Abstract
Objective:
Assess the prevalence of child behavior, academic and sleep concerns, and parent stress and depression symptoms during COVID-19; test associations of parent-child wellbeing with child school format; and examine effect moderation by child race/ethnicity and material hardship.
Methods:
305 English-speaking parents of elementary school-aged children completed online surveys regarding demographics, child school format, behavior, learning-related experiences, and sleep, and parent stress and depression symptoms. Multivariable linear and logistic regression analyses examined associations of school format with child and parent outcomes.
Results:
Children were 5.00–10.99 years old, with 27.8% underrepresented minority race/ethnicity. Per parental report, 27.7% attended school in-person, 12.8% hybrid, and 59.5% remote. In multivariable models, compared to children receiving in-person instruction, children receiving remote instruction exhibited more hyperactivity [ß 0.94 (95% CI: 0.18, 1.70)], peer problems [ß 0.71 (0.17, 1.25)] and total behavioral difficulties [ß 2.82 (1.11, 4.53)]; were less likely to show academic motivation [OR 0.47 (0.26, 0.85)] and social engagement [OR 0.13 (0.06, 0.25)]; more likely to show schoolwork defiance [OR 2.91 (1.56, 5.40)]; and had a later sleep midpoint [ß 0.37 (0.18, 0.56)] and higher odds of co-sleeping [OR 1.89 (1.06, 3.37)]. Associations of remote learning with behavior difficulties were stronger for children without material hardships.
Conclusion:
Children receiving remote and hybrid instruction were reported to have more difficulties compared to children receiving in-person instruction. Children with material hardships showed more behavior challenges overall, but less associated with school format. Therefore, planning for a return to in-person learning should also include consideration of family supports.
Keywords: remote schooling, COVID-19, child behavior, family well-being
INTRODUCTION
Over 55 million students in the U.S. were impacted by the shift to remote (i.e., computer-based) learning during Spring 2020 of the COVID-19 pandemic.1 During the 2020–2021 academic year, school districts took varied approaches to providing socially distanced learning, including sequences of all-remote, hybrid (i.e., a few days in-person per week, the rest remote), or in-person learning, as local infection rates fluctuated. While children and parents experienced multiple stressors during COVID-19, including isolation,2,3 loss,4 and financial/employment changes,5,6 little research has examined how family well-being varied by school format.
Prior work has largely focused on family adjustment to home confinement during Spring 2020, and found heightened child anxiety and depression;7,8 parenting stress9 and depression;3,10 and harsher discipline practices.3 Studies also documented worsened child behavior in the U.S. and internationally,11,12 including increased hyperactivity and conduct symptoms.12 Among parents, 27% endorsed worsening mental health11 and more negative mood following COVID-19 associated restrictions.5 Child sleep problems, including delayed bedtime,12–14 increased sleep onset latency,13 and delayed rise time12–14 have also been well-described during the COVID-19 pandemic, but it is not known whether their presence depends on school instruction format.
Recent survey research conducted by the U.S. Centers for Disease Control and Prevention (CDC) in October - November 2020 found that families receiving remote or hybrid instruction had lower parent well-being (e.g., more emotional distress, loss of work) and child well-being (e.g., less time outside, worse emotional health) compared to families receiving in-person instruction.6 This work did not examine mechanisms underlying child emotional challenges, such as learning-related experiences (e.g., motivation for learning, task-related defiance), or child sleep patterns. Elucidating mechanisms is important in order to plan appropriate family support services.
In contrast to prior work, which occurred towards the beginning of the COVID-19 pandemic,3,5–9,11–15 this study sought to examine differences in parent and child well-being in Winter 2021, in a single state, in which some school districts had returned to in-person instruction, while others remained remote or hybrid. We focused on elementary school-aged children, because at this age, children build foundational academic skills and require more adult support during instruction. We also sought to examine differences in school format related challenges by the presence of material hardships (e.g., food/housing insecurity), which have been prominent stressors during the pandemic.11,15 Lastly, we sought to examine differences in school format related challenges by child race/ethnicity, as structural inequities16,17 have led to a greater burden of stress for underrepresented minorities.15
METHODS
Study Design
We conducted a survey of parents of elementary school-aged children residing in the state of Michigan during February-March 2021. The survey was deemed exempt from review by our institution’s Institutional Review Board. Parents received a $30 gift card for participation.
Participants
Study information was posted on a statewide university research participant registry, through geographically targeted social media advertisements (initially to all Michigan counties, then restricted to counties with lower enrollment), and via study flyers distributed by clinicians, parent-teacher organizations, and non-profit organizations serving families throughout the state in order to recruit as diverse and representative a sample as possible. Interested parents contacted the research team and were screened for eligibility criteria: 1) Parent or legal guardian ≥ 18 years of age; 2) child aged 5.00–10.99 years (to capture children enrolled in Young 5’s/Kindergarten through 5th grade); 3) parent lives with child > half the week; 4) English speaking; 5) Michigan resident. Each participant reported on a single child. Children of all abilities were included, including those requiring special education services and/or therapies. Of 413 parents screened, 313 parents were eligible and provided online informed consent, of which 305 completed survey questions about school format, learning, behavior, and sleep and were included in the sample.
Survey
Demographic information.
