The COVID-19 Pandemic’s Disruptions to Preschool Children’s Cognitive,Social-Emotional,and Executive Functioning Skills

Abstract

Using a longitudinal design that analyzed data captured before and after pandemic-related preschool closures, we compared the kindergarten readiness of children whose preschool experiences were interrupted by the pandemic with the kindergarten readiness of prepandemic children whose preschool experiences were more typical. Pandemic-related disadvantages were observed for early math, social-emotional, and executive functioning, with the pandemic-affected cohort showing disadvantages of between 11 and 18 percentage points. Developmental disadvantages were observed across all income levels, and there was suggestive evidence that the loss of preschool affected the social-emotional and executive functioning skills of children from lower- and higher-income groups differently, although these differences were not always statistically significant. Policy implications are discussed.

Keywords

early childhood achievement gap achievement regression analyses pandemic social-emotional learning executive functioning preschool

Introduction

One of the most troubling aspects of the COVID-19 pandemic was its disruption to schools. Through a shelter-in-place decree, nearly every U.S. jurisdiction ordered or recommended the closing of schools by March 29, 2020 (EmpowerK12, 2020), which meant that students spent nearly one third of the 2019–20 academic year at home. At the outset of the school closures, educators were worried that teachers’ and students’ lack of experience with remote instruction would lead to a developmental disadvantage or learning lag (Dorn et al., 2021), which is defined as developmental or academic progress that is slower than that demonstrated by earlier cohorts of students who were not affected by the pandemic (Pier et al., 2021). Policymakers were particularly anxious that achievement declines would be most severe for children residing in lower-income households because they tend to have inadequate physical learning space within the home (Johnson et al., 2023) and poorer access to technological infrastructure (Haderlein et al., 2021).

Early childhood educators and advocates also were particularly apprehensive about the potential effects of pandemic-related school closures on preschool-aged children (Weisenfeld, 2021). Because early childhood is considered a sensitive period in children’s development in which failure to learn basic skills makes it harder to learn those skills in the future (Black et al., 2017), early childhood educators and advocates were worried about the compounding effects that preschool closures might have for young children’s learning. They noted that given this sensitive period, learning lag in early childhood might have more longer-term consequences than learning lag at other grade levels along the education continuum (Kaffenberger, 2021; Uğraş et al., 2023). In addition, many public preschool programs, which were specifically designed to reduce inequities in educational opportunities and have been found to have larger positive effects on the learning and development of historically marginalized children (Ansari, 2018), were closed due to the shelter-in-place decree (Delap et al., 2021). Thus, policymakers feared that the closures of preschool programs would increase disproportionalities in children’s learning and development (Barnett & Jung, 2021).

Although some educators endorsed the potential of remote learning to help mitigate some of the expected loss in learning as a function of preschool closures (Timmons et al., 2021), others questioned the appropriateness and effectiveness of virtual learning (Dong et al., 2020) because young children learn best through play and play-based learning that emphasizes social interactions and hands-on experiences (Barnett & Jung, 2021). Furthermore, unlike older children who are more experienced with technology use, preschool-aged children who were provided with remote learning opportunities during school shutdowns needed adult support at home for remote learning. However, many parents reported difficulties with providing support for remote learning (Kelly et al., 2023), a challenge that was particularly severe for families juggling their own remote work (Barnett & Jung, 2021) and for families who worked outside the home during the shutdown, which were disproportionately more likely to be lower-wage-earning families and families of color (Jay et al., 2021). These challenges contributed to low levels of participation in remote learning by preschoolers and their families, when offered the opportunity (Ford et al., 2021). Given these conditions, the learning lag may be severe among preschool-aged children, yet little is known about the effects of preschool closures on young children’s learning and development.

Cognitive Lags in Preschool-Aged Children Since the Pandemic

Much of the evidence of learning lag in achievement has been derived from studies on older students, where numerous studies have confirmed steep declines in the achievement of elementary and secondary grade students (Brandley, 2022; Curriculum Associates, 2023; Dorn et al., 2021; Halloran et al., 2021; Kogan & Lavertu, 2021; Kuhfeld et al., 2022; West & Lake, 2021; Uğraş et al., 2023). Although still in the nascent stages, emerging research also has started finding evidence of a learning lag among kindergartners. Markowitz et al. (2023) found that the proportion of kindergarteners in Louisiana who were deemed literacy ready in 2020 was 7 percentage points lower than the proportion of kindergartners who were deemed ready in 2019. Similar results were reported in Virginia, where 25% of entering kindergartners in the fall of 2020 were categorized as being at risk for failure in reading, which was an increase of 50% from the previous year (McGinty et al., 2021). Examining the early literacy skills of the 2020–21 kindergartners in 41 states, Amplify Education (2021) found that 47% of the pandemic-affected kindergarteners needed intensive intervention by midyear compared with the 28% of prepandemic kindergarteners from the 2019–20 academic year. Notably, few studies have examined pandemic-related changes in early math for kindergarteners, even though studies on older students have found learning lags to be more severe in math than in reading, writing, or literacy (Dorn et al., 2021; Kuhfeld et al., 2022; Pier et al., 2021).

Research that examines whether the pandemic had differential effects on preschool-aged children by family income level is also scant, but multiple studies at the upper elementary and secondary grades have found that students residing in lower-income households demonstrated considerably steeper achievement declines than did students residing in higher-income households (Dorn et al., 2021; Engzell et al., 2021; Fahle et al., 2023). Pier et al. (2021) found students who were not economically disadvantaged were between 1 and 2 months behind in English language arts and math, whereas economically disadvantaged students were approximately 3 months behind in both subjects. Similarly, Dorn et al. (2021) found that students in lower-income households were 6–7 months behind in math and reading, whereas students in higher-income households were behind by only half as much. As a result of the unequal rates of learning lag, achievement differences between lower- and upper-income students have been exacerbated.

Disadvantages in Social-Emotional and Executive Functioning Skills

Early educators were not only apprehensive that school closures would compromise children’s academic skills, but they also were worried that school closures would harm children’s burgeoning social-emotional and executive functioning skills (Cameron & Tenenbaum, 2021). These fears stemmed from research suggesting that children’s abilities to understand their own emotions and the emotions of others, control their attention, regulate their impulses, and cope with stress develop within the context of social interactions with their peers and relationships with their teachers (Bukowski et al., 2011; Park, 2016). Because the pandemic deprived children of an expanded social environment in which they could practice their emerging skills, both parents and educators were concerned about potential upticks in children’s emotional health problems and in challenging behaviors that could impede later school success (Egan et al., 2021).

Existing research has confirmed pandemic-related disadvantages in children’s social-emotional skills, with parents reporting that their preschool-aged children were more reliant on them and had more difficulty performing routine tasks since the start of the pandemic (Watts & Pattnaik, 2022). Similarly, in a study of changes in young children’s well-being since the pandemic, Shorer and Leibovich (2022) found that 25% of parents indicated an increase in agitation among their children, and 19% indicated an increase in separation fears. Jiao et al. (2020) also reported increases in parental reports of irritability and clinginess among their children. Teacher perceptions were consistent with parent perceptions as multiple studies have shown that teachers believed that the pandemic-affected preschoolers needed more support with social-emotional development than did previous cohorts of preschoolers who were not affected by the pandemic (Murphy et al., 2023; Tracey et al., 2022).

Similar declines could be observed in children’s executive functioning, which studies have shown can be depressed during times of stress (Hackman et al., 2010; Korzeniowski, 2023). Given that many studies have found that children’s and parents’ stress levels were elevated during the pandemic (Gassman-Pines et al., 2020; Schwartz et al., 2021), it is not surprising that multiple studies have reported disadvantages in executive functioning among pandemic-affected cohorts. Navarro-Soria et al. (2023), for example, found that 6 months after school closures, parents reported a significant decrease in their children’s emotional regulation, cognitive flexibility, and planning. Other studies have described pandemic-induced difficulties with attention (Jiao et al., 2020; Schwartz et al., 2021) and working memory (Lavigne-Cervan et al., 2021). In a statewide survey administered in Arkansas during the spring of 2021, 30% of teachers indicated that more preschool-aged children were struggling with attention problems than was the case prior to the pandemic (Smith et al., 2021).

