Premium
Significant vs meaningful differences in the effects of tangible reinforcement on intelligence test achievement and reliability of tmr subjects
Author(s) -
Busch John Christian,
Osborne William Larry
Publication year - 1976
Publication title -
psychology in the schools
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.738
H-Index - 75
eISSN - 1520-6807
pISSN - 0033-3085
DOI - 10.1002/1520-6807(197604)13:2<219::aid-pits2310130220>3.0.co;2-f
Subject(s) - reinforcement , psychology , reliability (semiconductor) , intelligence quotient , wechsler adult intelligence scale , test (biology) , comprehension , set (abstract data type) , audiology , achievement test , developmental psychology , statistics , cognition , standardized test , social psychology , power (physics) , mathematics , mathematics education , psychiatry , computer science , medicine , paleontology , physics , quantum mechanics , biology , programming language
The effects of tangible reinforcement on a set of four different intelligence test measures for 72 trainable mentally retarded children was investigated. Performance under a standard and reinforced (M & M) condition, differences in split‐half reliability, power estimates, as well as the effect size resulting from treatments, were considered. The Lorge‐Thorndike vocabulary, WISC Arithmetic, Picture Arrangement, and Comprehension subtests were administered. Reinforced administration resulted in significantly superior performance on all but the Comprehension test, although magnitude of treatment effects was low. Reliabilities under the reinforced condition were higher for all but the Lorge‐Thorndike measures. Confidence bounded effect sizes even at their maximal value did not result in “meaningful” differences in performance, however.