Premium
PREDICTING LEVELS OF FIRST YEAR UNIVERSITY PERFORMANCE
Author(s) -
WILSON J. D.
Publication year - 1971
Publication title -
british journal of educational psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.557
H-Index - 95
eISSN - 2044-8279
pISSN - 0007-0998
DOI - 10.1111/j.2044-8279.1971.tb02248.x
Subject(s) - psychology , personality , clinical psychology , demography , social psychology , sociology
S ummary . An analysis of the first year academic performance of 1,015 arts and science students at Aberdeen University identified 324 ‘merit’ and 121 ‘fail’ students. The results of school examinations and headmasters' estimates of promise were obtained for the whole group, and test scores of ability, personality, motivation and study methods were collected for about half the students, by interview and questionnaire, over the session. Variables significantly related to performance (analysed separately by sex and faculty) were treated as ‘indicators’ of merit, and ‘symptoms’ of failure. Ratio scores were then calculated on an individual basis, firstly, for all students, on the number of symptoms each showed and secondly, for students with a good academic record, on the number of indicators. The distributions of ratio scores were dichotomised to give the best prediction of performance. Symptoms and indicators varied with sub‐groups, but the overall pattern of results was similar. 45 per cent of all the students were identified from their high ratio scores as ‘at risk’ of failing, but of those who actually failed, 80 per cent fell into the ‘at risk’ category. The prediction of merit performance was more accurate than the prediction of failure. A further analysis suggested how groupings of symptoms might give a more accurate prediction.