Logo image
Construct validity of the item bank of the computerized adaptive testing system for measuring self-care performance (CAT-SC) in children with developmental disabilities
Conference paper

Construct validity of the item bank of the computerized adaptive testing system for measuring self-care performance (CAT-SC) in children with developmental disabilities

Szu-Shen Lai, Yu-Pei Tsai, Yu-Lan Chen, Yu-Ching Lin, Cheng-Te Chen and Kuan-Lin Chen
9th World Congress of the International Society of the Physical and Rehabilitation Medicine 9th World Congress of the International Society of the Physical and Rehabilitation Medicine
2015

Abstract

validity;testing system;self-care performance;CAT-SC;developmental disabilities
Introduction Currently available pediatric self-care assessments have several limitations, such as poor efficiency and feasibility, because they are lengthyand time-consuming. Computerized adaptive testing (CAT), which combines adaptive testing and item response theory by using computer technology, can eliminate these limitations. Therefore, we have been developing a CAT system for measuring self-care performance (CAT-SC) to precisely and efficiently measure self-care performance in children with DD aged 6 months to 12 years. The CAT-SC is being developed in 6 steps: (1) construction of the item bank; (2) field testing; (3) examination of model fitting and item calibration; (4) determination of the stopping rules; (5) construction of a web-based assessment platform; and (6) validation of the efficiency and psychometric properties. At this point, the item bank of the CAT-SC has been constructed, and the purpose of this present study was to examine the construct validity of the item bank of the CAT-SC. Materials and Methods A total of 215 caregivers of children with DD were interviewed with the 73-item CAT-SC item bank. The children were aged 6 months to 12 years and exhibited the entire range of self-care performance. The construct validity of the CAT-SC item bank was assessed by confirmatory factor analysis (CFA) techniques. Results Due to the gender difference, 5 items was removed from the item bank for the first CFA. However, the first CFA showed an inadequate fit. After removing one low factor loading item, the CFA showed that the data had an adequate fit, as judged by various goodness-of-fit (GOF) indices, including the comparative fit index (CFI=0.987), the Tucker-Lewis index (TLI=0.986), and the root mean square error of approximation (RMSEA=0.061). Conclusion The results of the CFA showed a good fit of the unidimensional model of the CAT-SC item bank and thus confirmed that the CAT-SC item bank has good construct validity for measuring self-care performance in children with DD. Therefore, this item bank will be used in the subsequent steps of developing the CAT-SC in the future and the CAT-SC is expected to ease the assessment burden of measuring children’s self-care performance. Keywords: confirmatory factor analysis, unidimension, computer adaptive testing

Metrics

1 Record Views

Details

Logo image