This article evaluates the discriminant validity of a model using both the Fornell and Larcker criterion and the Heterotrait-Monotrait (HTMT) ratio of correlations method. The study uses data from a previous research on value-based excellence in higher education institutions (HEIs) in Malaysia, involving 429 respondents. The results show that while the Fornell and Larcker criterion indicates acceptable convergent, divergent, and discriminant validity, the HTMT criterion reveals significant multicollinearity issues among the latent variables. Specifically, several constructs, such as culture-productivity, culture-employee, and culture-stakeholder, exhibit problems with discriminant validity. The study concludes that the instrument used in the survey needs revision to address these multicollinearity issues, and recommends adopting the HTMT criterion for more accurate and reliable results. The HTMT criterion is noted for its high sensitivity and specificity in detecting discriminant validity problems, but further empirical evidence is needed to validate this approach.This article evaluates the discriminant validity of a model using both the Fornell and Larcker criterion and the Heterotrait-Monotrait (HTMT) ratio of correlations method. The study uses data from a previous research on value-based excellence in higher education institutions (HEIs) in Malaysia, involving 429 respondents. The results show that while the Fornell and Larcker criterion indicates acceptable convergent, divergent, and discriminant validity, the HTMT criterion reveals significant multicollinearity issues among the latent variables. Specifically, several constructs, such as culture-productivity, culture-employee, and culture-stakeholder, exhibit problems with discriminant validity. The study concludes that the instrument used in the survey needs revision to address these multicollinearity issues, and recommends adopting the HTMT criterion for more accurate and reliable results. The HTMT criterion is noted for its high sensitivity and specificity in detecting discriminant validity problems, but further empirical evidence is needed to validate this approach.