Remote Access – Eine Welt ohne Mikrodaten??
Use of microdata is severely hampered in many areas of research. This is in particular true for data from statistical offices. One way to circumvent this problem is to anonymize the data such that both confidentiality is guaranteed and informational content of the data is not too much distorted by the anonymization procedure. However many researchers prefer the use of 'original' data. Therefore in recent years remote access/execution ('Fernrechnen') has become quite popular where the original micro data are used in the statistical analysis but are not available to the researchers. Clearly, this alternative takes more time since program files have to be sent to the statistical office. However, the euphoria for this approach has cooled down a bit since it has become apparent that here also problems of condentiality exist. Most obvious is the fact that residuals cannot be provided. See, for example, Gomatam et al. (2005). However, there are very different kinds of 'disclosures' which are discussed in the paper. The paper also draws attention to the use of saturated models which bear the risk of reproducing confidential tabular data. Analysis of variance is the relevant tool in reproducing magnitude tables whereas the corresponding micro-econometric models can be used to reproduce frequency tables: Logit models give the results in case of a nominal variable and Poisson regression is the approach in case of count data.We also shortly discuss possible disclosure risk in the standard multivariate procedures (factor analysis, principal components, cluster analysis and multidimensional scaling). It is clear from the many examples given in the paper that the remote access/execution option will ask for a large amount of statistical expertise in the statistical office in order to check for disclosure risk. Additionally, there will be a tendency not to provide statistical results to the researcher if critical variables such as region or sector are demanded as regressors in the program file. Perhaps a much cruder classification of regions and sectors will be allowed which in a way is the situation used in providing anonymized data.
|Date of creation:||Jun 2010|
|Date of revision:|
|Contact details of provider:|| Postal: |
Phone: (+49) 7071 98 96 -0
Fax: (+49) 7071 98 96 -99
Web page: http://www.iaw.edu/
More information through EDIRC
References listed on IDEAS
Please report citation or reference errors to , or , if you are the registered author of the cited work, log in to your RePEc Author Service profile, click on "citations" and make appropriate adjustments.:
- Roderick McDonald & E. Burr, 1967. "A comparison of four methods of constructing factor scores," Psychometrika, Springer, vol. 32(4), pages 381-401, December.
When requesting a correction, please mention this item's handle: RePEc:iaw:iawdip:66. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Rolf Kleimann)
If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.
If references are entirely missing, you can add them using this form.
If the full references list an item that is present in RePEc, but the system did not link to it, you can help with this form.
If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your profile, as there may be some citations waiting for confirmation.
Please note that corrections may take a couple of weeks to filter through the various RePEc services.