Sustaining Effects of the NC Pre-K Program: Contributions of School Context in Kindergarten

• In Event: 3-200 - Kindergarten Transition and Context: Factors that Influence the Effects of Pre-k Programs

Integrative Statement

The “sustaining environments” hypothesis posits that the positive impact of pre-kindergarten (Pre-K) programming will only be sustained in the long-term if children subsequently experience high-quality educational environments in elementary school (Bailey, Duncan, Odgers, & Yu, 2017). The current study tested this hypothesis in relation to the impact of participation in the NC Pre-K Program on children’s language, math, and executive function skills in the spring of kindergarten—one year after completion of the Pre-K program. Policy-relevant measures of kindergarten classroom and elementary school quality were examined as moderators of NC Pre-K program effects on children’s end of kindergarten skills.

We undertook secondary analyses of data from a recent statewide evaluation of the NC Pre-K program (Peisner-Feinberg et al., 2017) in combination with statewide administrative data from the NC Department of Public Instruction. NC Pre-K, which serves 4-year-olds primarily from low-income families, is considered one of the more established high quality state-funded pre-k programs (Friedman et al., 2018). The sample included two groups of children in 24 school districts in North Carolina using propensity score matching—those who had attended NC Pre-K during the previous academic year (treatment; n = 255) and those who had not attended NC Pre-K (comparison; n = 257). Quality measures included four categories: classroom structural quality (teacher degree; class size), school-wide structural quality (school size; % of teachers with advanced degrees), school-wide achievement (third grade student reading/math proficiency; growth in student reading/math proficiency), and school-wide strain (% student poverty; % teacher turnover rate; % student attendance rate). Separate single-level regression models were estimated for each child outcome. Interaction terms were calculated to test for the moderation of treatment status by each classroom and school quality measure.

We found that the positive effect of NC Pre-K participation on children’s language competencies was larger if children attended kindergarten classrooms with smaller class sizes or schools with lower rates of economic disadvantage; the positive effect on children’s math competencies was larger if children attended schools with a higher percentage of teachers with advanced degrees or larger schools; and the positive effect on children’s executive function skills was larger if children attended classrooms with teachers with elementary education degrees or schools with higher rates of growth in student achievement (see Table 1). For example, the positive effect of NC Pre-K participation on children’s vocabulary skills (WJ III Picture Vocabulary) was moderated by kindergarten class size, resulting in a 0.21 increase in the treatment effect for every 1 SD unit decrease in class size (β = –0.21, p = 0.01; see Figure 1). Sensitivity analyses revealed that the treatment effect was not statistically significant at the average class size of 20 children (ES = 0.04, p = 0.55), but was statistically significant for class sizes of 18 children (ES = 0.21, p = 0.02) and 17 children (ES = 0.29, p = 0.01). In sum, these results provide evidence in support of the “sustaining environments” hypothesis; implications for educational policy will be discussed.

Authors

• Robert Carr, University of North Carolina at Chapel Hill
  Presenting Author

• Ellen Peisner-Feinberg, University of North Carolina-Chapel Hill
  Non-Presenting Author

• Rachel Kaplan, NC Division of Child Development and Early Education
  Non-Presenting Author

• Irina L. Mokrova, The University of North Carolina at Chapel Hill
  Non-Presenting Author