Salient. Victoria University of Wellington Student's Newspaper. Volume 31, Number 1. March 3 1968
Cloud On Parkyn
Cloud On Parkyn
This article, the first of three by Jonathon Cloud, reviews the results of recent research into university education in New Zealand, and attempts to isolate some of the major defects of the present system.
Individual (e.g. psychological) factors were dealt with in a more intensive study by Dr J. J. Small, and hit recommendations will be mentioned later.
Since the personal characteristics and circumstances of the students do not explain why they fail Parkyn takes a closer look at the examination process itself. He discusses two important aspects: the reliability or consistency of examinations, and the standard or pass/fail level. He argues that the average reliability of the usual two-paper examination is about .85 (corrected split-half consistency coefficient). This strikes me as rather high, and overseas studies have usually suggested a lower figure.
But in any case, the fact is that somewhere betwene 8 and 32 per cent of students fail through the unreliability of examinations and examiners.
Parkyn writes: "Whatever the exact numbmay be, these full-time failing students would in fact be no lower in attainments than an equivalent number of passers. They were the unlucky students, whose fate rested on the hazard of fallible examinations."
To raise the reliability and avoid these "mistakes" within the present framework would require much longer examinations — such as six papers for a Stage 1 subject. An alternative would be to average out examination marks with marks from tests and essays during the year. This would mean about nine or ten essays in each subject to supplement the present two-paper final examination, and they would all have to carry out approximately equal weight.
The second aspect is the extent to which a consistent pass rate is maintained over several years and between different universities. The general conclusion is that, while the standards remain fairly constant within a given department over time, there is little consistency between different subjects at the same university, or between different universities when considering the same subject.
Briefly, pass rates depend mainly on the idiosyncratic expectations of the individual departments. They do not depend on variations in the intrinsic difficulty of certain subjects or on variations in the academic ability of the candidates. Furthermore, it seems unlikely that examiners can maintain an absolute standard of judgement throughout the marking of several hundred scripts.
These factors — unreliability and inconsistent standards — show that the examination process itself can be regarded as causing many unjustified failures. Of the 33 per cent of first-year students who fail, less than half can be accounted for by lack of academic ability, deterioration of intellect, or personal problems. The failure of some 16 or 17 per cent of students can be traced directly to examiner unreliability and fluctuations in standards.
Maintaining a high standard is obviously important both to the university and to the community; but Parkyn argues that we can lower the failure rate without lowering the standard. The assessment procedure must be reorganised so as to pass those presently discriminated against by the examination system.
Parkyn suggests that universities as a whole, rather than individual departments, should determine the standard and the failure rate. This should in no case exceed 18 per cent of the first-year fulltime students.
Even this proposal may err on the conservative side. Sir James Mountford, Vice-Chancellor of Liverpool University, writes:
"As a general proposition I would suggest that when more than 10 per cent of a class fails, something is seriously wrong with the selection of students, or the teaching they have received, or the examining to which they have been subjected; and that when the failure rate reaches 25 per cent, it is time for what may euphemistically be called a staff reorganisation."
Dr J. J. Small's work, published in 1966, is basically a highly focused supplement to the Parkyn report. Small studied 99 first-year full-time students at Canterbury, 66 of whom had had some failure. He tried to determine, by an intensive testing and interviewing programme, the effects of unfortunate personal circumstances, aptitudes, personal traits, changes in interest, neglect of work, and so on. Accepting the actual pass rate as the criterion of success (which, in the light of the Parkyn report, is questionable), Small writes: "The main conclusion ... is that failure is due to a variety of factors, of which the most important single one seems to be an intellectualacademic one." He adds, "the performance of students is so idiosyncratic that a reduction in the failure rate would not be easily achieved by general procedures. Attempted improvements should, therefore; be based on the principle of meeting individual needs." Small consequently recommends:
(a) Providing educational and psychological guidance based on a system of academic and personal assessment before matriculation;
(b) Strengthening the work of liaison officers and improving the circumstances of the transition from school to university; and
(c) Improving living conditions and working conditions at the university.
But such changes could not be expected to reduce failure rates dramatically. In a recent report to N.Z.U.S.A., Sue Markham and I wrote:
"It is to be hoped that the universities will follow these recommendations; but it would be tragic if they stopped where Small has. It is not just the performance of students which is idiosyncratic. There seems to be a view prevalent among university administrators and staff that if only the Government would spend more money on accommodation, psychological guidance, and so on, everything would be all right ... It is Just con
P. C. Campion
90 Courtenay Place
(Opposite St James Theatre)
Telephone 55-009 For An Appointmentpage 7
ceivable, however, that improvements might result from better teaching methods, more consistency about standards, and less emphasis on final examinations."
Parkyn himself believes that the factors studied by Small account for only a small proportion of the failures: "Possib'y 4 per cent of the first-year fulltime students failed for such personal reasons." While not denying the importance of Small's conclusions, Parkyn recommends that:
(a) All departments should attempt to estimate the reliability of their tests and examinations, and where necessary improve this reliability;
(b) Examiners should make allowances for the inconsistency of the examination when deciding which students [unclear: shoul]
(c) The academic staff should reach common agreement upon a Stage l pass rate which could be adopted as a tentative reference point for the whole university; and
(d) The failure rate for the Whole of the full-time first-year group should not exceed about 18 per cent.
This, it seems to me, is the very least we can expect from our universities.
Markham, S. and Cloud. J. "Report to N.Z.U.S.A. on First-Year Failure Rates," V.U.W.S.A. Education Subcommittee, 1967.
Mountford, Sir James. "Success and Failure at the University," Universities Quarterly, May 1957.
Parkyn, G. W. Success and Failure at the University: Vol. I, "Academic Performance and the Entrance Standard" (N.Z.C.E.R., 1959); Vol. II, "The Problem pf Failure" (N.Z.C.E.R., 1967).
Small, J. J. Achievement and Adjustment in the First Year at University (N.Z.C.E.R.. 1966).