Research Potential of Official Microdata

Using selected research questions from the areas of household and family, participation in education, income, migration, and socio-economic status, this article outlines the analytical potential of the microcensus scientific use files dating from 1989 to 1997 as model examples and summarizes suggestions for improving both survey program and data provision.

Schimpl-Neimanns, Bernhard (2002): Anwendungen und Erfahrungen mit dem Scientific Use File des Mikrozensus. ZUMA - Arbeitsbericht 2002/01 (363 kB)

To facilitate easier user access of anonymized microdata by the research community and to develop sustainable solutions over time, since the end of 1996, a pilot project has been in operation that is financed by the Federal Ministry of Education and Research (BMFB). It was developed by the Federal Statistical Office in cooperation with GESIS. The article outlines the current status of the project due to expire by the end of 2002 or 2003 respectively, with particular emphasis on experiences with the microcensus and on future prospects.

Köhler, Sabine; Schimpl-Neimanns, Bernhard; Schwarz, Norbert (2000): Pilotprojekt zur Erleichterung der Nutzungsmöglichkeiten von faktisch anonymisierten Mikrodaten. Wirtschaft und Statistik 1: 30-37.

The critical review outlines the key results of the Advisory Board’s activity, dated 1989, which have been partially implemented. However, legislators have not adopted many recommendations that today might still be regarded as pointing the way ahead for the microcensus. The reasons for the discrepancy between the Scientific Advisory Board’s recommendations and their implementation can be seen in the current legal framework, which not only contributes to the inflexibility of official statistics in Germany, but also hardly offers any opportunities for the research community to participate in decision-making about the federal statistics program.

Schimpl-Neimanns, Bernhard (1999): Zur Umsetzung von Empfehlungen des Wissenschaftlichen Beirats für die methodische und inhaltliche Weiterentwicklung des Mikrozensus - ein Rückblick. S. 47-51 in: Statistisches Bundesamt (Hrsg.): Kooperation zwischen Wissenschaft und amtlicher Statistik - Praxis und Perspektiven. Beiträge zum Symposium am 31. Mai/1. Juni 1999 in Wiesbaden. Schriftenreihe Forum der Bundesstatistik, Band 34. Stuttgart: Metzler-Poeschel.

By reference to examples, and with both practical considerations and a critique of methodology, an additional essay highlights the analytical potential of the 1995 microcensus. References are made to problems of operationalization in relation to research questions focusing on areas such as the labor market, households, families, and social structure. Non-response, variance estimation, and weighting or projection are discussed from a methodological perspective.

Schimpl-Neimanns, Bernhard (1998): Analysemöglichkeiten des Mikrozensus. ZUMA-Nachrichten 42: 91-119. PDF-Datei (405KB) (202 kB) 

Since the late 1980s, the microcensus has been subjected to various changes (survey of the newly-formed German states, new sampling plan, conversion of some questions to voluntary-response questions, new microcensus law since 1996, etc.). These changes, along with the fact that accessing data was simplified substantially since the 1995 microcensus, meant that it was necessary to contribute to the debate on reforms over recent years as well as future developments and thus to demonstrate that the microcensus can be openly accessed and used (“research potential”) by a broader spectrum of scientists and researchers.

Lüttinger, Paul; Riede, Thomas (1997): Der Mikrozensus: amtliche Daten für die Sozialforschung. ZUMA-Nachrichten 41: 19-43. PDF-Datei (413KB) (215 kB)

Doubts were frequently expressed about the information value or validity of data drawn from official GDR statistics for analyses of transformation processes. Objections were primarily associated with publications on educational and income inequality. In acknowledging the political selectivity of published statistics, following German unification the decision was partly taken to refrain from attempts to reconstruct the GDR’s social structure on the basis of official GDR data. Yet it is often overlooked that any explanation of social change must be based on knowledge of the original status quo before 1990. The GDR microdata project therefore focused on a critical examination of existing official microdata, among other things, with regard to their validity. In the context of this project, both the research status and the research interest of transformation studies were outlined with regard to educational and income inequality. Against this backdrop, the debate subsequently focused on analysis options and limitations of surveys based on GDR statistics which were still accessible as separate data, and problems relating to the comparison of East and West Germany were highlighted.

Schimpl-Neimanns, Bernhard; Wirth, Heike (1994): Bestandsaufnahme und Nutzungsmöglichkeiten amtlicher Mikrodaten der DDR für Sekundäranalysen zur Bildungs- und Einkommensungleichheit. ZUMA-Arbeitsbericht Nr. 94/06.

Secondary analysis of official data enjoys a long-standing tradition in social research, dating back to the early days of sociology. On the one hand, the introduction of sampling in official statistics, and on the other hand, the revolution in computer technology launched a new phase of secondary analysis of official data that individual researchers are now able to process. Thus, the same potential exists for analyses that are both flexible and as closely as possible aligned to a theoretical research question as in the case of data collections originally conducted by academia. An informative overview is given here of the situation in the USA, where secondary analysis of official microdata is an established method of empirical social research. Equally, an outline is given of the, for various reasons, much more cautious relationship between official statistics and research in Germany. An overview is provided of the totality of microdata currently available in Germany. A more detailed description is provided of the potential of official German microdata as sources of social sciences research based on employment statistics, income and expenditure survey, and the microcensus.

Alba, Richard; Müller, Walter; Schimpl-Neimanns, Bernhard (1994): Secondary Analysis of Official Microdata. In Borg, I. & Mohler, P.Ph. (Eds.), Trends and Perspectives in Empirical Social Research (pp. 57-78). New York: de Gruyter.

The paper provides an elaborated description of the work that is done in one of the various departments within the Center for Survey, Research and Methodology (ZUMA), the Department of Microdata. The first project deals with the identification risk of microdata; the second describes some results dealing with the representativity of surveys.

Lüttinger, Paul; Wirth, Heike; Hippler, Hans-J. (1993): Different Aspects in Using Microdata: Data Anonymity and Non-response Bias. Journal of Official Statistics 9 (1): 217-233