SF-36 Health Survey - Tests of data quality, scaling assumptions, and reliability in a community sample of Chinese Americans

BACKGROUND. Chinese Americans are one of the fastest growing ethnic groups in the United States; however, language and cultural obstacles have challen ged health workers and policy makers seeking to understand the health statu s and needs of this population. OBJECTIVES. This study is the first to use a large-scale probability design to evaluate the 36-item Short-Form Health Survey (SF-36) in a Chinese popu lation (n = 1,501). METHODS. Using the International Quality of Life Assessment Project protoco ls, we examine summated-rating scaling assumptions, item-internal consisten cy, item-discriminant validity, and reliability. RESULTS. Similar to previous studies, our tests indicated that the SF-36 ge nerally met minimum psychometric criteria with high reliability and satisfa ctory scaling success rates for most scales. However, the performance of th e vitality and mental health scales was less satisfactory with regard to di scriminant validity and scaling success rates. Notably, our results indicat e that VT3 and VT4 ("feel worn out" and "tired," respectively) formed a sep arate "fatigue" cluster more highly correlated with the mental health scale . However, MH4 and MH5 ("downhearted and blue" [reverse coded] and "been a happy person") were more highly correlated with the vitality scale, suggest ing that it may be more meaningful to reorganize the vitality and mental he alth items along the dimensions of wellbeing and distress. CONCLUSIONS. These results are interpreted within a cultural framework; how ever, additional work is needed to better understand the relationship betwe en vitality and mental health for Chinese Americans.