Between February 11 and March 18, 2021, parents completed an online survey, via unique REDCap database18 link, reporting their child’s age, grade in school, sex, and race/ethnicity; their own age, sex, educational attainment, employment, marital status; county of residence and household size and income (from which income-to-needs [ITN] ratio was calculated). Food insecurity, housing insecurity, and financial stress were assessed by using items from the Pediatric Adverse Childhood Experiences and Related Life Events Screener19 and the American Academy of Family Physicians Social Needs Screening Tool.20 We calculated a cumulative hardship score by summation of any endorsed food, housing, or financial stressors; job loss; receipt of public assistance during this school year, and ITN below 200% of the federal poverty level. Each type of hardship was counted as one point on the hardship score, as we had no a priori reason to weigh any hardship more than others. Cumulative hardship was modeled as a continuous variable when used as a covariate and dichotomized into any vs. none for effect moderation analyses.
School format.
Parents reported whether their child’s school was public, private, charter, or homeschooling. Parents selected the main type of schooling their child received from August/September 2020 through the present day: all in-person; all hybrid; mostly in-person with brief periods of remote learning (state-mandated in late November through December 2020); mostly hybrid with brief periods of remote learning; all remote until January/February 2021 (when some schools returned to in-person); all remote; or homeschooling (refer to Appendix 1 for survey questions). Remote learning was defined as “accessing the school curriculum, assignments, and learning activities through a computer or electronic device.” For comparisons of school format, we classified children by what type of school they had attended for most of the year (mostly in-person; mostly hybrid; mostly remote) and excluded 16 homeschooled participants. Parents of children receiving remote instruction also indicated whether they were offered an in-person/hybrid learning option and their reasons for declining. Parents of children receiving remote instruction were asked whether the child takes part in group learning (i.e., “pod”) with a tutor or teacher.
Child behavior.
Parents completed the Strengths and Difficulties Questionnaire (SDQ),21 a validated 25-item survey with subscales for Hyperactivity (Cronbach α= 0.81), Conduct Problems (α = 0.68), Peer Problems (α = 0.63), Emotional Symptoms (α = 0.74), Prosocial Behaviors (α = 0.80) and a Total Difficulties score (α = 0.83). Scores were examined as continuous variables.
Learning-related experiences.
We developed a 10-item survey based on clinical experience with elementary-aged children during COVID-19, regarding the parent’s perception of how well the child knows their teacher and classmates, child’s motivation for learning, and opportunities for friendships (rated on a 5-point Likert scale; each item dichotomized into Agree/Strongly Agree vs. Neutral/Disagree/Strongly Disagree for analyses). Parents reported whether their child took part in therapies outside of school (e.g., speech therapy, tutoring, counseling, etc.) and/or initiated medications management for behavior or attention concerns.
Child sleep.
Parents reported their child’s usual bedtime and waketime on weekdays and weekends. From this, we calculated sleep duration, sleep midpoint (median between sleep onset and awakening), and weekday-to-weekend shift. Parents also reported their child’s sleep latency (dichotomized as <30 minutes or ≥30 minutes); number of overnight arousals on a typical night (dichotomized none vs. any); co-sleeping (defined as sleeping in the same bed as a parent (dichotomized never vs. ever); and whether sleep behaviors had worsened during the pandemic.
Parent stress and Depression symptoms:
Parents completed the Parental Stress Scale22 (α = 0.88) and Centers for Epidemiologic Studies Depression Scale23 (CES-D, α = 0.92), both widely used validated measures.
Data Analysis
Data were analyzed using R version 4.0.5. We performed descriptive analyses to examine variable distributions and inspect for outliers. We conducted Chi-square and Wilcoxon rank sum tests to examine associations of school format with child academic, behavior, and sleep outcomes. In multivariable models, we included all possible confounders that demonstrated statistically significant associations (p <.05) with either school format or child outcomes in bivariate tests (child age, child race/ethnicity, parent education, household ITN, and hardship score). As a secondary analysis, to examine possible mediation, we added parent depression symptoms. Because the results were unchanged, the findings presented are those obtained without adjustment for parent depression symptoms. Child race/ethnicity (dichotomized as non-Hispanic white vs all underrepresented minority categories) and hardship (dichotomized as hardship score of 0 vs. ≥1) were examined for effect moderation for child behavioral outcomes by testing interaction terms with school format.
RESULTS
Descriptive Statistics
Sample.
Sample characteristics are shown in Table 1. Underrepresented minority children were Hispanic, any race (11.3% of the sample), Black non-Hispanic (6.3% of the sample), Asian/Pacific Islander (4.7% of the sample) or more than one race (5.0% of the sample). Parents reported a variety of hardships during the pandemic, including 32.1% having to reduce work hours or quit their job for childcare, 19.7% receiving public assistance, 7.2% often/always not having enough money to pay bills, and 9.5% worried about having enough food.
