This study investigated the extent to which learning approaches can be accounted for by personal factors (i.e., demographics, ability, and personality). The participants were 443 students in a university in mainland China. The Revised Two-factor Study Process Questionnaire, the NEO Five-Factor Inventory-3, and the short form of Raven's Advanced Progress Matrices were respectively applied to test students' learning approaches, personality, and ability. The results of correlations and structural equation modeling indicated that male students were more likely to be deep learners than female students; relative to Year-one students, Year-three students were more likely to use the surface learning approach and less likely to use the deep learning approach. Openness to experience and conscientiousness had strong positive effects on the deep learning approach. Neuroticism had positive effect, whereas conscientiousness had negative effect on the surface learning approach. Approximately 44 % of the variance in the deep learning approach and approximately 18 % of the variance in the surface learning approach could be explained by the three personal factors. Personality was the strongest predictor of learning approaches, whereas ability was the weakest predictor. The implications of the results were discussed.