Multivariate prediction of academic performance by Hong Kong University students

Verbal intelligence, English-language skills, personality, and attitude scales were used as predictors of academic performance in 230 male and female arts students at the University of Hong Kong. A series of bivariate, multiple, and canonical correlation analyses were performed. The results showed that verbal intelligence and attitudes, excepting study orientation, were not predictive of performance. English-language skills had the most predictive value, accounting for about 10% of the variance of performance measures. Personality variables failed to predict performance when composite criterion measures (GPAs) were used; however, they proved to be of predictive value when results of individual academic subjects were used as criterion measures. The study points to the importance of using noncomposite criterion measures in prediction and of considering the cultural context of achievement.