Extending Previous Research

Although the existing body of research provides a better understanding of how students fared during the pandemic and its aftermath, there remains avenues for further research. First, much of the pandemic-related achievement research has been studied among older children, and little is known about the magnitude of cognitive learning lags among preschool-aged children, especially in early math. Second, there is a dearth of pandemic-related studies that have examined the effects of the pandemic using direct cognitive assessment data collected from preschool-aged children (Weiland et al., 2021). Third, many of the studies conducted during the early stages of the pandemic employed retrospective surveys that asked for parents’ and teachers’ perceived changes in children’s skills and well-being since the start of the pandemic. However, these responses may be subject to recall bias. Finally, many of these latter studies lacked a comparison group, so it is unclear how to interpret the magnitude of changes.

The purpose of our study was to address some of the limitations in the literature by capitalizing on a longitudinal study design that collected data before and after the COVID-19 mandated preschool closures. Because crisis events are difficult to predict, studies such as ours that are able to examine changes in outcomes before and after a crisis on the same set of children are extremely rare (Weems et al., 2007). Our study design was further enhanced by the inclusion of a prepandemic comparison group, which provided a basis for understanding the degree of changes in pandemic-affected children’s skills. Consequently, our study focused on the following research questions:

How do the cognitive, social-emotional, and executive functioning outcomes of preschool children who were affected by the pandemic compare with those of prior cohorts of children who were not affected by the pandemic?

Do these results differ by family income levels?

Methodology

Sample

This study analyzed data collected as part of an evaluation of the Denver Preschool Program (DPP). The DPP provides a tuition credit for all 4-year-old children who reside in the City and County of Denver to attend a preschool of their families’ choosing. Families of all income levels are eligible for the tuition credit, and the tuition credit is scaled to family income, household size, number of hours of preschool in which a child is enrolled, and the quality of the preschool program selected. In the past 5 years, DPP has served approximately 4,700 children annually.

For more than a decade, DPP has conducted an annual evaluation in which a sample of children is randomly selected and indictors of their school readiness are assessed. For the purposes of this study, we compared the outcomes from the cohort affected by the pandemic (i.e., those enrolled in DPP in 2019–20) with the outcomes from the three cohorts of DPP children enrolled in preschool in the years immediately prior to COVID-19 (i.e., during the 2016–17, 2017–18, and 2018–19 academic years). These cohorts were chosen as a basis of comparison because they were assessed on the same timeline as the pandemic-affected cohort. That is, each cohort was evaluated at the beginning of preschool (in the fall) and at the beginning of kindergarten (in the fall of the following year). Thus, all cohorts started preschool under typical circumstances, but two thirds of the way through their preschool year, the pandemic-affected cohort had their schooling disrupted by the pandemic, and this disruption continued through to kindergarten. Our total sample consisted of 662 DPP children, of whom 31% had their preschool year affected by the pandemic (n = 204) and 69% who attended preschool prior to the pandemic (n = 458).

Child Outcomes

Starting with the 2017–18 academic year, we administered surveys to parents to assess children’s social-emotional and executive functioning outcomes. The surveys, which were available in both English and Spanish, were administered at the beginning of preschool and kindergarten. We fielded survey responses for an 8-week window, and reminders were periodically sent to families. Although paper copies were available, nearly all families responded to the surveys via an online format. Families were provided with a gift card for survey completion.

For all cohorts in our study, we administered standardized assessments directly to children to evaluate their cognitive skills at the beginning of their preschool and kindergarten years. During a 6-week window, the cognitive assessments were administered in English to all cohorts during the children’s preschool hours.¹ At kindergarten, the prepandemic cohorts were also assessed during school hours. However, due to the shelter-in-place decree during their kindergarten year, the pandemic-affected cohort was assessed at a time convenient for parents using remote procedures recommended by the test publishers for online administration during the pandemic. Thus, the assessment conditions between the prepandemic and pandemic-affected cohorts were markedly different at kindergarten.

Receptive Vocabulary

We administered the Peabody Picture Vocabulary Test 4 (Dunn & Dunn, 2007) to assess children’s receptive vocabulary skills. This task entailed children choosing among four pictures that best described the word spoken by the assessor. The internal consistency estimate of the test was 0.90.

Early Literacy

We assessed children’s early literacy using the letter-word identification subtest from the Woodcock-Johnson IV Achievement Battery (Schrank et al., 2014). The letter–word identification subtest examined children’s ability to recognize letters and simple words and had an alpha estimate of .80.

Early Math

Early math skills were assessed with the Woodcock-Johnson IV Achievement Battery, Applied Problems Subtest (Schrank et al., 2014). The applied problems subtest, which had an alpha of .82, requires children to perform simple math operations to solve problems.

Social-Emotional Development

We administered the 38-item preschool version of the Devereaux Early Childhood Assessment, 2nd edition (LeBuffe & Naglieri, 2012), to parents in either English or Spanish to assess children’s social-emotional skills. The test organizes items into two subscales, the first of which is protective factors, which assessed children’s initiative, self-regulation, and attachment/relationships with adults. The second subscale is behavioral concerns, which assessed aggression, withdrawal, emotional regulation, and attention problems. Higher scores on the protective factor subscale denote more favorable outcomes, whereas higher scores on the behavioral concerns subscale denote poorer outcomes. The coefficient alphas for the protective factors and behavioral concerns subscales were .92 and .89, respectively.

Executive Functioning

Executive functioning was assessed via the Child Executive Function Inventory (Thorell et al., 2013), a 24-item measure that was administered to parents in either English or Spanish. The test measures two constructs among very young children: working memory and inhibition. Working memory assessed children’s ability to plan and to remember simple instructions. Inhibition assessed children’s ability to refrain from engaging in tasks when instructed and to regulate their behaviors. The test was designed to assess difficulties with executive functioning, so higher scores on these subscales denote poorer outcomes. The coefficient alpha was .87 for working memory and .92 for inhibition.

Survey About Preschool Participation During COVID-19

Between March 16 and May 8, 2020, Denver residents were subjected to their strictest shelter-in-place restrictions, and only essential businesses could remain open. In late May, we administered a survey to parents in our pandemic-affected sample to obtain information about whether their preschool had remained open, the date the preschool closed (if the preschool had closed), whether the child was still attending (if the preschool had remained open), and the number of hours attended (if the child was still attending). We received responses from 88% of our pandemic-affected sample.

Model Covariates

Center Quality

DPP’s administrative records captured the type of preschool (i.e., district- or community-based program) and the quality of each center that DPP children attended. Program quality was measured by Colorado’s quality rating and improvement system. In Colorado, ratings from the quality rating and improvement system are recertified every 3 years. At the time when the pandemic closed preschools in our study, none of the affected centers in our sample were scheduled to undergo recertification of their quality rating. Thus, the quality ratings that had been assigned to the centers for the pandemic-affected group preceded the pandemic and therefore could be used as a baseline measure in the analysis.

The center-level quality rating is a multidimensional assessment of the quality of classroom learning environments using the Early Childhood Environment Rating Scale–Revised (Harms et al., 2005), the quality and frequency of family partnership opportunities, the level of staff qualifications, and classroom ratios and group sizes that together are aggregated across classrooms in a center. Centers’ ratings could range from a score of 1 (poor quality) to a score of 5 (excellent quality). Nearly 85% of children attended centers with quality ratings that were within the top two tiers of Colorado’s quality rating and improvement system.

