Databases and use
An initial examination showed that BIBB has generated more than 200 data sets since 2000 and that new databases are additionally being generated in the course of current research projects. Not all of these databases are of equal interest for secondary analyses. For this reason, the BIBB-FDZ first prepared those databases for which there is proven demand from external researchers. Care was also taken to cover as much of the entire range of topics so that documentation systems develop which make allowances for the heterogeneousness of BIBB's microdata without being arbitrary or too idiosyncratic.
It is possible to classify data on vocational education and training at least roughly (see diagram) using five stations in life (general schooling, first threshold, vocational training, second threshold, working life). The introduction of a further differentiation - between company-level data and individual-level data - produces ten 'cells'. During the first phase of its work, the BIBB-FDZ will offer at least one (processed) data set for each of these cells.
Structure of the databases at the BIBB-FDZ
.jpg "Structure of the databases at the BIBB-FDZ")
Source: www.bibb.de/de/50124.htm
The above-outlined work that went into setting up the BIBB-FDZ is offset, at least by national standards, by a considerable yield. Right from the start, the BIBB-FDZ sought to record and document requests/enquiries received from researchers with the aim of obtaining the most detailed picture possible of the different interests in using this data and the levels of its use. Analyses from the FDZ request/enquiry statistics were used for the following account of the yield from the BIBB-FDZ. The analyses cover the period from April 2008 to October 2009 (approximately 18 months).
During this period, BIBB-FDZ answered 278 requests/enquiries regarding subjects covered by BIBB microdata - in addition to conducting the previously-mentioned work to set up the BIBB-FDZ. Of these 278 requests, 234 were external requests/enquiries (some 84%) and 44 were internal requests/enquiries. These requests covered the entire range of FDZ services and products. As would be expected, most of the data-related requests/enquiries pertained to accessing data, the terms of use and the like. However, the BIBB-FDZ also received a significant number of methodology-related requests/enquiries (such as regarding certain regression) and requests/enquiries regarding specific topics/content (such as the effects that continuing training activities have on whether the individual finds employment). The range of research topics extends from, for example, studies on the potential for creativity in the workplace all the way to potential earnings and opportunities to earn income in the wake of technical and organisational changes in the work environment.
Looking at the individual BIBB data sets, the 2006 BIBB/ BAuA survey and preceding surveys (BIBB/IAB surveys) currently account for the greatest number of requests/enquiries by far. Approximately 43 per cent of all external requests/enquiries pertain to the use of this data set or lead to its use. Another eight per cent revolve around the BIBB/IAB surveys. About 27 per cent of all requests/enquiries do not concern a specific data set and thus can be classified entirely as concerning content-related or methodological consultation services. At eight per cent of all requests/enquiries, there is still a relatively large number of requests/enquiries regarding the data set pertaining to the "costs and benefits of vocational education and training" from the year 2001. It must be remembered when interpreting these user statistics that the data sets mentioned here were the first to be processed at or for the BIBB-FDZ. This does not necessarily mean however that the demand for other data sets that are not mentioned here is less important. In one or two years' time, a more complete picture will be available than can currently be offered during the current set-up phase.
The processing time for individual requests/enquiries received by the BIBB-FDZ averaged approximately half an hour - albeit with a large standard deviation. In other words, there was an above-average number of shorter processing times and an above-average number of longer processing times. The FDZ request/enquiry statistics do not however include requests/enquiries with processing times of less than ten minutes. The number of requests/enquiries was not the identical with the number of licence agreements. The BIBB-FDZ signed a total of 68 licence agreements with more than 130 users for the different types of data access. Even when the vast majority of the licence agreements was also for the BIBB-BAuA survey from 2006, it must be noted that this figure is remarkably large given the period of the BIBB-FDZ's operation (18 months) which was covered by the analysis. Based on this, the question of whether there is an actual need for an FDZ at BIBB can be answered with an unconditional 'yes'. This is also made clear by a glance at the institutions that use BIBB data: These institutions include not only numerous German universities and most well-known social and economic research institutes, but also renowned foreign universities such as the University of Leuven (Belgium) and the elite American universities Princeton and Stanford.
.JPG){kind=logo}
