Marshall Sarah C, Mungas Dan, Weldon Minda, Reed Bruce, Haan Mary
Berkshire Medical Ctr.
Psychol Aging. 1997 Dec;12(4):718-725. doi: 10.1037/0882-7974.12.4.718.
The purpose of this study was to determine if the Mini-Mental State Examination (MMS; M. F. Folstein, S. E. Folstein, & P. R. McHugh, 1975) demonstrates item bias with respect to measuring cognitive functioning of older Hispanics and non-Hispanics. Assessment of differential item functioning (DIF) of individual MMS items across 3 language/ethnicity groups (English test administration/non-Hispanic ethnicity, English test administration/Hispanic ethnicity, and Spanish test administration/Hispanic ethnicity) was performed by using a logistic regression procedure. Fifteen of the 26 MMS items were significantly related to total score and were shown to provide unbiased measurement across the 3 groups. Normative data are presented for older Hispanics (n = 365) and non-Hispanics (n = 388) on the raw MMS, a 15-item version in which items with significant DIF were eliminated, and a total score statistically adjusted for effects of education and age.
本研究的目的是确定简易精神状态检查表（MMSE；M.F.福尔斯坦、S.E.福尔斯坦和P.R.麦克休，1975年）在测量老年西班牙裔和非西班牙裔的认知功能方面是否存在项目偏差。通过使用逻辑回归程序，对26个MMSE项目在3个语言/种族群体（英语测试施测/非西班牙裔种族、英语测试施测/西班牙裔种族和西班牙语测试施测/西班牙裔种族）中的差异项目功能（DIF）进行了评估。26个MMSE项目中有15个与总分显著相关，并显示在3个群体中提供了无偏差的测量。给出了老年西班牙裔（n = 365）和非西班牙裔（n = 388）在原始MMSE、剔除了具有显著DIF项目的15项版本以及对教育和年龄影响进行了统计调整的总分上的常模数据。
