Item and Scale Differential Functioning of the Mini-Mental State Exam Assessed Using the Differential Item and Test Functioning (DFIT) Framework
Survey items or multiple-item scales on statistical analyses can produce misleading results when the probability of an individual endorsing an item or category is influenced by an individual's membership in a particular group. In the present study the authors evaluated the Spanish version of the Mini-Mental State Exam (MMSE) for both differential item functioning and differential scale functioning using the Differential Item and Test Functioning (DFIT) framework. Data for these analyses came from a case registry study of dementia developed among individuals 65 years and older living in North Manhattan, New York. Comparisons were made between item responses of individuals interviewed in Spanish and those interviewed in English. Using the DFIT framework, nine of the 21 items that make up the MMSE showed significant differential item functioning. As a result, comparisons between English and Spanish respondents based on these items individually could lead to the over- or under-identification of cognitive impairment. Assessments of cognitive functioning among respondents to the English and Spanish versions of the MMSE are comparable at the scale level, however. This study does not address why certain items function differently across the groups analyzed. Qualitative methods, such as focus groups and cognitive interviews, are necessary to assess the reasons underlying differential item functioning.