Resumen de The effect of multiple‐choice pre‐testing on post‐test performance in programmed instruction
Michael J. Apter, David R. Boorer
The effect of pre‐testing on post‐test performance in programmed instruction was investigated by comparing three groups of subjects (aged 10‐11): (1) Ss took the same pretest as post‐test; (2) Ss took a different pre‐test from post‐test; (3) Ss took no pre‐test at all. All subjects were given a specially written program followed by a post‐test which was the same for all groups. All the tests were multiple‐choice: this contrasts with previous work using contructed responses and allows the possibility of post‐test results being made worse rather than improved by pre‐testing. The results, however, showed no significant differences between the groups. It is suggested, especially in the light of a significant relationship between taking a pre‐test and reporting enjoying the program, that there may have been two opposing effects at work, one increasing motivation and the other increasing incorrect learning.