Table 1:
Sociodemographic characteristics of 305 parents and elementary school-aged children enrolled in the present study
| Characteristic | Mean (SD), range or n (%) |
|---|---|
|
| |
| PARENT | |
|
| |
| Parent sex | |
| Female | 272 (89.2%) |
| Male | 33 (10.8%) |
|
| |
| Parent/guardian age (years) | 38.4 (5.6), range 22–62 |
|
| |
| Parent education | |
| High school/GED/less than HS | 21 (6.9%) |
| Some college or 2-year college degree | 70 (23.0%) |
| 4-year college degree | 79 (26.0%) |
| More than 4 year college degree | 134 (44.1%) |
|
| |
| Parent marital status | |
| Married/live with a partner | 262 (86.2%) |
| Other | 42 (13.8%) |
|
| |
| Parent occupation/location | |
| Stays at home or unemployed | 103 (34.3%) |
| Works full/part-time outside the home | 99 (33.0%) |
| Works full/part-time from home | 98 (32.7%) |
|
| |
| Parent reduced work hours due to childcare needs during COVID-19 | 98 (32.1%) |
|
| |
| CHILD | |
|
| |
| Child age (years) | 7.3 (1.6), range 5–10 |
|
| |
| Child grade in school | |
| Young 5s/Kinder-ready | 13 (4.3%) |
| Kindergarten | 58 (19.0%) |
| 1st | 75 (24.6%) |
| 2nd | 52 (17.0%) |
| 3rd | 48 (15.7%) |
| 4th | 38 (12.5%) |
| 5th | 21 (6.9%) |
|
| |
| Child race/ethnicity | |
| Asian or Pacific Islander | 14 (4.7%) |
| Black/African American, non-Hispanic | 19 (6.3%) |
| Hispanic or Latinx, any race | 34 (11.3%) |
| More than one race, non-Hispanic | 15 (5.0%) |
| White, non-Hispanic | 218 (72.7%) |
|
| |
| Only child | 53 (17.4%) |
|
| |
| School format | |
| Mostly full-time in-school, face-to-face instruction | 80 (26.2%) |
| Mostly hybrid instruction | 37 (12.1%) |
| Mostly virtual/remote instruction | 172 (56.4%) |
| Homeschooling | 16 (5.2%) |
|
| |
| School type | |
| Private | 18 (5.9%) |
| Charter | 22 (7.2%) |
| Public | 249 (81.6%) |
| Homeschooling | 16 (5.2%) |
|
| |
| HOUSEHOLD | |
|
| |
| Income-to-needs ratio | 3.3 (1.7), range 0.14–7.18 |
|
| |
| Hardship score | 0.9 (1.4), range 0–6 |
In this sample, 27.7% of parents reported their child attended mostly in-person school, 12.8% mostly hybrid, and 59.5% mostly remote. There were no differences in school format by age, sex, grade, parent educational attainment, or parent marital status. There was a higher proportion (p<.0001) of underrepresented minority children vs. non-Hispanic white children in the remote learning group (36.7%) compared to the in-person learning group (12.5%). Parents of children receiving remote learning were more likely to be unemployed or working from home (35.7%) than parents of children attending in-person learning (27.5%, p=.04). Of 18 children attending private school, 83.3% were receiving in-person instruction, compared to 23.3% of children attending public school (p<.0001). Of 172 children receiving remote instruction, 25 (14.5%) attended ‘pods’ with a tutor, and 97 (56.4%) reported receiving an in-person option but chose remote instruction due to infection concerns.
Child Behavior and Emotional Symptoms.
Parent-reported behavior outcomes are shown in Table 2. In multivariable models, children receiving remote instruction were reported to have significantly higher SDQ Hyperactivity, Peer Problems, and Total Difficulties scores compared to children receiving in-person instruction, and higher Conduct Problems and Total Difficulties compared to children receiving hybrid instruction. SDQ scores did not differ significantly between children receiving hybrid and in-person instruction.
Table 2.
Child Emotional and Behavioral Difficulties by School Format
| Format of Child’s School Instruction, 2020–2021 | Adjusteda effect estimates or odds ratios (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| SDQ Scores (median [IQR]) | Overall† (n=305) | In-person (n=80) | Hybrid (n=37) | Remote (n=172) | Remote vs. In-person | Hybrid vs. In-person | Remote vs. Hybrid |
| Conduct Problems | 1.0 [0.0; 3.0] | 1.0 [0.0; 2.0]+ | 1.0 [0.0; 2.0]+ | 1.0 [1.0; 3.0]+ | 0.43 (−0.05, 0.90) | −0.25 (−0.93, 0.43) | 0.68 (0.05, 1.30) |
| Emotional Symptoms | 2.0 [1.0; 4.0] | 2.0 [1.0; 4.0]+ | 1.5 [1.0; 4.0]+ | 2.5 [1.0; 5.0]+ | 0.68 (0.004, 1.35) | 0.20 (−0.77, 1.17) | 0.48 (−0.42, 1.37) |
| Hyperactivity | 5.0 [3.0; 7.0] | 4.0 [2.0; 6.0]* | 3.0 [2.0; 7.0]* | 5.0 [3.0; 7.5]* | 0.94 (0.18, 1.70) | 0.07 (−1.03, 1.16) | 0.87 (−0.14, 1.88) |
| Peer Problems | 2.0 [1.0; 3.0] | 1.0 [0.0; 3.0]* | 1.0 [0.0; 2.0]* | 2.0 [1.0; 4.0]* | 0.71 (0.17, 1.25) | 0.17 (−0.61, 0.95) | 0.54 (−0.18, 1.25) |
| Prosocial Behaviors | 8.0 [6.0;10.0] | 8.0 [6.0;10.0] | 9.0 [7.2;10.0] | 8.0 [6.0;10.0] | −0.27 (−0.87, 0.32) | 0.18 (−0.68, 1.03) | −0.45 (−1.24, 0.34) |
| Total Difficulties | 11.0 [7.0;16.0] | 9.0 [5.5;14.0]** | 7.5 [4.1;15.5]** | 12.0 [8.0;17.0]** | 2.82 (1.11, 4.53) | 0.16 (−2.31, 2.62) | 2.66 (0.40, 4.93) |
p < .10
p <.05
p<.001
p<.0001
Models adjusted for child age, race/ethnicity, income-to-needs ratio, cumulative hardship score, and parent education
Includes 16 homeschooled children that were excluded from comparative analyses
SDQ = Strengths and Difficulties Questionnaire; IQR = inter-quartile range
Child Learning-Related Experiences.