Child Demographics

DPP’s records also provided the following information about children and their families: child’s birth date, child’s race and ethnicity, child’s gender, family income, household size, primary language(s) spoken at home, hours of preschool received per week, and name of preschool attended. Income was verified through W-2 forms, paystubs, and other financial statements. DPP classifies children into one of six income tiers based on family income and household size, with Level 1 denoting the lowest income tier, Level 5 denoting the highest income tier, and Level 6 denoting families who have opted out of income verification and therefore have unknown income.

Because we randomly sampled at the child level, our sample was generally representative of DPP’s population as a whole. In our analytic sample, children were nearly 4.5 years old when they started preschool, half the sample was female, and half the children lived in households whose family income was below Denver’s median income for the year in which they participated. White children comprised approximately half our study, Hispanic children comprised nearly 30%, Black children comprised roughly 12%, Asian children comprised nearly 3%, and children of mixed racial identities comprised the remaining 6%. Approximately 86% of families indicated that their primary language at home included English. In addition, nearly two thirds of children attended preschool full time, and nearly half attended a community-based preschool.

Timing of the Assessments

Conceivably, there may be differences in outcomes stemming from variation in the timing of the assessments. That is, parents who responded to the survey in the late fall may have provided higher social-emotional and executive function ratings than parents who responded in the early fall because their children may have derived greater benefit from being enrolled in preschool for more weeks than the children of parents who responded in the early fall. Similarly, children whose cognitive assessments were scheduled later in the 6-week window may have benefited from the additional weeks of preschool relative to those who were assessed earlier in the assessment cycle. To account for this possibility, our analysis controlled for children’s age at the time of the assessment, the number of days that had elapsed between the first day of kindergarten and the assessment, and the number of months that had elapsed between the children’s preschool assessment and their kindergarten assessment.

Analytic Approach

Missing Data

We considered data measured at preschool entry (or baseline) to be covariates. At baseline in the fall of preschool, we observed no missing data on the center-level characteristics, less than 5% missing data across the child-level demographic variables, no missing data on the cognitive assessments, and 22% and 11% missing data on the survey measures for the prepandemic and pandemic-affected cohorts, respectively. For the cognitive outcomes measured at kindergarten, we obtained data from 99% of the prepandemic sample and 81% of the pandemic-affected sample. For the survey outcomes measured at kindergarten, we received responses from 84% of the prepandemic sample and 92% of the pandemic-affected sample. Importantly, there were no cohort-related differences in the demographic characteristics of respondents or nonrespondents.

To address missing data, we included the following variables in our imputation models: center-level characteristics, child-level demographics, variables related to the assessment schedule, and outcome measure scores obtained at preschool and at kindergarten. We addressed missing data by using a multivariate distribution to impute 40 sets of values, with the stipulation that the imputed values reflect the observed distribution (Schafer & Graham, 2002). We included the outcome scores obtained at kindergarten in our imputation models to improve precision (von Hippel, 2007), but our analytic models only examined the fully observed outcome data. Model results were aggregated across these multiply imputed datasets using standard procedures (Schafer & Graham, 2002).

Propensity Weights

To increase baseline equivalence between the prepandemic and pandemic-affected samples, which would allow us to better attribute differences in outcomes to the pandemic as opposed to extraneous factors, we created propensity weights. We used a generalized boosted regression model to produce propensity weights (McCaffrey et al., 2013). For each prepandemic child, we calculated a weight equal to the odds of their propensity score using the equation

w_{i} = P (T_{i} = 1 | X_{i}) / [1 - P (T_{i} = 1 | X_{i})]

(1)

where w_i represents the propensity weight, T_i represents child i who was affected by pandemic, and X _i represents the child-level covariates (i.e., child demographics, variables related to the timing of the assessments, and relevant outcome measured at preschool entry) as well as the center-level covariates (i.e., center quality and type of center). The pandemic-affected children were assigned a weight of 1.0. To ensure that the two groups were optimally balanced on each outcome at baseline, we generated outcome-specific propensity weights. For example, the propensity-weight models for receptive vocabulary consisted of the child’s receptive vocabulary score obtained at preschool entry in conjunction with the child- and center-level covariates. Analogously, the propensity-weight models for early literacy consisted of the child’s early literacy score obtained at preschool entry in conjunction with the child- and center-level covariates. We generated separate propensity weights for each of the cognitive, social-emotional, and executive functioning outcomes.

Because the pandemic affected the entire population of 4-year-old children, our propensity models estimated the average treatment effect. We assessed covariate balance by examining the absolute standardized difference (ASD) for each covariate. Adopting a conservatively stringent threshold, we considered covariates to be balanced if the ASD was below .10 (Harder et al., 2010).

Regression Models

We used the propensity weights in a regression model of the form

Y_{i j} = α + β_{1} X_{j} + β_{2} W_{j} + β_{3} T_{i j} + ε_{i j}

(2)

where Y represents the outcome for child i nested in center j, X represents the vector of child-level covariates (i.e., child demographics, variables related to the timing of the assessments, and relevant outcome measured at baseline entry), W represents the vector of center-level characteristics (i.e., type of center and center quality), T represents the indicator variable denoting whether the child was part of the 2019–20 pandemic-affected cohort, and $ε$ represent a random error term. To account for the clustering of children within the same center, we adjusted the standard errors via the Huber–White estimator (Freedman, 2006).² We also standardized nondichotomous variables to have a mean of zero and a variance of one.

We included the same covariates in our analytic models as were included in our propensity models, which allowed us to safeguard against model misspecification and obtain more consistent treatment estimates (Bang & Robins, 2005). In Equation (2), the variable of interest is T. To the extent that the coefficient associated with T is statistically significant, we have evidence for pandemic-related changes in children’s outcomes.

To assess whether children of different income levels may have been differentially affected by the pandemic, we limited the analysis to children whose families submitted verifying documents for their income. That is, we excluded families who were in Level 6 and had unknown income. We classified families who were living at or below 300% of the federal poverty line in the year they participated as lower-income families (i.e., Levels 1, 2, and 3) and families who were above that threshold as higher-income families (i.e., Levels 4 and 5). (Appendix A provides descriptive statistics for the lower- and higher-income samples.) Conducting a similar analysis as previously, we used propensity weights in a regression of the form

Y_{i j} = α + β_{1} X_{i j} + β_{2} W_{j} + β_{3} T_{i j} + β_{4} T_{i j} I_{i j} + ε_{i j}

(3)

where Y, X , W , and T hold the same values as in Equation (2), I is an indicator variable denoting whether the child’s family was living at or below 300% of the federal poverty line, and standard errors were adjusted via the Huber–White estimator. In Equation (3), we have added an interaction term (associated with the coefficient β₄) indicating whether the child was affected by the pandemic and whether the child’s family was classified as lower income. If this interaction term was statistically significant, there was evidence that the pandemic had differential effects on children in lower- and higher-income households.

Effect Sizes and Percentile Differences

To facilitate interpretations of the magnitude of developmental disadvantages, we report Hedges g, which is the standardized mean difference between the pandemic-affected and prepandemic cohorts. In addition, we converted the effect size to an expected percentile difference using the conversion tables given by Marzano Research Laboratory (2009). That is, the effect size was translated to a value on a normal distribution and converted to a percentile difference for a student scoring at the 50th percentile. For example, if we were to observe a learning lag with an effect size of .20, this is roughly equivalent to an 8 percentage point difference such that 58% of the pandemic-affected children would score below the mean score of the prepandemic children.

Results

Preschool Closures During the COVID-19 Pandemic

As essential workers, preschool teachers were allowed to continue teaching in person, albeit with strict limitations on the number of children they could care for in a classroom. Despite preschools being allowed to remain open, centers were closed for 88% of parents. Of the parents who indicated that their preschool remained open, 59% reported that they chose not to send their child to school. Thus, approximately 92% of our pandemic-affected child sample was not attending any preschool program in the spring of 2020. Of the few parents who continued to send their child to preschool, nearly half reported that their child attended preschool for fewer than 20 hours per week. Taken together, the results suggest that most of the pandemic-affected cohort was receiving little to no preschool services during the pandemic. For the purposes of this study, we excluded the approximately 8% of children who attended preschool during the pandemic from our analysis, although the inclusion of these children did not change our findings.

