It surveys how this term is defined in the published literature — what the sources agree on, where they diverge and what remains contested — and carries its own numbered bibliography. It does not describe what we run. Programme policy is on the core pages.
Our own use of the word is often narrower than general usage; the short definition and the boundary are in the glossary, under Entrepreneurship. The review date below matters, because the literature moves.
A definitional review. Citations follow IEEE style; see References.
Abstract
Entrepreneurship education (EE) covers a wide range of interventions united less by content than by an aspiration: that learners become more capable of, or more inclined toward, creating ventures. The field’s central measurement problem is that its dominant outcome variable — entrepreneurial intention — is a weak proxy for the behaviour it stands in for. Where rigorous designs have been used, average effects are small, strongly dependent on programme duration, and in at least two well-identified studies negative on the outcomes the programmes were meant to improve. This article sets out the definition, the standard typology, what the evidence reports, and the inferences the literature does not support.
I. Definition
There is no settled definition. What the field agrees on is a typology of purpose, usually given as three prepositions:
| Type | Teaches | Typical outcome measured |
|---|---|---|
| About entrepreneurship | entrepreneurship as subject matter — what ventures are, how they work | knowledge |
| For entrepreneurship | preparation to start a venture — planning, finance, pitching | intention, self-efficacy |
| Through entrepreneurship | learning other things by running a venture-like activity | non-cognitive skills, transferable competences |
The distinction matters because the three are evaluated against different outcomes, and results from one are routinely quoted in support of another. A “through” programme that improves persistence is not evidence that an “about” programme transmits business knowledge — and, as §III shows, the best-identified study on the question found precisely that dissociation.
Competence frameworks attempt to stabilise the field’s vocabulary; the most widely adopted in Europe is EntreComp, which decomposes entrepreneurial competence into areas and progression levels [7]. Practice-oriented syntheses serve a similar function [8]. These are specifications, not findings — they say what could be taught, not what teaching achieves.
II. What the outcome variable actually measures
Most EE studies do not measure venture creation. They measure entrepreneurial intention — a self-reported disposition, usually captured immediately after the intervention.
This is the field’s structural weakness, and it has three parts:
- Intention is a weak predictor of behaviour, and the gap is widest for exactly the population EE targets: young people with limited capital and few years in which the intention could have been acted on.
- Measurement is proximate to the intervention. Post-programme enthusiasm is measurable; whether it survives is usually not measured at all.
- Higher intention is treated as unambiguously good, which §III shows it is not — an intervention that raises intention without raising capability may be steering learners toward a worse outcome than the one they would have chosen.
III. What the evidence reports
A. Average effects are small
A meta-analysis of entrepreneurship education’s effect on intention and self-efficacy reports an intervention effect of d ≈ 0.215 against a control effect of d ≈ 0.008 [3]. The contrast is real but the magnitude is modest, and it is measured on the proxy variable described above.
In the adjacent domain of school financial education, a meta-analysis reports effects on financial behaviour of approximately 0.07 SD [4] — smaller again, and behaviour is the harder and more meaningful outcome.
B. Dosage is decisive
The clearest practical finding in the literature concerns duration. In a large South Korean study, a single three-hour workshop produced no significant gains on any measured entrepreneurial variable, while an extended programme of ten two-hour sessions produced significant gains (p < .001) across opportunity discovery and exploitation, entrepreneurial orientation, creativity, social problem-solving and intention [2].
The design is quasi-experimental rather than randomised, and single-country, so the magnitude should be held loosely. The direction is nonetheless consistent with the wider pattern: one-off interventions test null. Short workshops are the most commonly delivered form of EE and the form least supported by evidence.
C. Mindset moves; knowledge does not
The most on-point randomised trial ran a hands-on simulated-venture programme with approximately 2,751 Dutch children aged 11–12 [1]. It found:
- robust gains on non-cognitive entrepreneurial skills — persistence, creativity, forward-looking and analytical behaviour;
- no gain on business knowledge — profit and loss, shares, production;
- a negative effect on entrepreneurial intentions.
All three results matter, and the third is usually dropped when this study is cited. The dissociation between skills and knowledge is the strongest available evidence for the through type over the about type at pre-adolescent ages.
D. A well-identified programme produced negative downstream effects
A randomised trial across approximately 200 schools and 3,000 students in Rwanda (2016–2021) found gains that faded, and negative downstream income and employment effects among marginal entrants — learners who were induced into venturing by the programme rather than self-selecting into it [5].
Two caveats belong with the finding: the full-sample income effect is near-null, so the negative result is a marginal-subgroup finding rather than an average one; and the source is a working paper. Both qualify the result without removing it. The mechanism is the important part: an intervention that successfully increases venture entry can make some entrants worse off, because the marginal entrant is by construction the one least suited to the path.
