Transfer of Variables between Different Data Sets, or Taking “Previous Research” Seriously

Bojan Todosijević*

*Corresponding author for this work

    Research output: Contribution to journalReview articleAcademicpeer-review

    4 Citations (Scopus)


    Given two methodologically similar surveys, a question not asked in one survey could be seen as a special case of the missing data problem. Hence, the transfer of data across data sets (“statistical matching” or “data fusion”) could be achieved applying the procedures for Bayesian multiple imputation of missing values. To tackle the problem of conditional independence, which this approach creates, a simulated data set could serve as the “third data set” that conveys information about the relationship between variables not commonly observed. This paper presents a model for transferring data between different data sets based on multiple imputation (MI) approach. The results show that statistical matching based on MI principles can be a useful research tool. The entire enterprise is interpreted in the sense of taking the “previous research” into account seriously.

    Original languageEnglish
    Pages (from-to)20-39
    Number of pages20
    JournalBMS Bulletin of Sociological Methodology/ Bulletin de Methodologie Sociologique
    Issue number1
    Publication statusPublished - 2012


    • Dutch election study
    • Multiple imputation
    • Statistical matching

    Cite this