Reliability and Validity of the Reduced Spanish Version of the Iowa Infant Feeding Attitude Scale

Abstract

Objective. To translate the Iowa Infant Feeding Scale (IIFAS) into Spanish, to test its psychometric properties, and to explore item reduction for the Spanish version. Design. Instrumental study. Setting. Six hospitals in eastern Spain. Participants. A convenience sample of 1,354 pregnant women was recruited. Methods. The IIFAS was translated using forward and back translation. Dimensional structure, internal consistency, and construct validity of the Spanish IIFAS version were assessed. Criteria to improve the reliability of the scale and the predictive power of the items were used to reduce the scale. Results. Statistics on the psychometric properties suggested the need for item reduction. Cronbach's alpha coefficient of the nine-item final version of the Spanish IIFAS (IIFAS-S) was 0.792. The confirmatory factor analysis showed a unidimensional structure. Demographic response patterns and correlations with the IIFAS-S scale provided further evidence of construct validity. Pregnancy IIFAS-S scores significantly predicted breastfeeding rates and exclusivity. Conclusion. The results of this study indicated that the Spanish translation of the IIFAS is a valid and reliable measure of women's attitudes toward breastfeeding.