Translating and validating a Mandarin Chinese version ofthe Computerized Revised Token Test

Introduction: Speech-language pathology is a relatively new clinical discipline in Taiwan. There is a paucity of standardized assessment tools available in Mandarin Chinese. For example, there is one standardized test for aphasia, the Concise Chinese Aphasia Test (CCAT). Because no single assessment tool can serve all assessment purposes, it is important to develop new assessment tools to meet various clinical needs. One test that is culturally neutral, provides easy translation and has a long history of successful psychometric development is the Revised Token Test (RTT). This test has been computerized (Computerized-RTT - CRTT) and was selected for translation and validation in Mandarin Chinese. The primary purpose of this study was to investigate the concurrent validity of the translated CRTT by comparing the performance of normal participants and persons with aphasia (PWA) on this test, to their performance on the CCAT. Method and Procedures: The translation of the CRTT (CRTT-Mandarin) was first validated. This translation was recorded auditorily by a native speaker and the recording was assessed for articulatory and prosodic accuracy. The CCAT and the CRTT-Mandarin were administered to 11 PWA, and 19 normal individuals. All participants were native Chinese speakers, between 38 and 80 years of age, and without premorbid history of speech, language, hearing or learning disorders. Participants were assigned the CRTT-Mandarin and the CCAT randomly and all participants finished the two tests within ten days.Results: Statistical analyses revealed significant differences between groups on the overall and subtest scores of the CRTT-Mandarin. Correlation coefficients computed between the CRTT-Mandarin overall and between subtest scores and each of the CCAT subtest and overall scores were low and nonsignificant for both participant. However, when the groups were combined, the correlations were predominantly high (>.70) and significant (p<.05).Discussion: The CRTT-Mandarin distinguished comprehension performance between PWA and normal controls. This finding, along with the high correlations between the CRTT-Mandarin and the CCAT, provide preliminary concurrent validity for the CRTT-Mandarin. Small sample sizes are a substantive limitation of the study. Future work will establish additional concurrent validity, reliability, sensitivity, specificity, and other psychometric data for the test using larger sample sizes