Poster #87 - What can variability in preschoolers’ growth in pre-literacy skills tell us?

• In Event: 3-206 - Poster Session 13
  In Poster Session: PS 13 Section - Education, Schooling

Integrative Statement

Preschoolers’ pre-literacy skills, including alphabet knowledge, phonemic awareness, and expressive and receptive language comprehension, are key building blocks for meeting the academic demands of kindergarten (Duncan et al., 2007; Kaminski et al., 2014). Variability in the development of these skills opens an opportunity for differential pedagogy to meet students’ needs (Greenwood et al., 2013; Justice et al., 2008). Of paramount importance is understanding the sources of variability in the rate of growth in these skills during preschool. From a public policy perspective, it is particularly important to examine socioeconomic, racial/ethnic, and cultural disparities in pre-literacy skill growth (Child Trends, 2015). This project explores factors associated with variability in preschoolers’ growth in pre-literacy skills over the course of an academic year.
This study is part of a larger, ongoing study and includes data collected during the 2017–2018 school year on 97 three-, four-, and five-year-old children from five classrooms in four public preschool programs in a small urban northeastern US community. Children were diverse, classified by their parents as Asian (7%), Black (18%), Hispanic (52%), and White (22%), and were equally male (53%) and female (47%). Annual family income ranged between $0 and $244,000, median=$36,179, which was significantly lower than the 2016 Census median household income for this community, $80,896, t(96)=-5.19, p<.001, d=0.53.
Children’s pre-literacy skills were assessed three times a year by trained school district staff using the Preschool Early Literacy Indicators (PELI; Kaminski et al., 2014). Pre-literacy skill levels and growth over time were predicted by children’s gender, race/ethnicity, age at Time 1 (3 or 4 years old), and family income (centered at the classroom level due to significant between-classroom variance). Multi-level growth modeling was used with occasion nested within children nested within classrooms.
The best-fitting model included random effects for the intercept-by-time covariance at the child level and for the time effect at both the child and classroom levels. Without covariates, pre-literacy scores varied over time, b=14.33, p<.001. Furthermore, the growth rate varied considerably between children, SD = 8.74, p = .009, as well as mildly between classrooms, SD=4.48, p=.05. In addition, the higher children’s initial pre-literacy score, the slower their rate of change, r=-.52, p=.006.
As expected, age category (4 vs. 3 years old) and family income, but not gender, were significantly associated with pre-literacy scores; ethnicity (White vs. Hispanic) was marginally significant (Table 1). Exploratory analyses were conducted on moderation of pre-literacy skill growth by gender, race/ethnicity, age category, and family income. These analyses revealed a significant moderation of pre-literacy skill growth by race/ethnicity (Figure 1). White children showed a steeper increase in pre-literacy skills over the course of the year relative to Asian and Hispanic students. The moderation model explained 18% of the total variance, 28% of the variance between children, 9% of the variance between classrooms, and 23% of the total between child and classroom variance in pre-literacy score growth. Future explorations with this data will incorporate up to 1,000 additional children and child- and classroom-level variables to further explain remaining variance.

Authors

• Craig S. Bailey, Yale University
  Presenting Author

• Cynthia Willner, Yale University
  Non-Presenting Author

• Paula Palermo, Norwalk Acts
  Non-Presenting Author