Does Entrepreneurship Education Really Work? A Review and Methodological Critique of the Empirical Literature on the Effects of University‐Based Entrepreneurship Education

Abstract

Does entrepreneurship education (E‐ed) really work to create business enterprise? We conducted a comprehensive review and methodological critique of the empirical research on the outcomes of university‐based E‐ed. We identified every empirical study conducted over the past decade, and found 12 that minimally met our methodologically “robust” (Storey Steps 4–6) standard. Our systematic critique of the studies' research methods found a variety of methodological weaknesses, undermining confidence in the belief that E‐ed can produce entrepreneurship. The implications for both practice and policy are discussed, and recommendations are made for conducting future E‐ed outcome research.

Notes

1. Keywords “Entrepreneurship Education,” “university,” “high‐tech entrepreneurship education,” “outcomes, impacts, and evaluation” were used alone and in combination. In addition, leading entrepreneurship scholarship repository websites were searched, and leading peer‐reviewed entrepreneurship journals in the field were manually searched by title. (Journals searched included: Entrepreneurship Theory and Practice, Journal of Small Business Management, Journal of Business Venturing, Academy of Management Learning and Education, Technovation, Journal of International Entrepreneurship, International Entrepreneurship Management Journal, Journal of Entrepreneurship, Journal of Entrepreneurship Education, Entrepreneurship and Regional Development, and several others.)

2. Unfortunately, no Step 6 experimental studies (with random assignment to conditions, or controls for self‐selection) were found.

3. However, they do not report statistical tests on these comparisons.

4. Concerns can also be raised by the use of change scores and use of separate pre‐posttests for the E‐ed and comparison groups.

5. Many of the more well‐known, widely cited studies could have been included in this analysis but were not, either because they did not have anything further to add or because of design flaws. In the former case, for example, Weber (2012) conducted a controlled study and found no effects of E‐ed on Business students’ E‐Intentions. The study was unique in that it avoided self‐selection problems (the E‐ed course was mandatory), but control group noncomparability essentially produced a one‐group pre‐post design. His statistical compensation efforts lead to additional problems (excessive t‐tests, small cell sizes). Lange et al. (2011) conducted an e‐mail survey of all the alumni of Babson College; and, Blackford, Sebora, and Whitehill (2008) conducted a similar e‐mail survey of all E‐ed graduates at a major public Midwestern university. Both studies employed a one‐group posttest‐only design methodology, however. Another well‐known study coauthored by NYU and Summit Consulting, LLC (2009) employed a noncomparison‐group, posttest‐only design, with mixed group comparisons (undergraduates and MBAs), and significant response bias problems, according to the authors. Dutta et al., 2011 attempted to replicate Charney et al.'s (reviewed here) methodology, but employed a one‐group posttest‐only, no‐comparison‐group design. Although one‐group studies are helpful for hypothesis building, they are insufficiently rigorous in providing the confidence that E‐ed does in fact produce entrepreneurship—the causal result we were looking for.

6. Unfortunately, since Chen, Greene, and Crick (1998) involved only a pretest comparison, 11 studies truly met this criterion. Of these almost all had comparison group problems.

7. Shadish et al. label this design the “untreated control group design with dependent pretest and posttest samples.”

Additional information

Notes on contributors

Elaine C. Rideout

Elaine C. Rideout is Adjunct Professor, Entrepreneurship Initiative, North Carolina State University.

Denis O. Gray

Denis O. Gray is Alumni Distinguished Graduate Professor, Department of Psychology, North Carolina State University.