable are very dangerous. We claim that no respondend can by identified. But when a respondend himself download the data she can find herself easily.Thus the decision to give the data to reliable researchers only is a protection against this danger.
For processing the data request the following is needed:
1.the full office address (including telephone and fax number) of
the person in charge of the research project (in the case of a
student or doctoral thesis, the supervisor);
2.the title of the research project or research program in which
you wish to use the data.
Contact <soepmail@diw.de> for further information.
-Virtual data centers and functionalities
The GSOEP
team is eager to bring in the data
into virtuell data centers. We are in contact with the NESTAR team in Norway
and we hope that the Ann Arbor meeting will be the start of a Virtuell Panel
Data Center.
-Improving contractual use of data (confidentiality and encryption)
Again we hope that the Ann Arbor conference will bring some improvements.
-Comparability of formats (evolution of data standards)
GSOEP data are in many different comparative data sets available:
· CNEF (Cornell University)
· CHER (former PACO) (CEPS Luxembourg)
· EPAG (Essex University)
· ECHP (EUROSTAT Luxembourg)
-
Institute for Social Research | University of Michigan | Privacy | Conditions of Use