Assessing Covariate Balance

Because we generated outcome-specific propensity weights, the covariate balance varied across each of the seven outcomes. For illustrative purposes, we present the covariate balance between the pandemic-affected and prepandemic cohorts on the first set of imputation for early math skills (Table 1). Prior to any weighting, approximately half the covariates showed ASD values at or below the stringent .10 threshold. After weighting, none of the ASDs were above .10, and the median ASD value across covariates was .02, which indicated excellent balance between the two groups. The covariate balance tables for the other six outcomes (which included an analogous outcome-specific baseline measure) showed similarly small ASDs after weighting (results available on request).

Table 1

Covariate Balance for Pandemic-Affected and Prepandemic Cohorts on Early Math.

Covariate	Before weighting			After weighting
Covariate	Pandemic	Prepandemic	ASD*	Pandemic	Prepandemic	ASD*
Child-level covariates
Early math score at preschool	100.36	98.54	.10	100.36	100.33	.00
Age at kindergarten assessment (mos)	66.70	65.67	.29	66.70	66.39	.09
Days elapsed since start of kindergarten	47.34	47.79	.02	47.34	47.35	.00
Months elapsed since preschool and kindergarten assessments	10.24	9.73	.55	10.24	10.16	.09
Income Level 1	14.3%	24.1%	.28	14.3%	16.7%	.07
Income Level 2	20.5%	17.8%	.07	20.5%	19.7%	.02
Income Level 3	13.0%	11.6%	.04	13.0%	12.3%	.02
Income Level 4	3.7%	5.5%	.09	3.7%	5.1%	.07
Income Level 5	39.1%	25.4%	.28	39.1%	35.7%	.07
Income Level 6	9.3%	15.6%	.22	9.3%	10.5%	.04
Female	53.4%	50.9%	.05	53.4%	52.7%	.02
White children	50.3%	45.8%	.09	50.3%	50.3%	.00
Hispanic children	26.1%	32.0%	.14	26.1%	24.8%	.03
Black children	15.5%	10.7%	.13	15.5%	15.9%	.00
Asian children	3.7%	2.4%	.07	3.7%	3.5%	.01
Other race children	4.3%	9.0%	.23	4.3%	5.6%	.06
Full-time attendee	59.0%	62.5%	.07	59.0%	62.7%	.07
Extended-time attendee	27.3%	23.5%	.09	27.3%	24.4%	.07
Part-time attendee	13.7%	14.0%	.01	13.7%	13.0%	.02
Primary home language English	89.4%	78.5%	.36	89.4%	87.0%	.08
Center-level characteristics
Community-based preschool	49.7%	45.6%	.10	49.7%	50.2%	.01
QRIS rating 1	0.6%	0.9%	.03	0.6%	1.0%	.05
QRIS rating 2	9.9%	3.7%	.21	9.9%	8.5%	.05
QRIS rating 3	7.5%	8.6%	.04	7.5%	7.1%	.01
QRIS rating 4	77.0%	82.2%	.12	77.0%	79.1%	.05
QRIS rating 5	5.0%	4.6%	.02	5.0%	4.3%	.03

ASD, absolute standardized difference; QRIS, Quality Rating and Improvement System

Smaller values indicate better covariate balance.

Estimating Pandemic-Linked Changes in Outcomes

Table 2 shows the sample sizes, the covariate-adjusted outcome means and standard deviations by cohort, and the standardized regression coefficient and associated standard error. The magnitude of learning lag or developmental disadvantage is given by the difference in the outcomes of the pandemic-affected children and the outcomes of the prepandemic children. Pandemic-affected children scored comparably to the prepandemic cohort on receptive vocabulary and letter–word identification but scored significantly poorer on early math skills. The effect size (g = –.29) translated to differences of 11 percentage points, which is equivalent to 61% of the pandemic-affected children scoring below the mean score of the prepandemic children.

Table 2

Pandemic-Affected DPP Participants in Comparison to Prepandemic DPP Participants.

Outcome	Pandemic-affected cohort			Prepandemic cohort			β (SE)	Hedges’ g	Percentage point difference
Outcome	N	Mean	SD	N	Mean	SD	β (SE)	Hedges’ g	Percentage point difference
Cognitive
Receptive vocabulary	160	111.97	13.20	457	110.77	14.39	.06 (.05)	.09	4
Early literacy	163	94.19	12.15	457	93.51	10.90	.01 (.07)	.12	5
Early math*	161	99.99	11.19	456	103.13	10.51	–.21 (.09)	–.29	11
Social-emotional skills
Protective factors*	177	48.49	5.98	304	50.76	6.11	–.27 (.12)	–.37	14
Behavioral concerns*	177	51.28	5.66	302	48.56	6.04	.20 (.10)	.46	18
Executive functioning
Inhibition problems*	175	30.24	3.92	304	28.45	4.17	.21 (.11)	.44	17
Working memory problems*	175	28.94	5.04	304	27.16	5.36	.23 (.11)	.34	13

Notes. The models took into account child- and center-level characteristics. For each outcome, higher scores denote more of the construct. Thus, higher scores on behavioral concerns, inhibition problems, and working memory problems denote less favorable outcomes.

Significant at the .05 level.

Pandemic-affected children also scored more poorly in terms of parent-reported social-emotional and executive functioning outcomes such that they demonstrated lower levels of protective factors, higher scores on behavioral concerns, and greater difficulties with inhibition and working memory than the prepandemic cohort. The magnitudes of developmental disadvantage in these areas were sizable, where the effect sizes for behavioral concerns, inhibition, protective factors, and working memory (g = .46, .44, –.37, and .34, respectively) translated to an approximately 18, 17, 14, and 13 percentage point difference, respectively, in favor of the prepandemic cohort.

Tables 3 and 4 show analogous results by income level. We first contrasted the outcomes of the pandemic-affected children whose family income was at or below 300% of the federal poverty line income against the outcomes of prepandemic children whose family income was also at or below 300% of the federal poverty line. Analogously, we compared the outcomes of the pandemic-affected children whose family income was classified as higher income against the outcomes of the prepandemic children whose family income also was classified as higher income. We then determined whether the magnitude of developmental disadvantage shown by the pandemic-affected children in lower-income households was statistically different from the magnitude of developmental disadvantage shown by the pandemic-affected children in higher-income households by examining the statistical significance of the interaction term in Equation (3). Due to the relatively small sample sizes, we focus on effect sizes as well as statistical significance.

Table 3

Pandemic-Affected DPP Participants from Lower-Income Households in Comparison with Prepandemic DPP Participants from Lower-Income Households.

Outcome	Pandemic-affected cohort			Prepandemic cohort			Β (SE)	Hedges’ g	Percentage point difference
Outcome	N	Mean	SD	N	Mean	SD	Β (SE)	Hedges’ g	Percentage point difference
Cognitive
Receptive vocabulary	76	101.20	12.44	244	100.26	14.67	.05 (.08)	.07	3
Early literacy	79	90.10	11.73	245	90.34	10.79	.01 (.09)	–.02	1
Early math*	77	93.67	10.11	244	97.91	11.17	–.19 (.11)	–.39	15
Social-emotional skills
Protective factors	82	48.10	5.98	160	50.48	5.65	–.23 (.15)	–.41	16
Behavioral concerns*	82	51.25	6.99	159	47.62	6.82	.31 (.13)	.53	20
Executive functioning
Inhibition problems	81	29.69	3.98	160	29.09	3.85	.15 (.15)	.15	6
Working memory problems	81	30.00	5.26	160	27.89	5.36	.27 (.16)	.40	16

Significant at the .05 level.

