Centre for Assessment
Information about assessment
Practice and coaching
Strictly speaking, practice and coaching are separate activities. Coaching takes place when children are actually taught the best way to answer the test questions. Practice is simply sitting down and attempting to answer questions that are similar to the ones in the real test.
In the past, much research was conducted by psychologists into the effects of practice and coaching. This research was summarised by Jensen (1980) and found, for example:
practice and coaching effects can be slightly greater for non-verbal tests than for verbal
more able pupils can gain more from practice than less able pupils
practice effects are greater for tests that have a time-limit compared to those without a time limit
The research also found diminishing returns with increased practice and coaching. For example the practice gain between the first and second sessions was usually as great as the total benefit from all further practice. This research went on to suggest that there is no significant gain after about five practice sessions and that there is no significant benefit from more than a few hours of coaching. Furthermore the typical gains in test scores resulting from practice were in the region of 4 to 5 standardised score points and the gains in score resulting from coaching were also around 4 to 5 standardised score points.
More recently, research has also been conducted by Bunting & Mooney (2001) into the effects of familiarisation/practice and coaching on verbal and numerical test scores. The scores obtained from the tests which were originally developed for the Northern Ireland transfer procedure in the 1980s, found that “the effects of familiarisation/practice did not produce a significant change in the means. Coaching for a period of 3 hours did produce a statistically significant shift in the means, though the individuals maintained their rank order. The effect of sustained coaching over a period of 9 months is shown to be substantial.”(Standardised score point gains however are not discussed in this study). The research did not incorporate multi level modelling to take account of the hierarchical nature of the data. Also, without a comparison group it is difficult to know whether some of the long term benefits of coaching might just be attributable to normal child development.
Jensen, A.R. (1980). Bias in mental testing. London: Methuen.pp 590-593
Bunting, B.P. & Mooney, E. (2001). ‘The Effects of Practice and Coaching on Test Results for Educational Selection at Eleven Years of Age’, Educational Psychology, 21, 3, 243-253.