How Do Test-Taking Behaviors Influence WISC-4 Scores?

Background: Wechsler Intelligence Scale for Children-Ⅳ (WISC-Ⅳ) is one of the most reliable and widely used assessment tools in the world (Wahlstrom et al., 2012). Recently, many studies focused on the profiles of clinical data and identified the common characteristics in each clinical group (Wechsler, 2010). However, some researchers have stressed the importance of comprehensive assessment including ecological validity since there are differences in terms of WISC profiles even among children with ASD and ADHD  (Oakland et al., 2012；Dairoku et al., 2011).According to Oakland et al. (2012), WISC-Ⅳ four domain scores are influenced by 3 test-taking behaviors such as “Avoidance”, “Inattentiveness”, and “Uncooperative mood”. Furthermore, other studies revealed that characteristics such as ADHD and ASD are related to these scores (Okada et el., 2010, Oliveras-Rentas et al, 2012). Therefore, it is evident that WISC-Ⅳ scores are influenced by various factors such as test-taking behaviors, ASD and ADHD. Okada et al. (2015) reviewed the existing test-taking checklist and created new checklist called Test-taking behaviors checklist (TBC). It comprises “Emotional state”, “Attentiveness”, “Cooperative mood”, “Comprehensive state”, “Social interaction”, and “Flexibility”, and they proposed that “Social interaction” and “Flexibility” are related to behaviors of children with ASD during the test.

Objectives: To examine the impact of test-taking behaviors on FSIQ (Full-Scale IQ) and four domain scores of WISC-Ⅳ.

Methods: WISC-Ⅳ, TBC, Pervasive Developmental Disorders Autism Society Japan Rating Scale (PARS) , and Attention Deficit-Hyperactivity Disorder Rating Scare (ADHD-RS) were administered to 74 children (5-16year old), who visited counseling room, medical institution, etc. In Japan, PARS and ADHD-RS are used to assess ASD / ADHD symptoms.

Results: Firstly, from principal component analysis and internal consistency analysis by Cronbach’s coefficient alpha, five scales except “Flexibility” of TBC are tested reliable for principal components score. However, “Flexibility ” is not appropriate for the calculation of scores. Secondly, we created models, comprising “Emotional state”, “Attentiveness”, “Comprehensive state”, and “Social interaction” that influence FSIQ and four domain scores. In these models we proposed that “Attentiveness” and “Comprehensive state” relate to age, PARS relates to “Social interaction”, and ADHD-RS relates to “Attentiveness”. The results of structural equation modeling (SEM) revealed that all models fit significantly and “Comprehensive state” is correlated with WISC-Ⅳscores in all models. In addition, “Social interaction” and “Attentiveness” are moderately correlated with PARS /ADHD-RS. 

Conclusions: In this study, the models were created with reference from previous studies, investigated analysis on the impact of FSIQ and the four domain scores of WISC-IV. Consequently, we found that test-taking behaviors influenced these scores to a small extent and “Comprehensive state” relates to all of these scores. Meanwhile, it revealed that “Social interaction” and “Attentiveness”, which were supposed to relate to PSI/WMI, did not relate these index scores. We will discuss the interpretation-system from the perspective of the impact of test-taking behaviors.