Table 4

Pandemic-Affected DPP Participants from Higher-Income Households in Comparison with Prepandemic DPP Participants from Higher-Income Households.

Outcome	Pandemic-affected cohort			Prepandemic cohort			β (SE)	Hedges’ g	Percentage point difference
Outcome	N	Mean	SD	N	Mean	SD	β (SE)	Hedges’ g	Percentage point difference
Cognitive
Receptive vocabulary	69	118.37	7.05	142	119.24	7.84	.13 (.15)	.11	4
Early literacy	69	97.34	10.30	141	95.79	11.83	.08 (.14)	.14	6
Early math*	69	106.20	9.87	141	109.63	9.24	–.29 (.13)	–.36	14
Social-emotional skills
Protective factors	75	48.11	6.54	101	50.89	6.01	–.17 (.17)	–.45	17
Behavioral concerns*	75	51.82	4.70	101	49.28	4.83	.30 (.14)	.53	20
Executive functioning
Inhibition problems*	74	30.57	3.65	101	27.49	4.02	.29 (.13)	.65	24
Working memory problems	74	27.88	5.05	101	27.68	5.81	.03 (.15)	.04	2

Significant at the .05 level.

The results for both the lower- and higher-income participants mirrored the trends for the overall population. That is, regardless of income, the pandemic-affected children showed developmental disadvantages with respect to early math and social-emotional and executive functioning skills. Relative to their prepandemic peers, children in lower-income households showed a 15 percentage point decline in early math skills (g = –.39), whereas children in higher-income households showed a 14 percentage point decline (g = –.36). Although the results were not statistically significant, there also was a trend for children from lower- and higher-income households to demonstrate a lag on protective factors, where they scored between 16 and 17 percentage points lower than their prepandemic peers (i.e., g = –.41 for children in lower-income householders and g = –.45 for children in higher-income households). With respect to behavioral concerns, there were statistically significant lags such that both children in lower- and higher-income households demonstrated a 20 percentage point disadvantage when compared with their prepandemic peers (g = .53).

Although children from lower- and higher-income families showed comparable magnitudes of impairment with respect to social-emotional and early math skills, this was not the case for executive functioning skills. Relative to their prepandemic peers, children from higher-income families showed significantly greater lag on inhibition (g = .65, or a 24 percentage point difference) than children from lower-income families (g = .15, or a 6 percentage point difference). The interaction term confirmed that this income-related difference in inhibition lag was statistically significant (β = .20; SE = .10; p = .046). There also was a trend for children from lower-income families to show a comparatively greater lag in working memory (g = .40, or a 16 percentage point difference) than children from higher-income families (g = .04, or a 2 percentage point difference), but the interaction term indicated that this seemingly large income-related difference was not statistically significant.

Discussion

The results of our study confirmed that pandemic-linked disadvantages in cognitive, social-emotional, and executive functioning were not limited to older students but also were observed at the preschool level. Unlike other studies (Amplify Education, 2021; Markowitz et al., 2023; McGinty et al., 2021), we did not observe learning lags in receptive vocabulary or early literacy for preschool-aged children, but we found evidence of learning lags in early math, where 61% of the pandemic-affected children scored below the mean score of the prepandemic children. This finding is consistent with earlier research, where the magnitude of learning lag tended to be larger in math than in English language arts (Kuhfeld et al., 2022). This may be because families of young children reported being more confident with fostering their child’s language development than their child’s math development (LeFevre et al., 2010). Thus, in the absence of preschool, many children may not have been provided with explicit opportunities to engage in mathematical thinking. We also found sizable pandemic-linked disadvantages in behavioral concerns, social-emotional protective factors, inhibition, and working memory, where the prepandemic cohorts showed advantages ranging from 13 to 18 percentage points in these areas over the pandemic-affected cohort.

In addition, the loss of preschool appeared to affect the social-emotional and executive functioning skills of children from lower- and higher-income groups differently, although the results were not always statistically significant. Pandemic-affected children from lower-income households showed a moderately large, albeit not statistically significant, disadvantage in their working memory, with a 16 percentage point disadvantage relative to their prepandemic peers from lower-income households. Prior studies have shown that working memory is highly susceptible to stress (Farah, 2017), which was disproportionately experienced during the pandemic by families in lower-income households (Scrimin et al., 2022).

Relative to their prepandemic higher-income peers, pandemic-affected children in higher-income households showed a particularly large and statistically significant lag with respect to inhibition, where the disadvantage was 24 percentage points. By way of contrast, pandemic-affected children in lower-income households showed a lag of 6 percentage points relative to their prepandemic lower-income peers. It is unclear why children in higher-income families demonstrated such large increases in inhibition problems, but it is important to note that these increases were largely driven by surveys administered to families. One possibility relates to higher-income families’ access to extracurricular activities. Prior to the pandemic, higher-income parents were more likely than lower-income parents to enroll their children in extracurricular activities (Pew Research Center, 2015). Studies have suggested that high-quality enrichment activities are ways in which higher-income parents attempt to confer advantages to the next generation (Covay & Carbonaro, 2010), and higher-income parents may have felt that the loss of these extracurricular opportunities represented a greater threat to their child’s development than did lower-income parents. These concerns may have translated to greater anxiety about how their children were progressing in terms of executive functioning skills.

The findings of our study underscore the wide-ranging deleterious effects of the pandemic, with developmental disadvantages observed on multiple dimensions of kindergarten readiness for children across the income spectrum. Given that children’s skills at preschool serve as the foundation for future learning (Institute of Medicine and National Research Council, 2015), it is imperative that policymakers provide support to address the pandemic-linked cognitive, social-emotional, and executive functioning delays. Yet some of the proposed initiatives emanating from the K–12 sector, such as summer learning programs, intensive tutoring, and expanded learning time, appear geared toward older students and may not be developmentally appropriate for young children. Furthermore, academics-intensive initiatives do not specifically focus on promoting children’s executive functioning or social-emotional skills, which in our study showed even more severe disadvantages. Many of the early pandemic-related investments in early childhood education focused more on initiatives that attempted to financially stabilize this economically vulnerable education service sector (Delap et al., 2021) and less on initiatives that improved the cognitive stimulation of children’s learning environments or supported children’s mental health. Thus, pandemic recovery initiatives may need to be better aligned with children’s developmental stage and encompass initiatives designed to support early cognitive skills as well as children’s social-emotional learning.

Limitations and Future Research

Our study has several limitations that circumscribe the inferences that can be drawn. Due to the shelter-in-place decree, we could not assess the pandemic-affected sample of children in person and instead assessed them remotely. Thus, it is unclear whether the lower math scores demonstrated by the pandemic-affected sample reflected learning lags, differences in testing modes, or a combination of both factors. Notably, we also administered the receptive vocabulary and early literacy measures remotely but did not observe learning lags on these areas. Nonetheless, it remains possible that there may be an interaction between subject-matter content and mode of administration that led to greater suppression in the pandemic-affected cohort’s math performance relative to their prepandemic peers, who took the math assessment in person.

Our study also used family reports of children’s social-emotional and executive functioning skills as opposed to direct assessments. It is possible that the stress that families experienced during the pandemic may have provided a filter through which they viewed their children’s behaviors and executive functioning skills that led to particularly low ratings of children’s skills in these areas. A recent study by Hindman and Bustamante (2019) found, for example, that teachers experiencing depression tended to view children’s behaviors more negatively (Hindman & Bustamante, 2019). Thus, it is possible that stress with parents operates in a similar way that biases families’ ratings of children’s social-emotional and executive functioning skills downward.

