Marks, get set, go: an evaluation of entry levels and progress rates on a university foreign language programme

Abstract

This paper highlights the problem of falling recruitment to foreign language (FL) degrees in higher education, outlining some of the recent changes in secondary FL education which have contributed in part to this situation. It draws attention to the lack of research into the impact that the fall in numbers of students taking a foreign language A level is having on the quality of the intake onto HE foreign languages courses. It reviews the findings and conclusions of the only major study in this area and then presents the results of a new longitudinal study of FL learning at a major UK university. The findings challenge the prevalent view that A level foreign languages are becoming the preserve of the academically elite and raise questions about the type of skills rewarded at foreign language A level. The study confirms previous research findings relating to poor progression rates in university FL learning, but questions the assumption that poor tuition is the root cause of this malaise.

Notes

*Corresponding author: English for International Students Unit, Department of English at the University of Birmingham, Edgbaston, Birmingham B15 2TT, UK. Email: [email protected]

Coleman (1996b) does draw on very restricted longitudinal data from two university contexts in his pilot study. In the first case (French at Portsmouth University), the sample is small (n=20) and the time scale is modest, middle of Year 1 to start of Year 2. In the second case (German at Portsmouth), the sample is larger (n=124), but the time scale is again restricted to one year of study. The data collated from this study is used primarily as a way of estimating the reliability of the C-test on a test/retest basis. Neither of these cases provides genuine longitudinal data related to the progress of a cohort through all years of university study.

The C-test consisted of five gapped tests, totalling 110 items. It had already been thoroughly trialled as part of the ELPS (n=3739) and had produced a normal distribution. The authors are grateful to Professor Jim Coleman for allowing them to use the ELPS C-test, and for other information on his study.

The grammar test consisted of 125 multiple choice items, covering 13 major topics in German grammar, e.g. adjective endings, word order, prepositions, passive, subjunctive. During the year preceding the main study, the test was piloted with 80 students and an item reliability analysis was conducted.

It is worth noting that the small percentage of students with a C or D grade at German A level were by no means the worst performers on the entry proficiency measures.

There are now three main examination boards in England. OCR=Oxford, Cambridge and RSA examinations. AQA=Assessment for Qualifications Alliance (incorporating the Associated Examining Board, City and Guilds and the Northern Examinations and Assessment Board). EDEXCEL=Edexcel Foundation.

These data derive from the sample's final-year language exams. The moderate correlation coefficient must be seen within the context of the narrow range of A level grades and degree classifications represented in the study. Correlations between raw scores for A level modules (not available) and final examination components would undoubtedly give a more accurate picture of the predictive validity of a foreign language A level.