Examining Socioeconomic Differences Among Children who are Struggling, Average, and Excelling in Language Skills

• In Event: 3-237 - New Perspectives on Early Language Input, Learning Outcomes, and SES

Integrative Statement

Socioeconomic (SES) differences in early language environments (e.g. Hart & Risley, 1995; Raviv et al., 2004) relate to language outcomes with some studies showing that these differences account for SES differences in later language ability (e.g. Hoff, 2003; Huttenlocher et al., 2010). Despite these findings, there is wide variability within SES strata (Gilkerson et al., 2017; Pan et al., 2005). Here, we compare children across SES who are struggling, average, and excelling language learners. By examining different SES groups with similar outcomes, we can see what are truly SES differences and determine mechanisms that underlie learning across SES groups. Across SES, are struggling, average, and strong language learners experiencing similar quantity and quality of talk?
Using data from the NICHD Study of Early Child Care and Youth Development, 120 children at age 3 were selected. Half came from low-income and half from high-income backgrounds; they had either low, average or high expressive language on the Reynell Developmental Language Scales (Reynell, 1991). Unlike in other studies, we matched children on their language ability across SES, meaning there was no difference in the language outcomes between children from high- and low-income backgrounds. We then asked what differences were observed in children’s early language environments at age 2. We compared groups on two measures: the quality of the language environment as assessed through fluency and connectedness, or the extent to which conversation between the parent and child was smooth and balanced between partners, and quantity of caregiver talk (words per minute). Both measures were calculated (with high reliability) from parent-child interactions in the three-box task, a semi-naturalistic play session involving a book and two toys.
We ran two sets of pairwise contrasts, one for fluency and connectedness and one for words per minute, between the low- and high-income groups for each level of expressive language. For fluency and connectedness, we found a trending difference between low- and high-income for the struggling language group (F(1,114)=3.039, p=.084) and no differences for the average or excelling language groups (See Figure 1). For words per minute, we found significant SES-differences in the struggling language group (F(1,114)=8.808, p=.004) and the excelling language group (F(1,114)=4.484, p=.036), but not for the average language group (see Figure 2).
By comparing low- and high-income children who are equivalent in language outcomes, we can better understand differences both within and across SES-groups. For average and excelling children, there are SES differences in the quantity of talk, which shows that SES-differences found in the literature (e.g. Hart & Risley, 1995) persist even among children who are all high achieving. For struggling children, those from low-income homes hear less language then their high-income peers and, although only trending, poorer quality. Perhaps struggling children from low-income backgrounds, but not low-income children as a whole, are especially at risk for hearing less talk and especially high-quality conversation. Struggling children from high-income families may hear enough talk, but not enough high-quality talk. These provocative results offer a novel perspective on how quantity and quality interact across social class.

Authors

• Lillian Masek, Temple University
  Presenting Author

• Kathy Hirsh-Pasek, Temple University
  Non-Presenting Author

• Roberta Michnick Golinkoff, Univ of DE
  Non-Presenting Author