Our study also was conducted in the fall of 2020, when children were still in the early stages of the pandemic. Therefore, we may not have captured the full detrimental effects of the pandemic on children’s outcomes. Studies have shown that learning lags became more severe over time, with the cohort of K–12 children who enrolled in school in the fall of 2021 scoring even more poorly than the cohort of K–12 children who had been enrolled in school in the fall of 2020 (Kuhfeld et al., 2022). It is possible that we would observe similar trends for the cohort of preschool children in the fall of 2021. That is, by contrasting the outcomes from the preschool cohort of 2019–20 (who had approximately one third of their preschool year disrupted by the pandemic) to the preschool cohort of 2020–21 (who had their entire preschool year disrupted by the pandemic), we would gain much-needed information about how sustained exposure to the pandemic may have affected children’s outcomes.

Due to limited sample sizes, we could not examine results separately by children’s sociocultural characteristics, aside from family income. Yet, we realize how important it is to examine potential differences in the effects of the pandemic on the developmental and learning outcomes for students of color. Studies suggest that Hispanic families in particular may have experienced disproportionate stress from the pandemic (Causadias & Neblett, 2024), which, in turn, could have disproportionately influenced their children’s developmental outcomes. In addition, research in the K–12 sector has found that Black and Hispanic students experienced more severe learning lags than their White peers (Lewis & Kuhfeld, 2023) due, in part, to inequities in access to favorable remote learning conditions and in access to teachers during the pandemic (Haderlein et al., 2021; Johnson et al., 2023). Racial disproportionalities were not only observed with respect to achievement but also with respect to emotional well-being, where students of color reported higher levels of anxiety during the pandemic (Jacobs & Burch, 2021). Notably, much of this research has been conducted for older students, and additional studies are needed to both understand how the pandemic may have affected the cognitive, social-emotional, and executive functioning of preschoolers of color and to identify the protective factors that enabled preschoolers of color to sustain healthy development during the pandemic.

Conclusion

Our study represents one of the first studies to examine developmental lags in kindergarten readiness skills for preschool children as a function of the pandemic. A unique aspect of our study is the use of a longitudinal design that capitalized on data captured before and after pandemic-related preschool closures, which allowed us to compare the kindergarten readiness of children whose preschool experiences were interrupted by the pandemic to the kindergarten readiness of similarly scoring prepandemic children whose preschool experiences were more typical. We found evidence that the pandemic-induced disruption in preschool experiences was associated with moderately large learning lags in early math and particularly steep disadvantages in social-emotional and executive functioning. These results suggest that elementary schools will need to implement developmentally appropriate solutions to address these lags in kindergarten readiness. With estimates that learning lags can cost students up to $70,000 in potential earnings over their lifetime (Hanushek, 2022), policymakers, educators, parents, and other stakeholders are faced with an unprecedented crisis that needs to be met with a comprehensive and sustainable approach to mitigate the long-lasting impacts on the nation’s youngest learners.

Footnotes

Appendix

Appendix A

Child- and Center-Level Characteristics by Cohort and Family Income Level

Covariate	Pandemic-affected		Prepandemic
Covariate	Lower income	Higher income	Lower income	Higher income
Child-level covariates
Female	52.7%	51.3%	47.8%	57.0%
White children	21.5%	76.3%	18.8%	80.3%
Hispanic children	49.5%	6.6%	50.2%	7.7%
Black children	21.5%	7.9%	19.6%	8.0%
Asian children	1.1%	5.3%	2.0%	1.4%
Other race children	6.5%	3.9%	9.4%	2.6%
Full-time attendee	65.7%	52.6%	71.8%	49.3%
Extended-time attendee	26.9%	30.3%	20.4%	30.3%
Part-time attendee	6.5%	17.1%	7.8%	20.4%
Primary home language English	69.0%	100.0%	62.0%	98.6%
Center-level characteristics
Community-based preschool	46.2%	51.3%	37.1%	61.6%
QRIS rating 1	0.0%	1.1%	0.4%	1.4%
QRIS rating 2	5.4%	13.5%	2.5%	7.0%
QRIS rating 3	4.3%	11.8%	5.7%	12.0%
QRIS rating 4	79.6%	72.4%	86.5%	73.9%
QRIS rating 5	9.7%	1.3%	4.9%	5.6%

QRIS, Quality Rating and Improvement System

Declaration of Conflicting Interests

The authors declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

This study received funding from the Denver Preschool Program (DPP) and from the National Institute of Child Health and Human Development under Grant R03HD106667. The contents are the responsibility of the authors and do not necessarily represent the official views of the funders. Any errors remain our own.

ORCID iDs

Vi-Nhuan Le

Cristal Cisneros

Open Practices Statement

The analytic code and nonproprietary survey measures administered in the study are available on openICPSR. The study has been cross-referenced under the COVID-19 studies subsection and has a unique identifier (). No aspects of the study design were preregistered.

Notes

Authors

VI-NHUAN LE is a principal research scientist at NORC at the University of Chicago. Her research interests lie in mathematics and science reform, educational assessment, postsecondary access and persistence, and early childhood education.

DIANA SCHAACK is an assistant professor of learning, developmental, and family sciences at the University of Colorado, Denver. Her research integrates educational and psychological theories into the study of the relationships, settings, and policies that foster positive child and teacher development within early care and education settings.

CRISTAL CISNEROS is senior director of evaluation and impact at the Denver Preschool Program. She is a first-generation Mexican and former Head Start student whose research interests include educational equity, critical race theory, and early childhood education.

JOLENE GREGORY directs the Research Review Board and is a lead researcher for research and evaluation at Denver Public Schools. She is currently engaging in research in a variety of areas including multilingual learners, early childhood education, special education, and school safety and discipline.

References

Amplify Education. (2021). COVID-19 means more students not learning to read. https://readytogether.sde.ok.gov/sites/default/files/2022-04/Amplify-mCLASS_MOY-COVID-Learning-Loss-Research-Brief_022421.pdf

Ansari

(2018). The persistence of preschool effects from early childhood through adolescences. Journal of Educational Psychology, 110(7), 952–973. https://doi.org/10.1037/edu0000255

Bang

Robins

J. M.

(2005). Doubly robust estimation and missing data and causal inference models. Biometrics, 61(4), 962–972. https://doi.org/10.1111/j.15410420.2005.00377.x

Barnett

W. S.

Jung

(2021). Seven impacts of the pandemic on young children and their parents: Initial findings from NIEER’s December 2020 preschool learning activities survey. National Institute for Early Education Research.

Black

M. M.

Walker

S. P.

Fernald

L. C. H.

Andersen

C. T.

DiGirolamo

A. M.

McCoy

D. C.

Fink

Shawar

Y. R.

Shiffman

Devercelli

A. E.

Wodon

Q. T.

Vargas-Barón

Grantham-McGregor

(2017). Early childhood development coming of age: Science through the life course. The Lancet, 389(10064), 77–90. https://doi.org/10.1016/S0140-6736(16)31389-7

Brandley

A. T.

(2022, November 28). Pandemic learning loss: Diving deeper into the data. Kem C. Gardner Policy Institute. https://gardner.utah.edu/blog-pandemic-learning-loss-diving-deeper-into-the-data/

Bukowski

W. M.

Buhrmester

Underwood

M. K.

(2011). Peer relations as a developmental context. In Underwood

M. K.

Rosen

L. H.

(Eds.), Social development: Relationships in infancy, childhood, and adolescence (pp. 153–179). Guilford Press.

Cameron

Tenenbaum

H. R.

(2021). Lessons from developmental science to mitigate the effects of the COVID-19 restrictions on social development. Group Process Intergroup Relations, 24, 231–236. https://doi.org/10.1177/1368430220984236

Causadias

J. M.

Neblett

E. W.

(2024). Understanding the impact of the COVID-19 pandemic on the mental health of Latinx children, youth, and families: Clinical challenges and opportunities. Journal of Clinical Child & Adolescent Psychology, 53(1), 1–9. https://doi.org/10.1080/15374416.2024.230414

10.

