Achieving Acceptable Response Rates in Behavioral Studies of Big-Four Accounting Firms

• In Event: Accounting History and Research

Abstract

Response rates to research instruments sent to accounting professionals have plummeted in recent decades. The cost of collecting a usable sample is high, with nonresponse bias a threat. Postal solicitation methods may be underestimated today, and may be more effective than email. In a pilot study targeting Big-Four CPA firm members, we use two of the most proven methods in Experiment 1: a small monetary gift and a postage stamp (vs. business reply) on the return envelope. The monetary gift shows a statistically significant and fairly large effect, but the stamped envelope is ineffective. As the lowest cost/response in the first experiment is about $22, in Experiment 2 we send an honorarium check of $20, sent with the request and contingent on completing the brief questionnaire, but the method is not cost-effective. Experiment 3 validates the results of Experiments 1 and 2 and provides improved confidence intervals for expected response rates. Experiment 4 compares the effectiveness of an alternative business gift to the effectiveness of $5 cash and clearly demonstrates the ineffectiveness of the alternative gift. Implications, limitations and suggestions for further research are discussed.

Authors

• Charles D. Bailey, James Madison University

• Thomas J Phillips, Louisiana State University