Towards an Infrastructure of Comparative Research on Europe
by Franz Kraus
During the last years, research on European societies and on European integration has increased. Concomittantly, discussions on proper infrastructures have intensified. Our newsletter, over a number of issues, will adress to that topic - partly through historical background information, partly through the treatment of specific aspects of actual interest. The sequence of contributions starts with a scetchy overview of the evoluton of the current infrastructure.
A quantitative analysis of human behaviour and societal structures requires information on the behaviour and attitudes of individuals and groups, on macro-societal contexts and institutional settings. Major sources of information today are official statistics, academic and private survey research and institutional documentations. Scholars who want to conduct cross-national research face a wide spectrum of potential problems: they need to know the current state of research, find the proper data sources, evaluate their comparability, build up a suitable database and apply adequate tools of analysis. Problems seem almost unsurmountable if a scholar considers time and space as important analytical dimensions.
In many cases the limitation of research to readily available information is sensible in such a situation and happens often even if research is carried out by international teams. But it is also obvious that progress in cross-national research requires not only intelligent use of available data but also the expansion of the data infrastructure itself. In addition, the maintenance of proper documentation services and of international forums for the cooperation of individual scholars, research institutes, national research councils and foundations is necessary.
National statistical institutes and associations
The problem is by no means new. Since the early days of quantitative, comparative analysis of social phenomena in the 17th and 18th century Europe, science has taken care of the cumulative development of a proper infrastructure of comparative research. On occasion of the establishment of national statistical institutes in almost all European countries during the 19th century (Flora 1975), many social scientists took over responsibilities. Frequently they were even in charge of the institutes. Statistical associations and learned societies supported the development of statistical tools and the standardization of enumeration programmes. In the mid-19th century, concern for international comparability of statistics and the international diffusion of statistical expertise had even led to the foundation of an international forum for discussion, the International Statistical Congresses.
International Statistical Institute (ISI)
By the end of the 19th century, the so-called pro-statistical era had come to its end (Landes 1972). In virtually all European countries, statistical institutes and advisory councils were in service and in many countries municipal statistical offices were established. The synchronous and regular carrying out of population censuses as well as the gradual introduction of economic censuses - both pillars of economic and soial statistics until today - was internationally agreed upon. In 1885, the International Statistical Congresses had become an organization with its own recources and governing bodies - the International Statistical Institute. Apart from the close cooperation between the statistical institutes in Scandinavia, the Institute, for a number of decades, was the only institution that promoted the development of national and international statistics (Macura/Cleland 1985). Data quality and comparability, over time and country, replacement of administrative records through stratified sample surveys, and publication of cross-national statistics and studies were major concerns of the ISI. Individual scholars, often supported by learned societies, produced international guides to official statistics, international bibliographies of government publications and data handbooks.
The rise of international official statistics
During the interwar period, ISI and the academic community continued their efforts to generate an international infrastructure for comparative research. An additional incentive came with the foundation of the League of Nations (1919) and the International Labour Organization (1919). Now, for the first time inter-governmental organizations took over responsibilities for the expansion and standardization of national statistics and cross-national reporting. ISI cooperated with these institutions in various fields and published voluminous international data handbooks, particularly in the fields of vital and of urban statistics.
Early developments after World War II
Revolutionary changes in information processing and rapidly increasing information needs for public policy and private decision making have paved the way for an unprecedented expansion of information.
In the field of official statistics, national services were re-organized, comprehensive enumeration programmes developed, and data increasingly collected through surveys both of persons and establishments. Concomittantly, efforts towards standardizing of enumerations, concepts and methods gained new momentum (Flora/Kraus/Noll/ Rothenbacher 1994). Decisive incentives came from the newly founded U.N. (1946), the OEEC (1948), the Nordic Council (1953), and the European Communities (1952-57). Under the auspices of U.N.s Statistical Commision a complex machinery of statistical cooperation was established, aiming at the improvement of the quality and international comparability of national statistics and an expansion to include new subject matters, particularly in the field of social statistics. International reporting soon became a permanent task within all these organizations.