IV. Scope of adoption
EE is widely delivered in school systems, in higher education and through development programming, and in low- and middle-income countries it is frequently bundled with financial education and delivered over a full school year. Programme dosage in these designs is substantially higher than the workshop format that dominates elsewhere, which §III.B suggests is the relevant difference.
Adoption has consistently outrun evaluation. Systematic reviews of EE in developing countries note that the volume of provision is not matched by studies capable of identifying effects [9].
V. What the literature does not establish
Three inferences appear frequently and are not supported.
That earlier is better. The finding that experiential EE moves mindset at ages 11–12 [1] does not license the comparative claim that non-cognitive skills are best developed early. That inference requires a cross-age comparison the study does not contain, and it has been explicitly rejected on re-examination. The defensible statement is bounded: at ages 11–12, experiential entrepreneurship education moves mindset, not knowledge.
That EE raises incomes. No study reviewed here establishes it, and [5] reports the opposite for a subgroup.
That intention gains are outcomes. They are proxies, and [1] found intention moving downward while skills moved upward — which is coherent if better-informed learners revise an unrealistic expectation, and which makes intention unusable as a headline success measure.
VI. Limitations
The evidence base is thinner than the field’s volume of activity suggests, and it is skewed in three ways.
Geographic skew. The rigorous studies cited here are Dutch [1], South Korean [2] and Rwandan [5]. Meta-analytic samples are dominated by high-income settings. For most countries, including most of South Asia, programme-level impact evidence is essentially absent — which means EE is often justified locally by evidence generated somewhere structurally different.
Design skew. Much of the literature is pre-post without a control group, and outcome measures are self-reported instruments administered by the people delivering the programme.
Outcome skew. The field measures what is cheap to measure. Intention is captured constantly; venture survival, income and employment rarely, and almost never far enough after the intervention for fade-out to be visible — which is exactly the interval in which [5] found its effects reverse.
A further limitation applies to this review: its quantities are drawn from a verified secondary compilation rather than from the primary reports, and no claim above should be extended beyond the figures as stated.
References
Reading note. Bibliographic details and reported quantities below are taken from a compilation whose principal claims were independently re-checked. The primary papers were not consulted directly for this article. Confidence markers reflect the design of each study, not the depth of reading here.
[1] J. Rosendahl Huber, R. Sloof, and M. Van Praag, “The Effect of Early Entrepreneurship Education: Evidence from a Field Experiment,” European Economic Review, 2014. Randomised controlled trial, n ≈ 2,751, ages 11–12, Netherlands. ⚠️ Full text not consulted for this article.
[2] K. Kim, S. Kim, Y. Lee, and H. Joung, “The Effect of Youth Entrepreneurship Education Programs,” SAGE Open, 2020, doi: 10.1177/2158244020956976. Quasi-experimental pre-post, n = 1,118 and n = 816, South Korea. ⚠️ Full text not consulted for this article.
[3] A. Martínez-Gregorio, L. Badenes-Ribera, and A. Oliver, “Effect of entrepreneurship education on entrepreneurship intention and related outcomes in educational contexts: a meta-analysis,” The International Journal of Management Education, vol. 19, no. 3, p. 100545, 2021. ⚠️ Full text not consulted for this article.
[4] T. Kaiser and L. Menkhoff, “Financial education in schools: A meta-analysis of experimental studies,” Economics of Education Review, vol. 78, 2020. ⚠️ Full text not consulted for this article.
[5] M. P. Blimpo and T. Pugatch, “Unintended Consequences of Youth Entrepreneurship Programs: Experimental Evidence from Rwanda,” GLO Discussion Paper 1332 / IZA Discussion Paper 16489; also Journal of Development Economics. AEA RCT Registry 1030. ≈200 schools, ≈3,000 students, 2016–2021. ⚠️ Working paper; full text not consulted for this article.
[6] Arrowhead Business Group randomised trial, Native American youth aged 13–16, PMC7177681. ⚠️ Full text not consulted for this article.
[7] M. Bacigalupo, P. Kampylis, Y. Punie, and G. Van den Brande, EntreComp: The Entrepreneurship Competence Framework, EUR 27939 EN. Luxembourg: Publications Office of the European Union, 2016, JRC101581.
[8] OECD, Entrepreneurial Education in Practice (E360), Part 1. Paris, France: OECD.
[9] “What is the impact of entrepreneurship education programs in developing countries? A systematic review and research agenda,” Allied Business Academies. ⚠️ Existence-verified only; not read.