Background: Successfully selecting students for undergraduate courses and for teaching is notoriously difficult. Traditionally, in England, A level grades and interviews have been used in conjunction with a variety of other approaches. Purpose: This paper examines the predictive validity of students' A level grades and of scores from a structured interview on their subsequent degree classification as well as the ratings from their final teaching placements. Sample: A cohort of 71 students who started studying for a Bachelor of Arts (BA) degree in Primary Education in 2007 and graduated in 2010. Design and methods: The A level grades and the ratings from structured interviews, carried out prior to the start of the course, were linked to the final degree classification and to the teaching placement grades. Results: A-level scores correlated significantly with degree classification but not with the placement grade whereas the interview grades correlated significantly with both outcome measures. Conclusions: These results are discussed in the light of the restricted range available for A-level grades. (Contains 8 tables and 2 notes.)