Order in the House! Associations among Household Chaos, the Home Literacy Environment, Maternal Reading Ability, and Children's Early Reading

Merrill-Palmer Quarterly, Oct 2008 by Johnson, Anna D, Martin, Anne, Brooks-Gunn, Jeanne, Petrill, Stephen A

The current study examines whether associations exist between household chaos and children's early reading skills, after controlling for a comprehensive battery of home literacy environment characteristics. Our sample included 455 kindergarten and first-grade children who are enrolled in the Western Reserve Reading Project. We go on to test whether these associations are moderated by maternal reading ability. Results suggest that the degree of household order is significantly and positively associated with the expressive vocabulary, Woodcock Reading Mastery, and phonological awareness skills of children whose mothers are above-average readers. By contrast, the number of books a child owns or brings home and how often a child amuses self alone with books are significantly associated with the expressive vocabulary, Woodcock Reading Mastery, and phonological awareness skills of children whose mothers are average-ability readers. These results suggest the potential for new approaches to encouraging literacy development in the home beyond those that depend solely on parental literacy.

The home environment is a primary context for children's early learning and socioemotional development (Bradley, Caldwell, & Rock, 1988; Bradley et al., 1989; Morrison & Cooney, 2001). In particular, scholars have employed the term "home literacy environment" (HLE) to refer to the subset of environmental factors thought to be most germane for literacy growth (Foy & Mann, 2003; Leseman & de Jong, 1998; Burgess, Hecht, & Lonigan, 2002). Two early attempts to operationalize the HLE yielded multifaceted inventories of home and family characteristics, including frequency with which a parent reads to child (also known as shared or joint reading), age when reading with child began, number of minutes spent reading to child yesterday, number of books child owns, frequency with which child asks to be read to, frequency with which child looks at books by self, frequency of trips to library with child, frequency with which mother reads to self, frequency with which father reads to self, amount caregiver enjoys reading to self, child's hours of television viewing per day, and the number of household newspaper, magazine, and child magazine subscriptions (Griffin & Morrison, 1997; Payne, Whitehurst, & Angeli, 1994). Numerous studies have since substantiated an association between the HLE and children's early literacy skills (e.g., Burgess et al., 2002; Frijters, Barron, & Brunello, 2000; Molfese, Modglin, & Molfese, 2003 Rashid, Morris, & Sevcik, 2005; Roberts, J�rgens, & Burchinal, 2005 Senechal, LeFevre, Thomas, & Daley, 1998; Senechal & LeFevre, 2002 Taylor, Clayton, & Rowley, 2004).

The dimensions of the HLE presented by Payne et al. (1994) and Griffin and Morrison (1997) are often treated by researchers as exhaustive. However, neither index explains more than a modest amount of variance in children's early reading skills. After controlling for primary caregiver IQ and education, an index combining nine HLE characteristics explained only 12% of the variance in a sample of 4-year-old children's expressive and receptive vocabulary (Payne et al., 1994). When Griffin and Morrison (1997) controlled for both child general cognitive ability and maternal education, their seven-item HLE index explained only 3% of the variance in kindergarten children's receptive vocabulary. We believe that the explained variance is small because there are aspects of the broader home environment not captured in traditional HLE studies. The present study considers other literacy-related behaviors as well as inputs not directly related to reading, shedding new light on the concept of the HLE.

Researchers looking for other aspects of the HLE that may explain additional variability in early reading development have found that children's interest in, or enjoyment of, reading is predictive of early reading skills (Burgess, 2005; Frijters et al., 2000; Scher, Baker, & Mackler, 1997; Scarborough, Dobrich, & Hager, 1991). These findings are important because they suggest a broader definition of the HLE than that conceived of by both Payne et al. (1994) and Griffin and Morrison (1997), which consisted almost exclusively of those facets of the home environment directly under parental control. Studies have also found that reading behaviors initiated by the child rather than the parent, such as the number of books child brings home from school and the frequency with which child initiates selfreading, are associated with early reading skill (Petrill, Deater-Deckard, Schatschneider, & Davis, 2005). It is likely that to some degree child behaviors are the result of parent behaviors such as encouragement and modeling; however, both sets of behaviors contribute to early reading development.

In addition to expanding the scope of the HLE to include child-directed measures, it is also important to examine other aspects of the home that, although not directly related to literacy instruction, may set the conditions for learning. In particular, there is reason to suspect that household chaos, which has demonstrated negative associations with a range of child cognitive outcomes, may also be associated with children's literacy growth. Chaotic environments have been defined as being very noisy, with a high degree of crowding and foot traffic (i.e., many people coming and going), and a lack of routine, predictability, and organization (Wachs, 1989, 2000).