Compared to children receiving in-person instruction, parents of children receiving remote instruction were less likely to report that the teacher knew their child well, the child knew their classmates well, was motivated to get ready for school in the morning, motivated to do schoolwork, loves learning, has enough opportunities to socialize, or has a best friend; and was more likely to show defiance in response to schoolwork (Table 3). Compared to hybrid, parents of children receiving remote instruction reported that their children were less motivated to get ready for school and less likely to start a new medication for behavior or attention. Parents of children receiving hybrid instruction were more likely to report that their child started a new therapy outside of school and less likely to report that the teacher understood the child well, the child knew their classmates, has enough socialization opportunities or a best friend, compared to children receiving in-person instruction.
Table 3:
Child Learning-Related Experiences by School Format
| Format of Child’s School Instruction, 2020–2021 | Adjusteda effect estimates or odds ratios (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| Learning-Related Experiences | Overall† (n=305) | In-person (n=80) | Hybrid (n=37) | Remote (n=172) | Remote vs. In-person | Hybrid vs. In-person | Remote vs. Hybrid |
| Teacher understands child well | 208 (71.5%) | 66 (84.6%)* | 24 (64.9%)* | 114 (66.7%)* | 0.37 (0.18, 0.78) | 0.28 (0.11, 0.72) | 1.35 (0.61, 2.97) |
| Child knows classmates well | 179 (62.2%) | 66 (84.6%)*** | 25 (67.6%)*** | 86 (50.6%)*** | 0.19 (0.10, 0.39) | 0.37 (0.14, 0.95) | 0.52 (0.24, 1.15) |
| Child is motivated to get ready for school in the morning | 167 (55.1%) | 57 (73.1%)*** | 28 (75.7%)*** | 73 (42.4%)*** | 0.28 (0.15, 0.52) | 1.10 (0.43, 2.82) | 0.25 (0.11, 0.59) |
| Child is motivated to do schoolwork | 167 (55.3%) | 53 (67.9%)** | 23 (63.9%)** | 80 (46.5%)** | 0.47 (0.26, 0.85) | 0.80 (0.34, 1.89) | 0.58 (0.27, 1.27) |
| Child shows defiance when tasked with schoolwork | 125 (41.4%) | 21 (26.6%)** | 13 (36.1%)** | 82 (48.0%)** | 2.91 (1.56, 5.40) | 1.70 (0.71, 4.06) | 1.71 (0.78, 3.74) |
| Parent and child have been arguing more than usual | 153 (50.7%) | 36 (45.6%) | 18 (48.6%) | 92 (53.8%) | 1.20 (0.67, 2.13) | 1.21 (0.53, 2.75) | 0.99 (0.47, 2.10) |
| Child loves learning | 220 (73.1%) | 65 (83.3%)* | 29 (78.4%)* | 114 (67.1%)* | 0.36 (0.17, 0.76) | 0.51 (0.18, 1.46) | 0.70 (0.29, 1.72) |
| Child has enough opportunities to socialize | 72 (23.7%) | 41 (51.9%)*** | 9 (24.3%)*** | 21 (12.2%)*** | 0.13 (0.06, 0.25) | 0.27 (0.11, 0.68) | 0.47 (0.18, 1.20) |
| Child has a best friend | 177 (59.0%) | 60 (78.9%)*** | 21 (56.8%)*** | 90 (52.6%)*** | 0.30 (0.15, 0.58) | 0.33 (0.14, 0.80) | 0.90 (0.43, 1.90) |
| Child has been engaging in negative self-talk | 67 (22.4%) | 15 (19.0%)+ | 4 (10.8%)+ | 46 (27.5%)+ | 2.02 (0.99, 4.12) | 0.65 (0.19, 2.21) | 3.10 (0.99, 9.74) |
| Child started supplemental therapies (SLT, OT) | 34 (11.2%) | 5 (6.3%) | 6 (16.2%) | 22 (12.8%) | 2.90 (0.92, 9.16) | 4.70 (1.17, 18.9) | 0.62 (0.22, 1.75) |
| Child started seeing behavioral therapist/psychologist | 23 (8.7%) | 8 (10.1%) | 5 (13.5%) | 26 (15.1%) | 1.67 (0.69, 4.07) | 1.65 (0.48, 5.66) | 1.01 (0.35, 2.95) |
| Child started new medication for behavior/learning | 12 (4.3%) | 4 (5.6%)+ | 4 (11.8%)+ | 4 (2.5%)+ | 0.39 (0.08, 1.83) | 3.03 (0.62, 14.9) | 0.13 (0.03, 0.64) |
p < .10
p <.05
p<.001
p<.0001
Models adjusted for child age, race/ethnicity, income-to-needs ratio, cumulative hardship score, and parent education
Includes 16 homeschooled children that were excluded from comparative analyses
SLT = speech-language therapy, OT = occupational therapy
Child Sleep.