In the field of private statistics the growth and expansion of information gathering was even more rapid. Polls and market research soon became an important source for social research as well. Moreover, with some delay, academics increasingly could afford their own sample surveys (Rokkan 1966).
This rapid growth of information offered a rich potential for social research but also required proper institutional innovations within the academic community.
The evolution of Social Science service institutions
The response of the international academic community in fact was very intelligent and thorough. On the initiative of leading social scientists concerned with comparative research, in the early 60s UNESCO - within its commitment to internationalize social sciences - decided to concentrate its efforts on the generation of an academic infrastructure for comparative research (Rokkan 1968). The ISSC was in charge of the programme execution and, after a series of prominent international conferences, organized a "Round Table Conference on Comparative Research". The conference adopted 8 recommendations for the strengthening of the institutional foundations of comparative research addressing 4 arenas of action:
The initiative was extremely successful, not only in intellectual, but also in institutional terms. Gradually three types of service institutions evolved during the last three decades (Tanenbaum and Mochmann, 1994): data archives, documentation services and survey research centres. With the exception of the ICPSR archive in the U.S., institutions were, however, built up primarily in response to national needs. In addition to these nationally oriented institutions, the social sciences managed, however, to build up various institutions for international cooperation - both at the level of infrastructural services (such as IFDO, CESSDA, ICSSD, ECSSID) and of research steering and promotion (for example ISSC, ESSC, ICPSR, ECPR and ECSR). Last not least, various research institutes have meanwhile been established in Europe which are explicitely dedicated to cross-national research. In addition to these institutional developments, a number of comparative databases was created, including data guides and handbooks. Through international cooperation, several cross-national survey programmes could be established in fields usually not covered by official statistics. The International Social Survey programme and the World Value studies are prominent examples for the establishment of a cumulating database with genuine information on attitudes, values and beliefs. The World Data Handbook of Political and Social Indicators represents such a case in the field of aggregate data. Regarding techniques of data analysis, many of the service institutions actively promote the international diffusion of the advancements in methodology and, more recently, in various workshops and summer-schooles clearly give increasing place to comparative methods.
Official statistics and new social science service institutions
In the field of official statistics, the progress made since the days of the Round-Table Conference at Yale was tremendous as well, both with respect to data gathering and dissemination (Bjerve 1985).
In Europe, extraordinary improvements in quality and comparability could be achieved, particularly within the context of todays European Union. Definitions, classifications and statistical units have gradually become mandatory for EU Member States. Meanwhile, a unified European statistical system has been evolving. EFTA has recently agreed to implement EU standards for its Member Countries. After the fall of the Iron Curtian close statistical cooperation was established with the formerly communist countries in Eastern Europe. Statistical subject matters were greatly expanded all over Europe, in particular within the field of social statistics. Administrative records were increasingly replaced by surveys, and, consequently, micro-level data have became a central source. What is more, large-scale surveys, such as micro-censuses and Labour Force Surveys, have diffused over almost the whole of Europe (Rothenbacher, 1994), offering an unprecended potential for comparative social analysis.
In addition to these political incentives, progress in information technology has also changed access to official statistics. The archives of statistical offices have meanwhile been transformed from repositories into electronic archives. New electronic storage media allow the dissemination of large quantities of data. And the public access to on-line databases which increasingly include also detailed meta-information , offer a new dimension of access to official statistics.
However, the efficient use of this unprecedented wealth of official statistics for comparative research requires new types of skills and services. Infrastructural knowledge must be accumulated and duplication of efforts be avoided. This is obvious in the case of large micro-data sets, but it actually also applies to the comparative analysis of aggregate statistics.