Covay

Carbonaro

(2010). After the bell: Participation in extracurricular activities, classroom behaviors, and academic achievement. Sociology of Education, 83(1), 20–45. https://doi.org/10.1177/0038040709356565

11.

Curriculum Associates. (2023). Annual report: State of student learning in 2023. https://www.curriculumassociates.com/research-and-efficacy/annual-report-the-state-of-student-learning-in-2023

12.

Delap

Franko

Nicolaou

Silva-Padron

Thornton

(2021). Measuring the impact of COVID-19 on Colorado’s early care and learning sector. Early Milestones Colorado. https://earlymilestones.org/wp-content/uploads/2021/03/COVID-Wave-1-Report-Web.pdf

13.

Dong

Cao

(2020).Young children’s online learning during COVID-19 pandemic: Chinese parents’ beliefs and attitudes. Children and Youth Services Review, 118, 105440. https://doi.org/10.1016/j.childyouth.2020.105440

14.

Dorn

Hancock

Sarakatsannis

Viruleg

(2021). COVID-19 and education: The lingering effects of unfinished learning. McKinsey & Company. https://www.mckinsey.com/industries/education/our-insights/covid-19and-education-the-lingering-effects-of-unfinished-learning

15.

Dunn

L. M.

Dunn

D. M.

(2007). PPVT-4: Peabody picture vocabulary test. Pearson Assessments.

16.

Egan

S. M.

Pope

Moloney

Hoyne

Beatty

(2021). Missing early education and care during the pandemic: The socio-emotional impact of the COVID-19 crisis on young children. Early Childhood Education Journal, 49, 925–934. https://doi.org/10.1007/s10643-021-01193-2

17.

EmpowerK12. (2020). Quantifying the impact of COVID-19 school closures on metro Atlanta student proficiency. https://redefinedatlanta.org/wp-content/uploads/2020/06/COVID-19-Impact-on-Atlanta-Student-Achievement.pdf

18.

Engzell

Frey

Verhagen

M. D.

(2021). Learning loss due to school closures during the COVID-19 pandemic. Social Sciences, 118(17), e2022376118. https://doi.org/10.1073/pnas.2022376118

19.

Fahle

E. M.

Kane

T. J.

Patterson

Reardon

S. F.

Staiger

D. O.

Stuart

E. A.

(2023). School district and community factors associated with learning loss during the COVID-19 pandemic. Center for Education Policy Research.

20.

Farah

M. J.

(2017). The neuroscience of socioeconomic status: Correlates, causes, and consequences. Neuron, 96, 56–71. https://doi.org/10.1016/j.neuron.2017.08.034

21.

Ford

Kwon

Tsotsoros

(2021). Early childhood distance learning in the U.S. during the COVID pandemic: Challenges and opportunities. Children and Youth Services Review, 131, 106297. https://ideas.repec.org/a/eee/cysrev/v131y2021ics019074092100373x.html

22.

Freedman

D. A.

(2006). On the so-called “Huber sandwich estimator” and “robust standard errors. The American Statistician, 60, 299–302. https://doi.org/10.1198/000313006X152207

23.

Gassman-Pines

Ananat

E. O.

Fitz-Henley

(2020). COVID-19 and parent–child psychological well-being. Pediatrics, 146(4): e2020007294. https://doi.org/10.1542/peds.2020-007294

24.

Hackman

D. A.

Farah

M. J.

Meaney

M. J.

(2010). Socioeconomic status and the brain: Mechanistic insights from human and animal research. Nature Reviews Neuroscience, 11(9), 651–659. https://doi.org/10.1038/nrn2897

25.

Haderlein

S. K.

Saavedra

A. R.

Polikoff

M. S.

Silver

Rapaport

Garland

(2021). Disparities in educational access in the time of COVID: Evidence from a nationally representative panel of American families. AERA Open, 7(1), 1–21. https://doi.org/10.1177/23328584211041350

26.

Halloran

Jack

Okun

Oster

(2021). Pandemic schooling mode and student test scores: Evidence from US states (NBER working paper no. 29497). https://www.nber.org/papers/w29497

27.

Hanushek

(2022). The economic cost of the pandemic. Hoover Institution.

28.

Harder

V. S.

Stuart

E. A.

Anthony

J. C.

(2010). Propensity score techniques and the assessment of measured covariate balance to test causal associations in psychological research. Psychological Methods, 15, 234–249. https://doi.org/10.1037/a0019623

29.

Harms

Clifford

Cryer

(2005). Early childhood environment rating scale–revised. Teachers College Press.

30.

Hindman

A. H.

Bustamante

A. S.

(2019). Teacher depression as a dynamic variable: Exploring the nature and predictors of change over the head start year. Journal of Applied Developmental Psychology, 61, 43–55. https://doi.org/10.1016/j.appdev.2018.09.004

31.

Institute of Medicine and National Research Council. (2015). Transforming the workforce for children birth through age 8: A unifying foundation. National Academies Press.

32.

Jacobs

Burch

A. E.

(2021). Anxiety during the pandemic: Racial and ethnic differences in the trajectory of fear. Journal of Affect Disorders, 292, 58–66. https://doi.org/10.1016/j.jad.2021.05.027

33.

Jay

Bor

Nsoesie

E. O.

Lipson

S. K.

Jones

D. K.

Galea

Raifman

(2021). Neighbourhood income and physical distancing during the COVID-19 pandemic in the United States. Nature Human Behavior, 4, 1294–1302. https://doi.org/10.1038/s41562-020-00998-2

34.

Jiao

W. Y.

Wang

L. N.

Liu

Fang

S. F.

Jiao

F. Y.

Pettoello-Mantovani

Somekh

(2020). Behavioral and emotional disorders in children during the COVID-19 epidemic. Journal of Pediatrics, 221, 264–266. https://doi.org/10.1016/j.jpeds.2020.03.013

35.

Johnson

C. C.

Walton

Strickler

Elliot

J. B.

(2023). Online teaching in K–12 education in the United States: A systematic review. Review of Educational Research, 93(3), 353–411. https://doi.org/10.3102/00346543221105550

36.

Kaffenberger

(2021). Modelling the long-run learning impact of the COVID-19 learning shock: Actions to (more than) mitigate loss. International Journal of Educational Development, 81, 1–8. https://doi.org/10.1016/j.ijedudev.2020.102326

37.

Kelly

E. M.

Harbin

S. G.

Schwartz

I. S.

(2023). Families’ experiences with online instruction and behavior support during COVID-19. Topics in Early Childhood Special Education, 44(2), 103–114. https://doi.org/10.1177/02711214231182023

38.

Kogan

Lavertu

(2021). The COVID-19 pandemic and student achievement on Ohio’s third-grade English language arts assessment. Ohio State University Press.

39.

Korzeniowski

(2023). Impact of COVID-19 pandemic on the development of children’s executive functions: Implications for school-based interventions. Journal of Biomedical Research and Environmental Sciences, 4(6), 1120–1135. https://doi.org/10.37871/jbres1776

40.

Kuhfeld

Soland

Lewis

(2022). Test score patterns across three COVID-19-impacted school years. NWEA’s Center for School and Student Progress.

41.

Lavigne-Cervan

Costa-Lopez

Juarez-Ruiz

Mier

Real-Fernandez

Sanchez-Munoz

Leon

Navarro-Soria

(2021). Consequences of COVID-19 confinement on anxiety, sleep and executive functions of children and adolescents in Spain. Frontiers in Psychology, 12(334), 565516. https://doi.org/10.3389/fpsyg.2021.565516

42.

LeBuffe

P. A.

Naglieri

J. A.

(2012). Devereux early childhood assessment for preschoolers, second edition: User’s guide and technical manual. Kaplan Early Learning.

43.

LeFevre

J.-A.

Fast

Smith-Chant

B. L.

Skwarchuk

S.-L.