Sleep problems were common (Table 4). Compared to before the pandemic, 31.1% of parents endorsed that their child takes longer to fall asleep, 15.4% had more overnight awakenings, and 22% had more nightmares during the COVID-19 pandemic. In multivariable models, children receiving remote instruction were reported to have later sleep midpoint compared to children receiving in-person instruction, and higher odds of co-sleeping compared to children receiving hybrid or in-person instruction.
Table 4:
Child Sleep Characteristics by School Format
| Format of Child’s School Instruction, 2020–2021 | Adjusteda effect estimates or odds ratios (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| Sleep variables | Overall† (n=305) | In-person (n=80) | Hybrid (n=37) | Remote (n=172) | Remote vs. In-person | Hybrid vs. In-person | Remote vs. Hybrid |
| Sleep latency > 30 min | 97 (31.9%) | 20 (25.0%) | 12 (32.4%) | 58 (33.9%) | 1.39 (0.74, 2.62) | 1.57 (0.65, 3.79) | 0.89 (0.40, 1.96) |
| Overnight awakenings (1+ vs none) | 115 (38.5%) | 30 (39.0%) | 11 (29.7%) | 66 (39.1%) | 0.87 (0.48, 1.59) | 0.64 (0.26, 1.56) | 1.36 (0.60, 3.08) |
| Nightmares | 67 (22%) | 13 (16.5%) | 7 (20.0%) | 43 (27.4%) | 1.75 (0.81, 3.75) | 1.64 (0.56, 4.79) | 1.07 (0.41, 2.80) |
| Co-sleeping (ever vs never) | 149 (48.9%) | 32 (40.0%)* | 13 (35.1%)* | 98 (57.0%)* | 1.89 (1.06, 3.37) | 0.84 (0.36, 1.95) | 2.26 (1.04, 4.89) |
| Sleep duration | 10.5 [10.0;11.0] | 10.4 [10.0;10.8] | 10.5 [10.1;11.0] | 10.5 [10.0;11.0] | 0.08 (−0.11, 0.27) | 0.20 (−0.07, 0.47) | −0.12 (−0.37, 0.13) |
| Sleep midpoint | 2.1 [1.6; 2.5] | 1.8 [1.5; 2.1]*** | 2.0 [1.1; 2.6]*** | 2.3 [1.8; 2.7]*** | 0.37 (0.18, 0.56) | 0.14 (−0.13, 0.42) | 0.23 (−0.02, 0.48) |
| Sleep shift | 0.5 [0.0; 1.0] | 0.5 [0.2; 1.0] | 0.5 [0.0; 0.8] | 0.5 [0.2; 1.0] | −0.04 (−0.23, 0.15) | −0.05 (−0.32, 0.23) | 0.004 (−0.25, 0.25) |
p < .10
p <.05
p<.001
p<.0001
Models adjusted for child age, race/ethnicity, income-to-needs ratio, cumulative hardship score, and parent education
Includes 16 homeschooled children that were excluded from comparative analyses
Parent Well-being.
As shown in Table 5, levels of parent depression symptoms were high, as 42.1% of parents met the CES-D depression cutoff score of ≥16; 38.0% of parents whose child attended in-person school, 25.0% of parents whose child attended hybrid school, and 48.5% of parents whose child attended remote school (p<.05 for bivariate difference). When adjusting for covariates, CES-D score ≥16 was not significantly associated with school format (OR 1.35 (95% CI 0.75, 2.44) for remote vs in-person; 0.61 (0.24, 1.53) for hybrid vs in-person, and 2.23 (0.95, 5.21) for remote vs hybrid). Parent stress scores were not associated with school format in bivariate or multivariable models (data not shown).
Table 5:
Parent Wellbeing by School format
| Format of Child’s School Instruction, 2020–2021 | Adjusteda effect estimates or odds ratios (95% CI) | ||||||
|---|---|---|---|---|---|---|---|
| Overall† (n=305) | In-person (n=80) | Hybrid (n=37) | Remote (n=172) | Remote vs. In-person | Hybrid vs. In-person | Remote vs. Hybrid | |
| Parenting Stress Scale | 39.0 [32.5;46.0) | 41.0 [32.5;47.0] | 38.0 [33.0;44.5] | 39.5 [32.5;46.0] | −0.34 (−3.05, 2.38) | −0.78 (−4.69, 3.14) | 0.44 (−3.16, 4.04) |
| CES-D^ >=16 | 127 (42.1%) | 30 (38.0%)* | 9 (25.0%)* | 83 (48.5%)* | 1.35 (0.75, 2.44) | 0.61 (0.24, 1.53) | 2.23 (0.95, 5.21) |
p < .10
p <.05
p<.001
p<.0001
Models adjusted for child age, race/ethnicity, ITN, hardship, parent education
Includes 16 homeschooled children that were excluded from comparative analyses
CES-D = Center for Epidemiologic Studies Depression Scale
Effect Moderation
Effect moderation by child race/ethnicity was found for SDQ Hyperactivity scores, with non-Hispanic white children receiving remote instruction reported to have more hyperactivity (1.41 [95% CI: 0.42, 2.40] vs. children receiving in-person instruction), while associations were non-significant for children from underrepresented minority groups receiving remote instruction (−1.64 [−3.92, 0.64]).
