Assessing the stability of graduates’ entrepreneurial intention and exploring its predictive capacity

Purpose: This is one of the first long‐term studies on entrepreneurial intentions (EIs) and their role in predicting start‐up. In entrepreneurship, the usefulness of intention models depends crucially on their stability over time. The purpose of this paper is to explore this issue in a cohort of university graduates. Design/methodology/approach: Using a structural equation model (SEM), it addresses three research questions: whether EI and perceptions are stable over time; whether the pattern of relationships of the variables in the model remains stable over a three‐year period; and whether intention accurately predicts entry into self‐employment. A two‐wave study was conducted with respondents being final year students at time 1 (T1), and part of the working/active adult population at time 2 (T2). Findings: The results suggest that construct scores and the pattern of relationships between the variables are notably stable over the three‐year period. The only partial exception is that of perceived behavioral control. Finally, EIs are found to significantly explain actual start‐up behavior. Implications are derived for future research and entrepreneurship education and policy. Research limitations/implications: The generalizability of the results cannot be claimed until these findings are replicated by additional samples from different segments of the population and in different contexts. In particular, the number of new entrepreneurs at T2 is small and results in this respect are to be taken with caution. Originality/value: The value of this paper resides in its clarifying the conditions under which EIs and their antecedents remain stable over time, thus enabling actual start‐up behavior.