AIM: To compare academic progress and performance of students admitted through two admission systems and to analyse the predictive power of different components in an alternative admission. SAMPLE AND METHODS: The subjects were students admitted to the dental programme at Malmö University, Sweden. The grade admission group was admitted on grades from secondary school (n = 126) and the alternative admission group via an alternative admission procedure (n = 157). The alternative admission procedure consisted of the following components: problem-solving matrices, spatial capacity tested with folding and tin models, manual dexterity, capacity for empathy and interview. Comparisons were made for academic progress (dropouts from the programme and study rate) and academic performance (examinations failed and outcomes of a comprehensive clinical examination). Spearman correlation was calculated for each component of the alternative admission procedure and academic progress as well as academic performance. Multivariate analyses were also carried out. RESULTS: Compared to the grade admission group, the alternative admission group presented lower rate of dropouts (3% vs. 20%, P < 0.001) and a larger proportion graduated within the expected time (88% vs. 60%, P < 0.01). There was no difference between the groups concerning academic performance. Capacity of empathy was correlated with study rate and outcomes of the clinical examination. The matrices predicted low proportion failed examinations and high students' self-assessments in the clinical examination. Predictive power of folding was limited and so was that of the interview. Manual dexterity was not correlated with academic progress or performance. CONCLUSIONS: Results support further development of admission selection criteria, particularly emphatic capacity that predicts important student academic achievements.