Early Childhood Research Quarterly 26 (2011) 475–483 Contents lists available at ScienceDirect Early Childhood Research Quarterly Exploring the underlying factor structure of the home literacy environment (HLE) in the English and Spanish versions of the Familia Inventory: A cautionary tale Jorge E. Gonzalez a,∗ , Aaron B. Taylor b , Anita S. McCormick a , Victor Villareal a , Minjung Kim a , Erica Perez a , Alicia Darensbourg a , Rebekah Haynes a a Department of Educational Psychology, Texas A&M, College of Education and Human Development, MS 4225, College Station, TX 77843-4225, United States b Department of Psychology, College of Liberal Arts, MS 4235, College Station, TX 77843-4235, United States a r t i c l e i n f o Article history: Received 12 May 2010 Received in revised form 6 December 2010 Accepted 15 December 2010 Keywords: Home Literacy Environment (HLE) Literacy Factor analysis Equivalence a b s t r a c t Few research-based measures of the family literacy environment are commercially available, especially in Spanish. One exception is the Familia Inventory (Taylor, 1995). The present study investigated the 10 subscales of this instrument and the factor structure they imply, using data from a low-socioeconomic (SES), largely Hispanic population. Confirmatory factor analysis (CFA) was used to test the structure of the instrument as well as whether it is invariant across English and Spanish language versions. The scale was administered to 232 parents of preschoolers. Results of the CFA revealed that the a priori dimensions of the Familia Inventory were not supported for this sample. Follow-up analyses suggested poor fit both within and between subscales. The scale fit poorly within both English and Spanish, precluding testing of invariance across language. Supplemental analyses using exploratory factor analysis (EFA) on a subset of the items that had adequate variance suggested that a four-factor solution accounting for 53% of the variance best represented the English data and that a two-factor solution accounting for 43% of the variance best represented the Spanish data. Implications are discussed. © 2011 Elsevier Inc. All rights reserved. Young children enter formal schooling differentially prepared to benefit from their educational experiences (Farver, Xu, Eppe, & Lonigan, 2006). When asked about the origins of these dif- ferences, teachers generally agree on some aspect of the home literacy environment (HLE) (Burgess, Hecht, & Lonigan, 2002). HLE may be conceptualized as the interrelated family resources and opportunities provided to children, combined with the parental skills, abilities, and dispositions that govern the provision of these opportunities (Burgess et al.). The HLE is thought to provide the demographic context within which literacy and language devel- opment occurs and has been linked to differences in children’s language and literacy skills (Snow, Burns, & Griffin, 1998; Weigel, Martin, & Bennett, 2006). But to understand the role of the HLE in children’s later language and literacy development, we must first be able to measure it. The importance of the HLE rests on the fact that the home is the setting where children first encounter the language and liter- acy experiences (Weigel et al., 2006) that presage the development of conventional literacy skills. The research is clear: literacy activi- ties in the more immediate environments of home and community largely influence a child’s literacy development (Gunn, Simmons, ∗ Corresponding author. Tel.: +1 979 845 2324; fax: +1 979 862 1256. E-mail address: jegonzalez@tamu.edu (J.E. Gonzalez). & Kameenui, 1998). Understanding the HLE is of particular interest because of the prevailing presumption that exposing children to an HLE rich in literacy activities is beneficial to their development (Burgess et al., 2002) and, in turn, may provide an indication of a child’s degree of risk for reading problems (Snow et al., 1998). Although conceptual models of the HLE vary, and no definition of this construct has been universally accepted, the importance of the HLE for children’s development is no longer disputed (Payne, Whitehurst, & Angell, 1994). However, in terms of measuring the HLE, few tools with reported score reliability are available in English. To our knowledge, even fewer are available in Spanish. The theoretical and conceptual literature indicates that the HLE is multidimensional (Bracken and Fischel, 2008). For exam- ple, Britto and Brooks-Gunn (2001) divided it into three domains: language and verbal interactions, learning, and the social and emo- tional climate. Similarly, Storch and Whitehurst (2002) posited three domains in their model of HLE: literacy environment, parental characteristics, and parental expectations. In a slight departure from these conceptualizations, in their study on HLE and child language comprehension and expression, Umek, Podlesek, and Fekonja (2005) found support for five factors: stimulation to use language, explanation; reading books to the child, visiting the library and the puppet theatre; joint activities and conversation; interactive reading; and zone-of-proximal-development stimu- lation. Further, Burgess et al. (2002) proposed several distinct 0885-2006/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.ecresq.2010.12.001