Children receiving remote instruction who did not experience material hardships were reported to have higher SDQ Emotional Symptoms (1.13 [0.17, 2.10] vs. in-person), Hyperactivity (1.43 [0.36, 2.50] vs in-person; 1.83 [0.50, 3.17] vs. hybrid]), and Total Difficulties (4.08 [1.64, 6.51] vs in-person; 4.39 [1.35,7.43] vs. hybrid]), while these associations were non-significant for children experiencing material hardships. Children receiving hybrid instruction with material hardships were reported to have higher SDQ Hyperactivity scores compared to children receiving remote instruction (2.63 [0.14, 5.13]).
DISCUSSION
In this cross-sectional survey of Michigan parents in Winter 2021, elementary school children enrolled in remote learning experienced greater behavioral, learning-related, and sleep difficulties compared with children receiving in-person instruction. Children in hybrid school formats were reported to have more difficulties with socially relevant learning experiences, such as knowing their classmates, compared to children receiving in-person instruction. Similar to previous studies, our findings highlight the multiple challenges that families have faced during the COVID-19 pandemic.
In this sample of school-age children, underrepresented minority children were more likely to attend a remote school format than non-Hispanic white children, as in the CDC survey.6 Prior work has found that Black and Hispanic children have a greater likelihood of low-quality remote instruction or no instruction at all during COVID-19, amounting to 10.3 months of lost learning for Black students vs. 6.0 for white students.24
Child Behavioral Well-being
We found that children receiving remote instruction were reported to have significantly higher SDQ Hyperactivity, Peer Problems and Total Difficulties scores compared to children receiving in-person instruction, and worse Conduct Problems and Total Difficulties compared to children receiving hybrid instruction. These results are similar to those of the CDC survey study, except that the CDC study found that hybrid learners had worse child mental/behavioral health compared with in-person learners,6 which we did not find. Hybrid models vary widely throughout the US, so it is unclear whether these discrepant findings reflect different child educational experiences. Greater behavioral challenges in children receiving remote instruction may be due to difficulties managing the remote platform and curricular workload, children’s tendency to show more dysregulation in home versus school,25 lack of positive peer role modeling, or the general stress experienced by children during home confinement. Parent depression symptoms did not appear to be driving associations between school format and SDQ scores.
Effect moderation analyses found that underrepresented minority children did not exhibit increases in hyperactivity in remote schooling compared to in-person. Underrepresented minority families have experienced heightened burdens this year due to increased public awareness of systemic racism in law enforcement and health systems, along with increased anti-immigrant and anti-Asian violence, which may have more relative impact than school format. Surprisingly, children with material hardships receiving hybrid instruction had more hyperactivity compared to children receiving remote instruction. This may have been due to difficulty transitioning from in-person to remote format; differences in expectations at school versus home; or life disruptions due to challenges in coordinating school, work, and childcare. The lack of associations between school format and other SDQ scores in children with more hardship may also reflect that chronic poverty is associated with more emotional and behavioral challenges at baseline, and therefore an appreciable difference was not detected in the home setting during remote learning. This is an area for further study.
Learning and Academics
Compared to children receiving in-person instruction, children receiving remote instruction were reported to be less motivated and more defiant. Motivation predicts children’s academic achievement in reading and mathematics,26 but is considered to be situational and contextual, rather than a static trait.27 Remote learning might therefore be considered a modifiable contextual influence on some children’s academic motivation, although there is likely inter-individual variation between children.
Although we assessed children’s learning-related experiences with non-validated questions created for this study, concerns about learning have been frequently expressed by parents28 during COVID-19 and can help elucidate mechanisms underlying elevated hyperactivity, as well as peer and conduct problems in children receiving remote instruction. For example, parents of children receiving remote instruction reported that their children were significantly less likely to know their classmates well, feel understood by the teacher, or have a best friend. Social competency is a critical life skill that children develop in their school interactions, while social isolation may fuel underlying mental health concerns.2 In classroom environments, children build a sense of community, identity, independence from parents, model behavior off peers, and respond to tailored classroom accommodations. Research has shown that elementary students who are enthusiastically engaged at the beginning of the academic year are more likely to use adaptive coping skills when faced with academic challenges throughout the school year; whereas dissatisfaction at the beginning of the academic year predicts maladaptive coping skills throughout the year.29 It is worth considering whether children on remote learning will need additional behavioral and social supports upon return to in-person instruction.