Bisanz

Kamawar

Penner-Wilger

(2010). Pathways to mathematics: Longitudinal predictors of performance. Child Development, 81(6), 1753–1767. https://www-jstor-org.web.bisu.edu.cn/stable/40925297

44.

Lewis

Kuhfeld

(2023). Education’s long COVID: 2022-23 achievement data reveal stalled progress toward pandemic recovery. NWEA’s Center for School and Student Progress.

45.

Markowitz

A. J.

Herring

W. A.

Fares

(2023, April 15). Changes in children’s kindergarten readiness in the wake of COVID-19: Statewide evidence from Louisiana. AERA Study Snapshot. https://www.aera.net/Newsroom/AERA23-Study-Snapshot-Changes-in-Childrens-Kindergarten-Readiness-in-the-Wake-of-COVID-19-Statewide-Evidence-From-Louisiana

46.

Marzano Research Laboratory. (2009). What is an effect size?

47.

McCaffrey

D. F.

Griffin

B. A.

Almirall

Slaughter

M. E.

Ramchand

Burgette

L. F.

(2013). A tutorial on propensity score estimation for multiple treatments using generalized boosted models. Statistics in Medicine, 32(19), 3388–3414. https://doi.org/10.1002/sim.5753

48.

McGinty

A. S.

Gray

Herring

Soland

Partee

(2021). Examining early reading skills in the wake of COVID-19: Virginia Fall 2020 state-wide screening findings and implications. PALS Office at the University of Virginia. https://pals.virginia.edu/pdfs/download/PALS_Fall2020_Findings.pdf

49.

McNeish

Stapleton

L. M.

Silverman

R. D.

(2017). On the unnecessary ubiquity of hierarchical linear modeling. Psychological Methods, 22(1), 114–140. https://doi.org/10.1037/met0000078

50.

Montrosse-Moorhead

Dougherty

S. M.

La Salle

T. P.

Weiner

J. M.

Dostal

H. M.

(2019). The overall and differential effects of a targeted prekindergarten program: Evidence from Connecticut. Early Childhood Research Quarterly, 48, 134–145. https://doi.org/10.1016/j.ecresq.2019.02.006

51.

Murphy

Giordano

Deloach

(2023). Pre-K and kindergarten teacher perception of school readiness during the COVID-19 pandemic. Early Childhood Education Journal, 52, 551–561. https://doi.org/10.1007/s10643-023-01462-2

52.

Navarro-Soria

Costa-Lopez

Collado-Valero

J. A.

Juarez-Ruiz

Mier

Lavigne-Cervan

(2023). Anxiety, sleep habits and executive function during the COVID-19 pandemic through parents’ perception: A longitudinal study. Psychology: Research and Review, 36, 8. https://doi.org/10.1186/s41155-023-00251-5

53.

Park

(2016). Emotional scaffolding as a strategy to support children’s engagement in instruction. Universal Journal of Educational Research, 4(10), 2353–2358. https://doi.org/10.13189/ujer.2016.041013

54.

Pew Research Center. (2015). Parenting in America: Outlook, worries, aspirations are strongly linked to financial situation. https://www.pewresearch.org/social-trends/2015/12/17/4-child-care-and-education-quality-availability-and-parental-involvement/

55.

Pier

Christian

Tymeson

Meyer

R. H.

(2021). COVID-19 impacts on student learning: Evidence from interim assessments in California. Policy Analysis for California Education. https://edpolicyinca.org/publications/covid-19-impacts-student-learning

56.

Primo

Jacobsmeier

M. L.

Milyo

(2007). Estimating the impact of state policies and institutions with mixed-level data. State Politics and Policy Quarterly, 7(4), 446–459. https://doi.org/10.1177/153244000700700

57.

Schafer

J. L.

Graham

J. W.

(2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. https://doi.org/10.1037/1082-989X.7.2.147

58.

Schrank

F. A.

McGrew

K. S.

Mather

Wendling

B. J.

LaForte

E. M.

(2014). Woodcock-Johnson IV tests of achievement: form A. Itasca, IL: Riverside Publishing.

59.

Schwartz

D. S.

Exner-Cortens

McMorris

C. A.

Makarenko

Arnold

Van Bavel

Williams

Canfield

(2021). COVID-19 and student well-being: Stress and mental health during return to school. Canadian Journal of School Psychology, 36(2), 166–185. https://doi.org/10.1177/08295735211001653

60.

Scrimin

Mastromatteo

L. Y.

Hovnanyan

Zagni

Rubaltelli

Pozzoli

(2022). Effects of socioeconomic status, parental stress, and family support on children’s physical and emotional health during the COVID-19 pandemic. Journal of Child and Family Studies, 31(8), 2215–2228. https://doi.org/10.1007/s10826-022-02339-5

61.

Shorer

Leibovich

(2022). Young children’s emotional stress reactions during the COVID-19 outbreak and their associations with parental emotion regulation and parental playfulness. Early Child Development and Care, 192(6), 861–871. https://doi.org/10.1080/03004430.2020.1806830

62.

Smith

Nakamura

Granja

Chow

Grindal

Perez

(2021). Early childhood education throughout the COVID-19 pandemic: The experiences of Arkansas educators. SRI International.

63.

Thorell

L. B.

Veleiro

Siu

A. F.

Mohammadi

(2013). Examining the relation between ratings of executive functioning and academic achievement: Findings from a cross-cultural study. Child Neuropsychology, 19(6), 630–638. https://doi.org/10.1080/09297049.2012.727792

64.

Timmons

Cooper

Bozek

Braund

(2021). The impacts of COVID-19 on early childhood education: Capturing the unique challenges associated with remote teaching and learning in K-2. Early Childhood Education Journal, 49(5), 887–901. https://doi.org/10.1007/s10643-021-01207-z

65.

Tracey

Bowyer-Crane

Bonetti

Nielsen

D’Apice

Compton

(2022). The impact of the COVID-19 pandemic on children’s socio-emotional wellbeing and attainment during the Reception Year. University of York.

66.

Uğraş

Zengin

Papadakis

Kalogiannakis

(2023). Early childhood learning losses during COVID-19: Systematic review. Sustainability, 15, 6199. https://doi.org/10.3390/su15076199

67.

von Hippel

P. T

. (2007). Regression with missing Ys: An improved strategy for analyzing multiply imputed data. Sociological Methodology, 37(1), 83–117. https://doi.org/10.1111/j.1467-9531.2007.00180.x

68.

Watts

Pattnaik

(2022). Perspectives of parents and teachers on the impact of the COVID-19 pandemic on children’s socio-emotional well-being. Early Childhood Education Journal, 51, 1541–1552. https://doi.org/10.1007/s10643-022-01405-3

69.

Weems

C. F.

Pina

A. A.

Costa

N. M.

Watts

S. E.

Taylor

L. K.

Cannon

M. F.

(2007). Predisaster trait anxiety and negative affect predict posttraumatic stress in youths after Hurricane Katrina. Journal of Consulting and Clinical Psychology, 75(1), 154–159. https://doi.org/10.1037/0022-006X.75.1.154

70.

Weiland

Greenberg

Bassok

Markowitz

Rosada

P. G.

Luetmer

Abenavoli

Gomez

Johnson

Jones-Harden

Maier

McCormick

Morris

Nores

Phillips

Snow

(2021). Historic crisis, historic opportunity: Using evidence to mitigate the effects of the COVID-19 crisis on young children and early care and education programs. Gerard R. Ford School of Public Policy, University of Michigan and Urban Institute.

71.

Weisenfeld

G. G.

(2021). Impacts of COVID-19 on preschool enrollment and spending. National Institute for Early Education Research.

72.

West

Lake

(2021). How much have students missed academically because of the pandemic? A review of the evidence to date. Center on Reinventing Public Education at the University of Washington, Bothell.