Social science service institutions in Europe differ greatly with respect to how they have adapted to these new developments in official statistics. Only a few of them managed to include official statistics as a service task: the explosion of academic surveys exhaust available resources in many cases. The resulting gap has meanwhile been partially closed by institutions established outside the national service movement. The Luxembourg Income Study (LIS-working paper, no. 7, 1988), with its continuously expanding holdings of micro-level data from official income and family expenditure surveys combines both research and archival functions with comparative orientation. And the same model is being applied by the evolving Luxembourg Employment Study, again with great success. An old question is raised again: should services be organized by subjects (cross-national, specialized institutions) or along national needs?
The rise of networking facilities
During the last decade, fundamental changes in computing environments again occurred. Particularly the diffusion of desk-top computers and the rise of international networking facilities (INTERNET) have opened new possibilities for accessing and integrating decentralized sources of information. These developments will also have important impacts on existing service institutions. Data sets of individual scholars will increasingly be available directly via net, freeing some resources at institutions with a general obligation to archiving. This opens new possibilities, and improving the support of national services for cross-national analyses would be a good decision. Work on the virtual integration of informaton holdings of national institutions (i.e. linkage of national services through networking) is already under way. CESSDA members, for example, are already about to link their machine-readable catalogues (Alvheim 1995; ESCR Data Archive Bulletin, 58, Spring 1995). Other examples of virtual services are the British Social Science Information Gateway or the efforts of CIESIN to establish a virtual international data archive through searching INTERNET for appropriate sources. Meanwhile, official bodies are going to use INTERNET as well to serve their clients information needs. In the U.S., many statistical agencies disseminate data through ftp-servers. The European Commission has set up a WWW-server offering information on the EUs goals, institutions and policies, including direct access to public databases.
Compared to the 50s, the progress made towards building an infrastructure of comparative research is impresssive in all four areas the Round-Table-Conference was concerned with. However, despite this tremendous success, much remains to be done. In most European countries is was not possible to establish traditional services. Existing institutions are limited to the richer parts of Western Europe. In Eastern Europe there is only one such institution: TARKI in Hungary. Moreover, services vary not only in organizational form and research orientation, but also in terms of basic functions and tasks. Support for the use of official statistics is in many countries clearly underdeveloped. LIS and LES are an important step in the right direction, but they do not suffice. Similar institutions for micro-level data from censuses of populations and establishments - invaluable sources for comparative research - are missing. National differences in data protection and copyrighting still are major obstacles to the formation of proper service organizations. What we urgently need is a combined effort to open up access to such sources at the European level. Official statististics are indispensable for the comparative analysis of societies and polities.
Alvheim, A. (1994): CESSDA satser pa WWW; in NSD-brukermelding, no. 5. Bergen: NSD.
Bjerve, P.J. (1985): International Trends in Official Statistics. In A. Atkinson/S.Fienberg (eds.): A Celebration of Statistics. New York: Springer.
Flora, P. (1975): Quantitative historical sociology, Current Sociology, XXIII (2).
Flora/Kraus/Noll/Rothenbacher (Eds.) (1994): Social Statistics and Social Reporting in and for Europe. Bonn: IZ
Kraus, F. (1994): Official statistics and international comparative research: the evolution of statistics on Europe, access to information, and tasks for a future European science information system. IASSIST 1994 conference paper.
Macura/Cleland (1985): Reflections on the World Fertility Survey. In A. Atkinson/S. Fienberg (eds.): A Celebration of Statistics. New York: Springer.
Rokkan, S. (Ed.) (1966): Data Archives for the Social Sciences. Paris: Mouton.
__(1968): Comparative Research across Cultures and Nations. Paris:Mouton.
Rothenbacher, F. (1994): Statistical sources for comparative European social research. In: International Social Science Journal (ISSJ), 142.
Tanenbaum, E. & E. Mochmann (1994): Integrating the European Database: Infrastructure services and the need for integration. In: ISSJ, 142.
CESSDA Council of European Social Science Data Archives
Dipl. Vw. Franz Kraus
University of Mannheim, MZES-EURODATA, D-68131 Mannheim
Phone: (+49) (0)621 - 292 1794
EURODATA Newsletter No.1, p.10-12