NAA enhancing the quality of marking project: The effect of sample size on increased precision in detecting errant marking

By Mudhaffar Al-BayattiBen Jones

Abstract

The quality of examiners’ marking is currently checked by sampling their marking and comparing the marks given with a more senior examiner’s marks.

The aim of this research was to investigate whether increasing the size of this sample re-marking would improve the quality control process significantly. 98 scripts were re-marked according to current quality assurance arrangements, and the relationship between the sample size of scripts and the standard error of the mean difference between Senior Examiner and Assistant Examiner marks was measured.

The pattern of decrease in the standard error, as the sample size of scripts increased, was found to be similar for three Assistant Examiners of varying expertise and experience, and a simulated marker.

There was little to be gained from increasing sample size beyond a certain number of scripts for each type of marker. The minimum number of scripts required to identify errant marking was estimated for each category of examiner, and the more experienced the examiner, the fewer scripts were required.

How to cite

Al-Bayatti, M. F. and Jones, B. (2005). NAA enhancing the quality of marking project : the effect of sample size on increased precision in detecting errant marking. London: Qualifications and Curriculum Authority.