Research

Abstract:

Early intervention (EI) and early childhood special education (ECSE) services for children with disabilities have expanded substantially across the U.S. over the past few decades, necessitating efforts to recruit and retain a qualified workforce to meet their needs. Despite widespread reports of staffing challenges in this sector, few contemporary studies provide large-scale evidence on this workforce. Using administrative data for all EI/ECSE employees in Oregon from 2008 to 2023, we provide longitudinal descriptive evidence on their composition, distribution, and stability. We show that the workforce has increased significantly, is growing more racially/ethnically diverse, and is more highly educated but less experienced than the state’s K-12 workforce. Turnover remained fairly constant during this period, with the exception of paraprofessionals and non-licensed staff whose retention steadily declined to historic lows. Finally, we show that staff are distributed somewhat inequitably throughout the state, with areas serving more low-income students having the highest child-staff ratios and fewer highly-educated teachers/interventionists. Together these analyses contribute the first longitudinal portrait of an EI/ECSE workforce, providing key insights into their staffing dynamics at scale. 

Link: https://doi.org/10.1080/19345747.2025.2611909

Abstract:

Importance:  Mothers and children in low-income households are more likely to experience worse mental and physical health than those from higher-income households.

Objective: To determine the effect of 4 years of monthly unconditional cash transfers on the mental health of mothers with low-income and the physical health of mothers and children.

Design, Setting, and Participants:  This was a parallel-group, randomized clinical trial conducted from May 2018 to July 2023. Mother-infant dyads were recruited (May 2018-June 2019) from postpartum wards in 12 hospitals in 4 cities: Omaha, Nebraska; Minneapolis/St Paul, Minnesota; New Orleans, Louisiana; and New York, New York. Data were analyzed from September 2023 to February 2025.

Interventions:  Mothers were randomly assigned to receive either a high-cash gift ($333 per month) or a low-cash gift ($20 per month) on debit cards. The cash gifts continued for the first 6 years of their children’s lives. Data analyzed here were collected after 4 years of monthly transfers.

Main Outcomes and Measures:  Outcomes were preregistered and measured around the child’s fourth birthday. Maternal outcomes included depression, anxiety, and body mass index (BMI). Child outcomes included age- and sex-adjusted BMI percentile and maternal report of child health (overall health, times sick in the past year, and presence of chronic health conditions).

Results:  A total of 1000 mother-infant dyads (mean [SD] maternal age, 27.0 [5.8] years) were included in this study. Among those mothers, 400 were randomly assigned to receive the $333 high-cash gift and 600 received the $20 low-cash gift on debit cards. Data were available from 891 mother-child dyads. No statistically detectable group differences were found in maternal depressive symptoms (effect size [ES], 0.04; 95% CI, −0.08 to 0.17; P = .51), anxiety (ES, 0.12; 95% CI, −0.02 to 0.25; P = .09), or BMI (ES, −0.06; 95% CI, −0.21 to 0.09; P = .42). In addition, there were no statistically detectable group differences in child BMI percentile (ES, −0.03; 95% CI, −0.17 to 0.12; P = .73) or overall child health (ES, 0.08; 95% CI, −0.07 to 0.22; P = .30).

Conclusions and Relevance: Monthly unconditional cash transfers totaling approximately $15 000 over 4 years to mothers with low incomes did not improve maternal mental health, maternal or child BMI, or maternal report of children’s health. These results could reflect the absence of causal connections between cash transfers and health, the possibility that impacts of early childhood income may not appear until later in life, or that an 18% increase in income is insufficient to overcome the structural vulnerabilities associated with poverty that contribute to health.

Link: https://doi.org/10.1001/jamapediatrics.2025.1612

Media Coverage: The Atlantic, The Oregonian

Abstract:

How can the United States reduce the number of children in our country who, after growing up in families with incomes below the poverty line, remain poor as adults? This article summarizes key conclusions of a 2024 report issued by the Committee of the National Academies of Sciences, Engineering, and Medicine on Policies and Programs to Reduce Intergenerational Poverty.  The committee sought to identify evidence-based policies and programs directed at low-income children, their families, or their neighborhoods that could reduce children’s chances of remaining low-income as adults. The committee found direct evidence of success in affecting correlates of adult poverty (e.g., low educational attainment and earnings, poor health) for more than two dozen programs and policies addressing health; education; income and parental employment; housing; and crime. Because we concentrate on interventions focused on child health, we include a brief discussion of the importance of the involvement of health care professionals in implementing policy changes resulting from the passage of the One Big Beautiful Bill Act (OBBB Act).

Abstract:

Fifteen percent of U.S. students receive special education services, a widespread intensive intervention with variable effects on student outcomes. Spurred by changes in federal policy, many states and school districts have adopted  Response to Intervention (RTI) as an approach to identify students for special education services. RTI seeks to provide a system for targeting interventions to children facing early academic challenges and identifying children with specific learning disabilities (SLD). This paper uses a difference-in-differences design to examine the effects of RTI adoption on Oregon elementary school students’ special education identification and state-standardized achievement test scores. We find that RTI adoption reduced special education identification by 1.4 percentage points (pp) (10%), primarily through lowering SLD identification by 0.5 pp (15%) and speech or language impairment identification by 0.7 pp (12%). We show that these changes to disability identification did not come at the expense of overall academic achievement and find evidence that RTI boosted reading test scores for some minoritized subgroups. These findings suggest that RTI is a promising approach to supporting students at scale.

