Using the Rasch Model for Psychometric Examination of Food Security and Acculturation Surveys

Thursday, 25 July 2013: 9:10 AM

Jill Francesca Kilanowski, PhD, MSN, BSN, APRN, CPNP
Department of Nursing, Capital University, Columbus, OH

Learning Objective 1: understand the use of Rasch modeling in psychometric analysis of instruments

Learning Objective 2: discuss the effect of acculturation on the US Household Food Security Module survey.


Rasch model is used to obtain measurements from two categorical variables to assess the probability of a specified response as a function of person and item parameters. That is the probability of a correct response is a logistic function of the difference between the person and item parameter item. The higher level of a person's ability relative to the difficulty of an item, the higher the probability of a correct response on that item. Rasch models have general applicablity and are used in psychometrics to evaluate theory and technique of measurement. The purpose of the study was to evalute the use of the Rasch model on measures of food security and accuturation. The examplar discussed will be the short-version USDA Household Food Security Survey Module (HFSSM) and the Short Acculturation Scale for Hispanics (SASH) in a Latino Midwest USA migrant farmworker population


Using  Winsteps software, we assessed: HFSSM structure, item severity levels, and fitness of sample population. We explored the invariance of HFSSM response through differential item functioning (DIF) analysis patterns between less/more acculturated households


With a convenience sample of n = 112, 97% migrant mothers, results of item infit scores were within adequate range (0.7 to 1.3). The data showed a balanced spread and cohesive order. The item infit statistics showed no substantial deviation from expectations for all the items. The mean item infit score was 0.97 which were close to perfect fit of 1 and 0. The unidimensionality structure of the HFSSM theoretical framework was supported with item severity scores. Food security score was not significantly correlated with acculturation (r = -0.10, p = 0.307).  


Rasch modeling was able to demonstrate the adapted HFSSM performed in agreement with the theoretical framework of food insecurity. The Rasch model was easy to use and interpretation of data was straightforward.