Interestingly, parents of children receiving hybrid instruction reported that their children were more likely to start medication for behavior compared to children receiving remote instruction, and more likely to start supplemental therapies outside of school compared to children receiving in-person instruction. One possible explanation for this finding is that parents of children attending school for a portion of the week may have received more feedback from teachers regarding child difficulties and need for intervention. Another possibility is that disruptions associated with transitioning between in-person and remote learning may have further contributed to attention, behavior and/or learning difficulties, for which families may have sought medication management or therapies.
Sleep
Child sleep is another important determinant of behavioral regulation and readiness to engage in learning. Our finding of children experiencing later sleep midpoints during the COVID-19 pandemic is congruent with prior studies.12–14 A study conducted in Germany with 4th graders found that later sleep midpoint was associated with poorer performance on standardized intelligence tests and work avoidance, whereas earlier sleep midpoint was associated with better grades.30 Our findings of prevalent co-sleeping in 5–10-year-olds during COVID-19 overall and higher odds of co-sleeping among children receiving remote instruction are novel. Prior work has established associations between anxiety and greater co-sleeping among school-aged children,31 which may explain this association, as the prevalence of anxiety has doubled among children and adolescents during the COVID-19 pandemic.32
Parent Well-being
We found no differences in parent depression or stress by school format. However, 42% of parents overall reported elevated depression symptoms, which could be due to social isolation3, significant life adjustments during the pandemic, and loss.4 Many parents in our sample reported material hardships, such as poverty, job loss, and food insecurity, which has been described in other cohorts5,11 and associated with more parent-child conflict.3 This information is crucial for informing parent support efforts as the pandemic continues. For example, primary care providers may consider screening parents for mental health concerns and material hardships, with the aim of connecting families in need with community mental health resources and social work assistance.
Limitations of this study include its cross-sectional nature, so causality between school format and parent-child well-being cannot be inferred. Although we attempted to recruit a geographically and racially diverse sample, our convenience sample was not broadly representative of Michigan residents. Two prior studies included nationally representative US samples,6,11 but were conducted earlier during the pandemic and did not include measures of child sleep or learning-related behaviors; however, child sleep and learning-related experiences were not measured using validated scales in our study. Reliance on parental report of child behaviors introduces the potential for single reporter bias. In addition, the SDQ asks parents to reflect on their child’s behavior over the course of the past 6 months, during which time the child may have received different learning modalities intermittently. This could potentially have led parents to report on more generalized traits or the child’s current status rather than reporting their child’s behaviors over the course of the academic year to date. Our school format variable did not quantify the proportion of the school year children received different formats, which leads to less precision in our results. Another limitation is that we combined underrepresented minority children into one category but recognize that children from different races/ethnicities have had different experiences during the pandemic, which deserves further study. Lastly, SDQ Peer Problems and Conduct Problems internal consistency scores were < .70 and should be interpreted with caution.
CONCLUSION
Our findings contribute to a growing body of literature examining remote schooling during the COVID-19 pandemic. As school districts make decisions about school format, they should consider the behavioral, emotional, and sleep-related difficulties young children experience when trying to participate in online instruction and provide appropriate special education services to address residual difficulties. Mechanisms underlying child difficulties during remote learning may include low sense of engagement with the classroom, low motivation, or increased defiance with online learning. Our findings may help clinicians guide parents in obtaining behavioral supports or making decisions regarding returning to in-person instruction.
Supplementary Material
Sources of Funding and Support:
This study was funded by the Towsley-White grant through the Department of Pediatrics, University of Michigan Medical School. REDCap and recruitment support were provided through the Michigan Institute for Clinical & Health Research (CTSA: UL1TR002240). Dr. Levitt received support through a National Institutes of Health (NIH) institutional T32 training grant [5T32HD079350-05; PI Julie C. Lumeng].
Footnotes
Author Disclosure Statement: Dr. Radesky is a paid consultant for Melissa & Doug Toys, Noggin/CBS/Viacom, and the Worldwide Early Development Movement, which are not relevant to the current study. The remaining co-authors have no conflicts of interest relevant to this article to disclose.
