- Greg J. Duncan
Search for EdWorkingPapers here by author, title, or keywords.
Greg J. Duncan
A survey targeting education researchers conducted in November, 2020 provides predictions of how much achievement gaps between low- and high-income students in U.S elementary schools will change as a result of COVID-related disruptions to in-class instruction and family life. Respondents were asked to suppose that the pre-COVID achievement gap was 1.00 standard deviations. The median forecast for the jump in math achievement in elementary school by spring, 2021 was very large – a change from 1.00 to 1.30 standard deviations. The predicted increase in reading achievement gaps (a change from 1.00 to 1.25 standard deviations) was nearly as large. This implies that many teachers will face classrooms of students with much more heterogeneous learning needs in the fall of the 2021-22 school year than usual. We gauged predictions for the success of efforts by teachers and other educators to make up for lost ground by asking for predictions of achievement gaps in the spring of 2022. Few of the respondents to our survey thought that achievement gaps would revert to their pre-COVID levels. In fact, median predictions of achievement gaps fell very modestly– from 1.30 to 1.25 standard deviations for math and from 1.25 to 1.20 standard deviations for reading. We discuss some implications of these predictions for school district strategies (e.g., tutoring and other skill- building programs focused on individual students) to reduce learning gaps exacerbated by the pandemic.
This paper uses meta-analytic techniques to estimate the separate effects of the starting age, program duration, and persistence of impacts of early childhood education programs on children’s cognitive and achievement outcomes. It concentrates on studies published before the wide scale penetration of state-pre-K programs. Specifically, data are drawn from 67 high-quality evaluation studies conducted between 1960 and 2007, which provide 993 effect sizes for analyses. When weighted for differential precision, effect sizes averaged .26 sd at the end of these programs. We find larger effect sizes for programs starting in infancy/toddlerhood than in the preschool years and, surprisingly, smaller average effect sizes at the end of longer as opposed to shorter programs. Our findings suggest that, on average, impacts decline geometrically following program completion, losing nearly half of their size within one year after the end of treatment. Taken together, these findings reflect a moderate level of effectiveness across a wide range of center-based programs and underscore the need for innovative intervention strategies to produce larger and more persistent impacts.
Using an additional decade of CNLSY data, this study replicated and extended Deming’s (2009) evaluation of Head Start’s life-cycle skill formation impacts in three ways. Extending the measurement interval for Deming’s adulthood outcomes, we found no statistically significant impacts on earnings and mixed evidence of impacts on other adult outcomes. Applying Deming’s sibling comparison framework to more recent birth cohorts born to CNLSY mothers revealed mostly negative Head Start impacts. Combining all cohorts shows generally null impacts on school-age and early adulthood outcomes.
We present a reanalysis of the Tennessee Voluntary Prekindergarten Program (TNVPK), a state-funded program designed to promote the school readiness of 4-year-olds from low-income families. Oversubscribed programs used a lottery to randomly assign prospective enrollees a chance to attend TNVPK. We found that assignment to the program had largely null effects on measures of behavior, attendance, and retention collected during elementary school. TNVPK increased enrollment in special education by 4% between kindergarten and grade 3, and generated negative but generally statistically insignificant impacts on third-grade state test scores. We explore reasons for fadeout as well as threats to internal validity.