Programmed instruction: an experiment with a programmed text: self-pacing versus group pacing

Previous research in the field of programmed instruction suggests that with groups of homogeneous ability moderate pacing is no less efficient than a self-paced rate of working. To investigate the effects of pacing on hetrogeneous ability groups 180 first year Secondary Modern School children from three schools (I.Q. range 75 - 125 Raven's Progressive Matrices) were randomly assigned to self-paced, moderate paced and fast paced groups to work a programme in physical geography. Raw gain scores on post-test and retention test, analysed by means of a two-way analysis of variance in which the contribution of schools differences to the variance was eliminated, showed the overall difference in variance among methods to be significant (P<0.05) on immediate post-test. The significance of differences between individual methods, estimated i n a series of t tests, confirmed the prediction that self-pacing and moderate pacing would not differ significantly, and that both these methods would be better than fast pacing (P<0.02 in each case). On analysis of retention test scores after an interval of four weeks, these differences among methods were not maintained. Re-analysis on a reduced sample in a three-way analysis did not reveal any tendency for high or low ability, defined in terms of I.Q. scores above or below the median, to have direct influence on scores on post-test or retention test, nor to interact significantly with other factors in the analysis. These results suggest that in programmed instruction externally imposed pacing may provide a means of overcoming differences in ability within groups. There is also from these results, some reason to question the assumption that self-paced rate of working is necessarily the ideal. The need for further investigation into long- term retention would seem to be indicated.