REFERENCES
- 1.Donohue JM, Miller E. COVID-19 and school closures. JAMA. 2020;324(9):845–847. [DOI] [PubMed] [Google Scholar]
- 2.Loades ME, Chatburn E, Higson-Sweeney N, et al. Rapid systematic review: the impact of social isolation and loneliness on the mental health of children and adolescents in the context of COVID-19. Journal of the American Academy of Child & Adolescent Psychiatry. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Lee SJ, Ward KP, Lee JY, et al. Parental social isolation and child maltreatment risk during the CoViD-19 pandemic. Journal of family violence. 2021:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Walsh F. Loss and resilience in the time of COVID-19: Meaning making, hope, and transcendence. Family process. 2020;59(3):898–911. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Gassman-Pines A, Ananat EO, Fitz-Henley J. COVID-19 and parent-child psychological well-being. Pediatrics. 2020;146(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Verlenden JV, Pampati S, Rasberry CN, et al. Association of Children’s Mode of School Instruction with Child and Parent Experiences and Well-Being During the COVID-19 Pandemic—COVID Experiences Survey, United States, October 8–November 13, 2020. MMWR Morbidity and Mortality Weekly Report. 2021;70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zhou SJ, Zhang LG, Wang LL, et al. Prevalence and socio-demographic correlates of psychological health problems in Chinese adolescents during the outbreak of COVID-19. European Child & Adolescent Psychiatry. 2020;29(6):749–758. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Xie X, Xue Q, Zhou Y, et al. Mental health status among children in home confinement during the coronavirus disease 2019 outbreak in Hubei Province, China. JAMA pediatrics. 2020;174(9):898–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Brown SM, Doom JR, Lechuga-Peña S, et al. Stress and parenting during the global COVID-19 pandemic. Child abuse & neglect. 2020;110:104699. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brooks SK, Webster RK, Smith LE, et al. The psychological impact of quarantine and how to reduce it: rapid review of the evidence. The lancet. 2020;395(10227):912–920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Patrick SW, Henkhaus LE, Zickafoose JS, et al. Well-being of parents and children during the COVID-19 pandemic: a national survey. Pediatrics. 2020;146(4). [DOI] [PubMed] [Google Scholar]
- 12.Cellini N, Di Giorgio E, Mioni G, et al. Sleep and Psychological Difficulties in Italian School-Age Children During COVID-19 Lockdown. Journal of pediatric psychology. 2021;46(2):153–167. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Oliviero B, Emanuela M, Mattia D, et al. Changes in sleep patterns and disturbances in children and adolescents in Italy during the Covid-19 outbreak. Sleep medicine. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Liu Z, Tang H, Jin Q, et al. Sleep of preschoolers during the coronavirus disease 2019 (COVID-19) outbreak. Journal of sleep research. 2021;30(1):e13142. [DOI] [PubMed] [Google Scholar]
- 15.Karpman M, Zuckerman S, Gonzalez D, et al. The COVID-19 pandemic is straining families’ abilities to afford basic needs. Washington, DC: Urban Institute. 2020;500. [Google Scholar]
- 16.Gravlee C. Systemic racism, chronic health inequities, and COVID-19: A syndemic in the making? American Journal of Human Biology. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Bailey ZD, Krieger N, Agénor M, et al. Structural racism and health inequities in the USA: evidence and interventions. The Lancet. 2017;389(10077):1453–1463. [DOI] [PubMed] [Google Scholar]
- 18.Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)—a metadata-driven methodology and workflow process for providing translational research informatics support. Journal of biomedical informatics. 2009;42(2):377–381. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Thakur N, Hessler D, Koita K, et al. Pediatrics adverse childhood experiences and related life events screener (PEARLS) and health in a safety-net practice. J Child Abuse & Neglect. 2020;108:104685. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.O’Gurek DT, Henke C. A practical approach to screening for social determinants of health. J Family practice management. 2018;25(3):7–12. [PubMed] [Google Scholar]
- 21.Goodman R. The Strengths and Difficulties Questionnaire: a research note. Journal of child psychology and psychiatry. 1997;38(5):581–586. [DOI] [PubMed] [Google Scholar]
- 22.Berry JO, Jones WH. The parental stress scale: Initial psychometric evidence. Journal of Social and Personal Relationships. 1995;12(3):463–472. [Google Scholar]
- 23.Radloff LS. The CES-D scale: A self-report depression scale for research in the general population. Applied psychological measurement. 1977;1(3):385–401. [Google Scholar]
- 24.Dorn E, Hancock B, Sarakatsannis J, et al. COVID-19 and student learning in the United States: The hurt could last a lifetime. McKinsey & Company;2020. [Google Scholar]
- 25.Winterbottom M, Smith S, Hind S, et al. Understanding similarities and differences between parents’ and teachers’ construal of children’s behaviour. Educational Studies. 2008;34(5):483–510. [Google Scholar]
- 26.Broussard SC, Garrison MB. The relationship between classroom motivation and academic achievement in elementary-school-aged children. Family and consumer sciences research journal. 2004;33(2):106–120. [Google Scholar]
- 27.Linnenbrink EA, Pintrich PR. Motivation as an enabler for academic success. School psychology review. 2002;31(3):313–327. [Google Scholar]
- 28.American Academy of Pediatrics. Parental Concern About Children Falling Behind in School During the COVID-19 Pandemic. https://www.aap.org/en/patient-care/family-snapshot-during-the-covid-19-pandemic/parental-concern-about-children-falling-behind-in-school-during-the-covid-19-pandemic/. Updated 10/13/2021. Accessed 11/8/2021.
- 29.Skinner EA, Pitzer JR, Steele JS. Can student engagement serve as a motivational resource for academic coping, persistence, and learning during late elementary and early middle school? Developmental Psychology. 2016;52(12):2099. [DOI] [PubMed] [Google Scholar]
- 30.Arbabi T, Vollmer C, Dörfler T, et al. The influence of chronotype and intelligence on academic achievement in primary school is mediated by conscientiousness, midpoint of sleep and motivation. Chronobiology international. 2015;32(3):349–357. [DOI] [PubMed] [Google Scholar]
- 31.Palmer CA, Clementi MA, Meers JM, et al. Co-Sleeping among School-Aged Anxious and Non-Anxious Children: Associations with Sleep Variability and Timing. Journal of abnormal child psychology. 2018;46(6):1321–1332. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Racine N, McArthur BA, Cooke JE, et al. Global prevalence of depressive and anxiety symptoms in children and adolescents during COVID-19: a meta-analysis. JAMA Pediatrics. 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.