Empirical Evaluation of Alternative Approaches

Abstract

In this chapter we consider a typical European statistical matching task where multiple categorical, continuous, and so-called semicontinuous variables concerning media and television behavior have been recorded in separate surveys in addition to the usual demographic and socioeconomic information. Our goal here is less to analyze or describe the relationship among variables in a meaningful way but to find out whether the procedures for imputing missing values preserve important features of marginal and joint distributions. We want to investigate the performance of the proposed alternative matching techniques discussed in Chapter 4 when applied to real data sets which typically do not follow simplifying assumptions. The validity of a matching technique is measured according to the four levels we introduced in Chapter 2. The data are provided from the television behavior panel run by the largest German market research company.¹ This data set is matched regularly with the purchasing behavior panel; the procedure actually applied by the GfK is described in section 3.3.5.

In a situation where social scientists are so hungrily looking for increasingly rich data bases, stastistical matching is dangerously attractive procedure for creating files containing the logical union of the variables involved in either of the component files [...] I would like to see a good deal of empirical evaluation of the validity of such joint distributions before I would suggest removing the label from this procedure “DANGEROUS — USE WITH CAUTION”. I.P. Fellegi (1977).