Factor analysis of the Minnesota Infant Development Inventory based on a Hispanic migrant population

The Minnesota Infant Development Inventory (MIDI), a parental report instru ment, was completed for 168 Hispanic migrant workers' infants ranging in ag e from 1 to 15 months. This study analyzed MIDI scores to determine the ext ent to which it measures the five domains postulated by its authors. Princi pal axis analysis with varimax and orthotran rotations was used to obtain p attern coefficients. The scree test and a parallel analysis were used to es timate the proper numbers of factors to extract. Three solutions were prese nted: (a) a one-factor solution, as indicated by the parallel analysis; (b) a three-factor solution, as indicated by the scree test; and (c) a five-fa ctor solution, as recommended by MIDI's authors. Based on several considera tions (e.g., lack of empirical support for the MIDI five domain scores, par simony, and ease of use), the one-factor solution was judged best for infan ts from Hispanic migrant families until future research supports other prac tices.