Link: https://doi.org/10.26300/72jm-2m09

Abstract:

Currently, 6.1 percent of K-12 students in the United States receive gifted education. Using education and IRS data that provide information on students and their family income, we show pronounced differences in who schools identify as gifted across the distribution of family income. Under 4 percent of students in the lowest income percentile are identified as gifted, compared with 20 percent of those in the top income percentile. Income-based differences persist after accounting for student test scores and exist across students of different sexes and racial/ethnic groups, underscoring the importance of family resources for gifted identification in schools. 

EdWorkingPapers Link: https://doi.org/10.26300/f3f6-vf51

CES Working Papers Link: https://www.census.gov/library/working-papers/2025/adrm/CES-WP-25-73.html

Media Coverage: Chalkbeat

Abstract:

Objective: To examine the association between frequent pediatric headache and academic, cognitive, and behavioral outcomes using sibling comparison and lagged dependent variable models that help account for shared family-level confounding and prior child functioning.

Background: Pediatric headache is associated with poorer educational and behavioral outcomes, but most prior studies compare unrelated children using cross-sectional designs, leaving estimates susceptible to residual confounding from unmeasured child and family characteristics. Whether these associations persist after accounting for shared family factors and preexisting differences in child functioning remains unclear.

Methods: We used nationally representative longitudinal data from the Panel Study of Income Dynamics Child Development Supplement (1997–2007) to estimate sibling comparison models with lagged dependent variables for 2,375 U.S. children aged 8 and older (3,205 person-year observations). Frequent headache was defined as self-reported headache occurring “almost every day” or “every day” in the past month. Outcomes included the Woodcock Johnson-Revised assessment in reading and mathematics, Digit Span working memory, Behavior Problems Index, and Positive Behavior Scale. Sibling comparisons controlled for stable observed and unobserved family characteristics shared by siblings born to the same mother, including socioeconomic background, parenting practices, and genetic factors. Lagged dependent variables measured prior to headache assessment accounted for preexisting differences in academic, cognitive, and behavioral functioning. Models additionally controlled for time-varying and time-invariant child-level covariates. Sensitivity analyses included sharpened q values adjusting for multiple hypothesis testing and Oster's δ to assess the threat of unobserved confounding.

Results: Overall, 7% of youth reported frequent headache. In unadjusted comparisons, frequent headache was associated with significantly worse functioning across all outcomes, with standardized mean differences ranging from −0.33 to −0.61 SD. After incorporating sibling comparisons, lagged outcomes, and child-level covariates, frequent headache remained significantly associated with lower math performance (−0.17 SD; 95% CI −0.30 to −0.04; p = .010), more problem behaviors (−0.23 SD; 95% CI −0.41 to −0.06; p = .010), and fewer positive behaviors (−0.19 SD; 95% CI −0.36 to −0.02; p = .030). Associations with reading achievement (0.02 SD; 95% CI −0.10 to 0.14) and working memory (0.00 SD; 95% CI −0.08 to 0.09) were not significant. Results were robust to adjustments for multiple comparisons (all significant q values < .05) and sensitivity analyses for unobserved confounding (Oster's δ: 1.24–1.79).

Conclusion: Frequent pediatric headache was independently associated with domain-specific academic deficits in mathematics and worse behavioral functioning, even after accounting for shared family-level confounding and prior performance. These results support routine screening for academic and behavioral difficulties during headache visits and highlight the need for rigorous evaluation of clinical interventions and school-based accommodations for youth with recurrent headache.

Abstract:

Disability identification plays a central role in allocating educational supports, yet little evidence documents how identification varies across the family income distribution. Using linked school and tax records from Oregon, we provide new evidence on income-based identification differences for two disability supports: special education and Section 504 plans. Students in the lowest income percentile are nearly four times as likely to receive special education supports as students in the highest income percentile (25% versus 7%). In contrast, high-income students are twice as likely to receive 504 plan supports. We further document substantial heterogeneity in these patterns by disability category, race/ethnicity, and grade. Together, the findings suggest that disability-related needs alone cannot explain the observed identification disparities and highlight how school and family processes shape disability classification and service provision.

EdWorkingPapers Link: https://doi.org/10.26300/dpwm-2673

CES Working Papers Link: https://www.census.gov/library/working-papers/2025/adrm/CES-WP-25-74.html

Media Coverage: The 74, Chalkbeat

Abstract:

Students with disabilities (SWDs) are central to many public investments in private schools, yet we know little about how private schools serve SWDs at scale. Using special education (SpEd) employees as one key measure of services, we provide the first estimates of SpEd staffing in private schools using administrative data. Leveraging novel population data on school employees in Nebraska from 2002-2025, we show that private schools are half as likely to have a SpEd teacher and one fourth as likely to have other SpEd licensed staff compared to public schools. Access to SpEd staff is highly geographically concentrated in the state’s private schools whereas SpEd staff are broadly available across the state’s public schools. At the same time, special educators employed in private schools are more experienced and racially diverse compared to their public-school counterparts, though they turn over at much higher rates. Together, these results provide new insights on SpEd staffing in private schools and raise questions about whether SWDs will have equal educational access as states continue to expand choice options.