Early Language Development of Boys and Girls: Assessment Using LENA in-Home Recordings

Background and Purpose. Language development begins at birth and is identified through clear stages beginning with utterances progressing to complex sentences by ages 3 and 4. As early as 18 months, children from low-income families are 2 months behind, and produce 30% fewer words than children from more advantaged families. Girls typically develop language earlier than boys, and book reading, adult-child interaction, and the parents’ social capital can influence early language skills. This study uses Language Environment Analysis (LENA) audio recordings to describe the home language environment of low-income children age birth to age 3 enrolled in a literacy intervention. Study questions are: (1) What is the extent of meaningful speech, adult-child vocalizations, and child vocalizations among the sample, (2) Are there gender differences in adult-child and child vocalizations, and (3) What is the relationship between the home literacy environment and vocalizations? Methods. The study includes 110 children attending a literacy-based play group in five low-income neighborhoods. Children were ages birth to <12 months (24%), 12 to 24 months (47%) and 25 to 36 months at baseline in 2013. Most children are male (55%), and 79% of families are low income with free lunch status. Most children are White (83%) and Latino (80%). Parents recorded the home audio environment using the LENA device. LENA^© is a “word pedometer” that children wear in a vest for a 16-hour period recording four audio types, child vocalizations, and adult-child reciprocal vocalizations. In the study, recordings averaged 14.3 hours. Descriptive analyses including means, proportions and percentile rankings were used to describe home language environments. Bivariate t-tests, correlations, and multivariate analyses were used to test gender and home environment influences on adult-child and child vocalizations. Results. The 14-hour recordings averaged 2.3 hours of meaningful speech (distinguishable vocalizations), 3.7 hours of distant or overlapping speech, 1.5 hours of television/electronics, and 6.8 hours of silence and background noise. Children averaged 1,756 vocalizations and 466 adult-child reciprocal vocalizations. Count data adjusted to uniform 12-hour recordings showed that child vocalizations ranged from about 300 to almost 5,000, and 44% of children had below-average vocalization and reciprocal adult-child vocalizations compared with nationally-representative data. Bivariate and multivariate analysis showed few related factors. Among 12 to 24-month-olds, boys produced more vocalizations than girls t(39) = 2.82, p = .01. No other statistically significant differences were found by gender. Age moderately predicted child and adult-child vocalizations. Child ethnicity, language, free lunch status, books in the home, and frequency of parent/child reading were not associated with child or reciprocal vocalizations. Conclusions and Implications. That boys in this study vocalize at rates similar to or higher than girls is unexpected, since girls typically acquire language skills before boys. In this low-income sample, both boys and girls had similar home language environments, with parents producing equal response vocalizations. Still, many children are already behind, with few indicators of what contributes to the disparity. Interventions that aim to increase the frequency and quality of adult speech uniformly with children are recommended to help bridge language gaps.