IMPACT OF EFFECTUAL PROPENSITY ON ENTREPRENEURIAL INTENTION

The study addresses how effectual propensity (EP)—an individual’s tendency toward effectual logic—affects entrepreneurial intention (EI) among potential entrepreneurs who have not yet created firms. Entrepreneurship is vital for economic development and is especially relevant in tourism, where new ventures drive destinations and employment. The Theory of Planned Behavior (TPB) posits EI is determined by personal attitude (PA), subjective norms (SN), and perceived behavioral control (PBC). Prior research on effectuation has focused on established entrepreneurs, leaving a gap on how effectual tendencies shape intentions before venture creation. This paper tests the direct and indirect effects of EP on EI within an extended TPB model using tourism students from the Universities of Cadiz and Seville. The objective is to evaluate EP as an antecedent of EI and examine mediating roles of PA and PBC.

Effectuation: Under uncertainty, entrepreneurs may follow causal (predictive, goal-driven) or effectual (means-driven, adaptive, control-oriented) logics. Effectuation emphasizes five principles: start with means, partnerships (co-creation), affordable loss, leveraging contingencies, and control (Sarasvathy, 2001; Dew et al., 2009). While extensively studied post-startup, few works examine effectual orientation in potential entrepreneurs. Effectual propensity (EP) captures pre-startup inclination toward effectual logic (Martín-Navarro et al., 2021). Theory of Planned Behavior (TPB): TPB explains intentions via personal attitude (PA), subjective norms (SN), and perceived behavioral control (PBC) (Ajzen, 1991). In entrepreneurship, PA reflects desirability of being an entrepreneur, SN the perceived social pressure/approval, and PBC perceived capability/control. Research model and hypotheses: The authors integrate EP into TPB, proposing nine hypotheses: H1 EP→PA (+); H2 EP→PBC (+); H3 EP→EI (+); H4 SN→PA (+); H5 SN→PBC (+); H6 SN→EI (+); H7 PA→PBC (+); H8 PA→EI (+); H9 PBC→EI (+). Rationale links effectual principles (means, partnership, affordable loss, contingencies, control) to TPB constructs and EI, supported by prior evidence on skills, risk propensity, cooperation, opportunity recognition, and internal locus of control.

Design: Cross-sectional survey of tourism degree students from the University of Seville and the University of Cadiz, Spain. Non-probabilistic sampling; self-administered questionnaire. Sample: 464 valid responses. University: 56.71% Seville, 43.29% Cadiz. Gender: 71.02% women, 28.98% men. Age: 56.04% (18–21), 37.58% (22–25), remainder >25. Nationality: 97.4% Spanish. Family entrepreneurship: 31.82% have entrepreneurial parents. Work experience: 34.20% have worked/are working; 46.84% never worked; remainder mixed/self-employed. Measures: 7-point Likert scales. EP measured as a higher-order construct with five subdimensions adapted from Martín-Navarro et al. (2021): means orientation (3 items), association orientation (4), affordable loss (3), contingency (4), control orientation (4). TPB constructs from the Entrepreneurial Intention Questionnaire (Liñán et al., 2011): PA (5 items), PBC (5), SN (4), EI (6). Items randomized to reduce response bias. Analysis: PLS-SEM using SmartPLS 3. Assessment followed standard procedures: indicator reliability (loadings >0.707), construct reliability (rho_A >0.7; composite reliability >0.8), convergent validity (AVE >0.5), discriminant validity via HTMT (<0.85). For the EP second-order construct, affordable loss and means orientation dimensions were removed due to insufficient loadings (<0.707). Structural model assessed via bootstrapping with 5,000 resamples, reporting path coefficients, t-statistics, p-values, and confidence intervals. Model fit via SRMR; explanatory power via R² and adjusted R². Indirect effects assessed. Importance-performance matrix analysis (IPMA) conducted for EI. Predictive performance evaluated with PLSpredict (RMSE, MAE, Q²predict).

