← Search

Nonresponse bias in survey‐based entrepreneurship research: A review, investigation, and recommendations

David J. Scheaf1; Andrew C. Loignon2; Justin W. Webb3; Eric D. Heggestad4

1 Hankamer School of Business Baylor University Waco Texas USA · 2 Center for Creative Leadership Greensboro North Carolina USA · 3 Belk College of Business, University of North Carolina at Charlotte Charlotte North Carolina USA · 4 Department of Psychological Science University of North Carolina at Charlotte Charlotte North Carolina USA

Strategic Entrepreneurship Journal 2023

Abstract Research Summary Entrepreneurship researchers commonly use survey‐based research designs. However, surveying entrepreneurs poses unique challenges. A principal concern for survey‐based research is nonresponse bias, which occurs when survey respondents systematically differ from those who were sampled but did not participate in the study. To address this concern, we conducted a systematic review of the literature to determine what practices are currently being used to address nonresponse bias (142 articles, 180 surveying efforts) and conducted multiple studies to determine the extent to which nonresponse bias can affect statistical results. Based on these efforts, we present a series of techniques and a checklist that entrepreneurship scholars and reviewers and editors can consider to mitigate the risk of nonresponse bias prior to, and following, their data collection efforts. Managerial Summary In an average survey of entrepreneurs, approximately 40% of the people contacted will respond. This raises an important question: Can we make conclusions about the entrepreneurs who did not respond? Answering this question requires one to consider nonresponse bias (i.e., if respondents differ from nonrespondents). We conducted a systematic review of survey‐based studies in entrepreneurship and conducted three field studies. Our results show that nonresponse bias can manifest in different ways across studies, but it is rarely discussed in the literature. We also show that response rates are poor proxies of nonresponse bias. Based on these efforts, we present several techniques for scholars, reviewers, and editors to consider in the hopes of mitigating the risk of nonresponse bias in their studies.

DOI
10.1002/sej.1453
Volume
17 (2)
Pages
291-321
Language
en
Export
BibTeX
Sources
crossref