One Construct or Two? Knowledge Management, University Entrepreneurship, And A Brief Validated Measure

Abstract

Research on higher education in Latin America increasingly combines questionnaire measures of knowledge management and of student entrepreneurship in structural models that posit influence relationships between them, yet rarely verifies a precondition of such models: that the constructs be empirically distinguishable. This study assessed the dimensional structure, reliability, and convergent and discriminant validity of an eighty-item, seven-dimension instrument designed to measure both variables, using responses from 1,092 students of private universities in Northern Lima, Peru, most of whom were employed and had prior entrepreneurial experience. Competing factor models were compared at the item level, and reliability, average variance extracted, and heterotrait-monotrait ratios were estimated. The hypothesized seven-factor structure did not substantively outperform a single-factor model; all twenty-one latent correlations among dimensions ranged from 0.86 to 0.97; and six of seven dimensions failed the convergent validity criterion despite high composite reliability. The instrument therefore measures essentially one general dimension, here termed knowledge-based entrepreneurial capacity. A fourteen-item short form preserving the content coverage of the original instrument was derived and validated, showing acceptable fit, satisfactory convergent validity, and high reliability. The findings caution against interpreting structural coefficients between closely related self-report measures as substantive effects, and provide researchers and university managers in the region with a parsimonious, freely reusable measurement tool and a reproducible validation protocol.