Measurement model: All retained constructs show adequate reliability and validity: rho_A >0.7; composite reliability generally >0.9; AVE ≥0.554. HTMT values below 0.85 indicate discriminant validity. In the EP higher-order construct, the affordable loss and means orientation dimensions were eliminated for low loadings. Structural model (path coefficients, significance): Of nine hypotheses, seven supported.

• H1 EP→PA: β=0.279, t=6.489, p<0.001 (supported)
• H2 EP→PBC: β=0.175, t=3.900, p<0.001 (supported)
• H3 EP→EI: β=0.077, t=2.407, p=0.016 (supported)
• H4 SN→PA: β=0.277, t=5.297, p<0.001 (supported)
• H5 SN→PBC: β=0.084, t=1.861, p=0.063 (not supported)
• H6 SN→EI: β=0.016, t=0.534, p=0.593 (not supported)
• H7 PA→PBC: β=0.425, t=10.345, p<0.001 (supported)
• H8 PA→EI: β=0.587, t=18.552, p<0.001 (supported)
• H9 PBC→EI: β=0.264, t=7.382, p<0.001 (supported) Explanatory power: R²(EI)=0.640 (moderate), R²(PA)=0.206 (weak), R²(PBC)=0.310 (weak). SRMR=0.052 indicates good model fit. Indirect effects (selected): EP→PA→EI: 0.164 (p<0.001); EP→PBC→EI: 0.046 (p=0.001); EP→PA→PBC→EI: 0.031 (p<0.001). SN shows significant indirect effects via PA on EI and PBC (e.g., SN→PA→EI: 0.162, p<0.001; SN→PA→PBC: 0.118, p<0.001), despite non-significant direct effects on PBC and EI. IPMA: PA exhibits the strongest contribution (importance) to EI, followed by EP. PBC contributes least to EI’s total effect. SN shows high performance but relatively low importance/effect on EI. Predictive validity: PLSpredict indicates positive Q²predict values (Attitude 0.196; PBC 0.156; EI 0.195) with acceptable RMSE/MAE, supporting out-of-sample predictability.

Integrating EP into the TPB framework improves understanding of antecedents to entrepreneurial intention among potential entrepreneurs in tourism. Findings show EP is a significant determinant of EI both directly and indirectly through PA and PBC, confirming that effectual tendencies—working with available means, partnering, leveraging contingencies, and a control orientation—shape favorable attitudes and perceived capability that translate into intentions. PA and PBC remain key predictors of EI as per TPB, with PA having the strongest effect in this sample. SN does not directly influence EI or PBC, but it indirectly affects both through PA, suggesting social approval primarily operates by shaping attitudes rather than perceived control or intentions directly. The model demonstrates good fit, substantial explanatory power for EI, and predictive validity, supporting its relevance for predicting entrepreneurial intentions in tourism education contexts.

The study contributes by introducing effectual propensity (EP) as a novel antecedent of entrepreneurial intention (EI) within an extended TPB model. Using a sample of Spanish tourism students, EP significantly influences EI directly and indirectly via personal attitude (PA) and perceived behavioral control (PBC). PA and PBC also exert strong direct effects on EI, while subjective norms (SN) influence EI indirectly through PA. The model shows good fit and predictive validity, indicating practical utility in tourism entrepreneurship contexts. Implications: For higher education, curricula can be tailored to strengthen students’ effectual orientation and entrepreneurial competencies. Policymakers can target early-stage entrepreneurial development with interventions that cultivate effectual thinking and supportive attitudes. Firms can assess EP among employees/managers/partners to inform entrepreneurial roles and collaborations. Future research: Replicate across diverse cultures, institutions, and demographic profiles; test EP as a moderator in intention models; refine measurement of EP’s subdimensions (including affordable loss and means orientation) and examine their roles across contexts.

Findings are based on a single cross-sectional study of tourism students from two geographically proximate Spanish universities, limiting generalizability. Non-probabilistic sampling and self-reported measures may introduce bias. Some EP subdimensions (affordable loss and means orientation) were removed due to low loadings, suggesting measurement refinement is needed. Future studies should use varied samples across countries and educational/cultural contexts and consider longitudinal designs.