Comparative Analysis of Within-Household Respondent Selection Techniques

A probability sample of households frequently must be translatedinto a sample of persons representing characteristics of theadult population. Researchers have developed at least 14 within-householdrespondent selection procedures or their variants to accomplishthis task. Although probability methods are preferable, theyoften increase nonresponse. Quasi-probability and nonprobabilitytechniques have been devised to increase cooperation and decreasecosts, although they sacrifice the advantages of randomness.The purpose of this paper is to analyze the results of 16 studiesthat have compared within-household respondent selection procedures,with respect to demographic representativeness, cooperationor response rates, and costs. Although there is an increasingneed to hold down costs yet maintain quality and accuracy asresponse rates are falling, little systematic, accessible evidenceexists to guide choice of respondent selection method.