Enhanced Ascription

Julian Bairm
Mediamark Research Inc.
and
Prof. Martin K. Frankel
Bernard M. Baruch College

In this essay, we would like to address a number of issues about ascription. Firstly, I am going to discuss the rationale for basic ascription and provide some examples of how this procedure works. I will then speak specifically about our efforts to refine the procedure through an enhanced ascription process.

Ascription is not a mysterious, arcane procedure. It is a process where survey respondents who have not answered all survey questions are assigned or given responses to these 'missing items.' Our study is conducted in two phases. The first part, a personal interview, measures magazine audiences, other media behavior and demographic characteristics. The second phase, a self-administered product information booklet or PIB, is completed and returned by slightly more than 60% of the first interview respondents. We are then missing the product behavior answers of some 40% of our respondents. To compensate for the missing data, MRI's computer program finds the best match between a product booklet non-respondent and a product booklet respondent. We then copy the responder's product booklet record to the non-responder. MRI' s matching process employs demographic as well as behavioral variables, since these characteristics are predictive of the answers to be ascribed. MRI limits the number of times any respondent can donate his/her record to non-respondents.