Return to search

The efficacy of N-stage testing versus intermediate testing in the formation of equivalence classes of chemical elements

A set of stimuli comprises an equivalence class if the three relations of reflexivity, symmetry, and transitivity are present (Sidman & Tailby, 1982). Although behaviorological researchers have suggested that the training and testing sequence may affect equivalence class formation, this has not been studied directly. The purpose of this study was to investigate the efficacy of two different arrangements of training and testing on the formation of classes of equivalent stimuli. Five middle school students were taught eight conditional discriminations with four classes of stimuli, three stimuli per class (A1B1C1, A2B2C2, A3B3C3, and A4B4C4). A fifth class of stimuli, A5B5C5, was used as control stimuli. Experimental stimuli consisted of five chemical elements, with three attributes per element (name, symbol, and atomic number). The formation of three-member equivalence classes was evaluated by testing students for symmetrical and transitive conditional discriminations involved in the training relations. Two phase arrangements were used with each student. Phase Arrangement I (incorporating n-stage testing) was as follows: (a) AM; (b) B1C1; (c) A1B1 and B1C1 mixed together; (d) test (A1C1, C1A1, A5C5, and C5A5); (e) AM; (f) B2C2; (g) A2B2 and B2C2 mixed together; (h) test (A2C2, C2A2, A5C5, and C5A5); (i) AM, B1C1, AM, and B2C2 all mixed together; and (j) test (AlCl, C1A1, A2C2, C2A2, A5C5, and C5A5). Phase Arrangement II was similar, except that test phases (d) and (h) were eliminated. Stimulus classes A3B3C3 and A4B4C4 were arranged analogously to provide a counterbalanced design. One student performed equally well on equivalence tests with both phase arrangements. Two students performed slightly better on equivalence tests with intermediate testing. One student's performance on equivalence tests demonstrated no equivalence class formation with n-stage testing and much more accurate, although varied, responding with intermediate testing. A 5th student who did not meet the pretest requirements for the study was nevertheless allowed to complete the experimental tasks because his test results were unusual. His first test score was far below chance level, but scores improved with subsequent testing. The results suggest that an intermediate testing arrangement may decrease intersubject variability and, for some individuals, may improve equivalence test performance. / Department of Special Education

Identiferoai:union.ndltd.org:BSU/oai:cardinalscholar.bsu.edu:handle/181642
Date January 1994
CreatorsVieitez, Doreen E.
ContributorsUlman, Jerome D.
Source SetsBall State University
Detected LanguageEnglish
Formatix, 280 leaves ; 28 cm.
SourceVirtual Press

Page generated in 0.0021 seconds