A Normative Study on Nelson’s Modified Card Sorting Test in Healthy Middle Age and Elderly Individuals in Taiwan

The Modified Card Sorting Test（MCST）, a modified version of the Wisconsin Card Sorting Test, proposed by Nelson（1976） is one of the most widely used tests for the evaluation of executive function. However, related studies on the MCST and the normative data for the test are scant in Taiwan. The objectives of this study were to evaluate the reliability and validity of MCST in Taiwanese healthy older adults, collect normative data for MCST in healthy older adults in Taiwan, and to examine the relationship between the test performance and demographic variables. The MCST was administered to 426 healthy individuals aged 45 to 88 years, with education level varied from 0 to 18 years. The Trail Making Test, the Color Trail Test, the Semantic Associated of Verbal Fluency, and Similarity and Matrix Reasoning subtests from the WAIS-III (Chinese version) were also administered to examine the validity of the MCST. Thirty of these 426 participants were retested 3 months later for examining the reliability of the MCST. The results revealed that the complete categories were positively correlated with education, but negatively correlated with age. The perseverative errors and the nonperseverative errors were positively correlated with age, but negatively correlated with education. However, no significant correlation was observed between the index scores and sex. Stability coefficients of the MCST ranged from .56 to .68, and the correlations between the MCST and the criterion validity tests were significant. Regression coefficients derived from demographic variables were used to adjust raw scores, and percentile rank norms for these adjusted scores were established. In conclusion, the MCST showed good test-retest reliability and validity, and the normative data are applicable for the healthy individuals aged from 45 to 88 years. For the effective use of the MCST in clinical settings, further studies of specific clinical populations are required.