Exploration of the Validation of the WAIS-R (Chinese Version) Short Form in Taiwan

• Main Authors: Chin-Chin Chang, 張勤金
• Other Authors: Wai-Cheong Carl Tam
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/30033695282973549194

碩士 === 中原大學 === 心理學系 === 88 === Abstract The purpose of this research is to explore the validation of the WAIS-R short form (including six subtests) commonly used in Taiwan. The Chinese WAIS-R ten-subtest version (excluding Vocabulary), revised by Dr. Mau-Sun Hua, was employed. There are 3 studies in this project. Study 1 : Since WAIS-R has not been standardized in Taiwan, WISC-III (Chinese version) is used as the validity criterion. Both the Chinese versions of the WAIS-R and WISC-III were administered separately to 36 high school students (age 16) with a time difference of about 41 days. A counterbalanced research design was used. Results indicated that there were significant differences between the two scales on Verbal IQ, Performance IQ, and Full Scale IQ respectively. Moreover, the WAIS-R (Chinese version) tented to overestimate the IQ scores. In studies 2 and 3 only the WAIS-R (Chinese version) was administered. The following 4 criteria were used to investigate the validation of the six-subtest short form: high correlation between the actual and estimated IQ scores, no difference between the mean IQ scores, high percentage of correct category classification, and absolute values of the difference between the actual and estimated IQ scores. Study 2 : The sample was 45 normal adults. Results indicated that the correlation between the ten-subtest and six-subtest versions for the Verbal IQ, Performance IQ, and Full Scale IQ were .94, .89,and .93 respectively, which were acceptably high. However, the correct category classification rate was a low 58%. The six-subtest short form overestimated the Full Scale IQ by 6.7 points. Study 3 : The sample was 45 schizophrenic patients. Results indicated that the correlations between the ten-subtest and six-subtest versions for the Verbal IQ, Performance IQ, and Full Scale IQ were .97, .96, and .98 respectively, which were acceptably high. The correct category classification rate was 84.4%. The six-subtest short form overestimated the Full Scale IQ by 3.8 points. In summary, the six-subtest short form showed better Full Scale IQ estimation for schizophrenic patients than normal adults. Nevertheless, further research is needed to validate its use for screening purposes.