State exams – trust and confidence

Sir, – Most large-scale examination systems adjust raw marks. The intention is not simply to maintain the normal distribution curve, but to ensure fairness to all candidates. This is particularly the case in qualitative subjects like English or history where there is a greater risk of examiner subjectivity.

A marking meeting and an agreed marking scheme attempt to alleviate this problem, but some examiners are simply more generous, and others less so. However, it is more complex than this. An examiner may be in line with normal distributions for most grades but too generous at the top, or too harsh on weak answers. This need not be a problem. At the end of the marking period, the computer can take account of that examiner’s standard deviation from the norm for each grade, and rectify this by the addition or subtraction of a number of marks. The worst of all worlds is where an examiner is inconsistent in both directions, so that entire batches of scripts have to be remarked.

This was normal practice in the large A-level examination board for which I was a chief examiner, and it had two advantages over the Leaving Certificate system. It is a process that does not eat into the short time allocated for marking, when examiners may be required to adjust their sample marks before they can start the marking proper. Moreover. it doesn’t threaten examiners’ judgment and confidence by asking them for a reassessment of their previous judgment, which is cited as one reason for examiners not seeking reappointment.

The A-level board also asked schools to submit estimated grades in advance of the actual exam. In most cases these were not used, but the computer could identify serious anomalies such as a two or three grade gap between the estimated and actual grade. This merited a remarking of the script by a senior examiner team but could also be used for the other purposes raised in recent discussions.

READ SOME MORE

The Leaving Cert system is unnecessarily cumbersome, but could be rectified in quite straightforward ways and restore both public and examiner confidence. – Yours, etc,

Prof ANNE JORDAN,

Emeritus Professor

of Education,

Waterford Institute

of Technology.