Open AccessDEVELOPMENT OF AN ADAPTIVE VERSION OF THE “KISSING & DANCING TEST" (KDT)
Author(s) -
Reynaldo Sandrini,
Maria Teresa Goulart,
Roberta Barade,
D. V. d ́Souza,
Patricia Claessens
Publication year - 2021
Publication title -
dementia and neuropsychologia
Language(s) - English
Resource type - Conference proceedings
SCImago Journal Rank - 0.54
H-Index - 21
ISSN - 1980-5764
DOI - 10.5327/1980-5764.rpda021
Subject(s) - cronbach's alpha , item response theory , task (project management) , test (biology) , reliability (semiconductor) , computer science , computerized adaptive testing , arithmetic , psychology , artificial intelligence , statistics , mathematics , psychometrics , engineering , paleontology , power (physics) , physics , systems engineering , quantum mechanics , biology
Background: KDT (Bak & Hodges, 2003) was adapted to Brazilian-Portuguese (Baradel, 2016) with modifications in the stimuli (photos vs pictures); number of foils (4 vs 2) and application (presentation in computer screen). The task has 51 items and was applied to 89 cognitively healthy elderly individuals. Objective: To develop and compare to short-versions of the KDT: fixed (FV) and adaptive (CAT). Methods: Using the ‘R for Statistical Computing’ environment, the answers were analyzed with the classical test theory (CTT) and with the item response theory (IRT). FV was developed with an algorithm based on the maximization of Cronbach’s alpha coefficient and ltm package. CAT was developed using ltm and mirtCAT, from the parameters of difficulty and specificity of each item, estimated by the IRT. Results: The short fixed test reduced the number of items from 51 to 27, obtaining a maximum Cronbach’s alpha of 0.74. The CAT is in its finalization phase and in the preliminary tests it showed an even greater decrease in the number of items, to a maximum standard error of 0.3, in the skill estimation. Conclusion: The application of psychometric technologies allowed a significant reduction in the number of items needed for the application of KDT, making it even more accessible for its target populations in Brazil.