Consideration of the incomplete block design for policy-capturing research

This article introduces and urges consideration of the incomplete block design, for policy-capturing research. The authors compare incomplete block designs with full factorial and fractional factorial designs along several dimensions, including survey length, analysis limitations, and impact on study participants. An initial empirical investigation of an incomplete block design and a full factorial design is performed in a policy-capturing study of 183 job seekers, As predicted, the two designs generated equivalent results on the main effects and most interactions; however, the designs yielded divergent results on some interactions. Study participants who received the incomplete block design were more positive about the survey length and experienced less stress and exhaustion than those receiving the full design. The authors discuss how methodologists might examine and compare policy-capturing designs in future research.