School-level Covariates Research Articles

Cumulative Neurotoxicological Air Pollution Exposure Is Associated with Lower Reading Improvement and Diminished Benefits of Literacy Interventions for Urban Elementary Students of Color.

The cognitive and behavioral deficits associated with air pollution exposure may have far-reaching negative effects on children's scholastic achievement. Moreover, air pollution may be conditioning the success of educational investments that support students who face greatest levels of societal adversity. This study examined the direct main effects of cumulative neurotoxicological exposure on annual reading improvement. We also tested the statistical interaction (i.e., moderation) of neurotoxicological exposure and academic intervention sessions on annual reading improvement for a large sample of ethnic minority (95%) elementary school children (n = 6080, k-6th grade) enrolled in a standard literacy enrichment program. These children were all behind grade level in reading and attended predominantly low-income schools (n = 85) in urban settings across the state of California. Multi-level modeling assessments accounted for random effects associated with school and neighborhood environments, and incorporated extensive individual, school, and community level covariates. Findings show individual elementary students of color to progress less in reading when exposed to greater accumulations of neurotoxin air pollution in their home and school environments, with the average deficit equivalent to 1.5 weeks of learning delay per year. Findings also show neurotoxicological exposure to diminish the efficacy of literacy intervention sessions received on reading improvement throughout the school year. Results suggest that pollution abatement can be a salient strategy to help bridge the child educational achievement gap. In addition to several methodological strengths, this study is one of the first to show that ambient pollution can undermine program efficacy of a literacy enrichment program.

Prescription Stimulant Medical and Nonmedical Use Among US Secondary School Students, 2005 to 2020

Recent information on the prevalence of prescription stimulant therapy for attention-deficit/hyperactivity disorder (ADHD) and nonmedical use of prescription stimulants (NUPS) at the school-level among US secondary school students is limited. To investigate the school-level prevalence of and association between stimulant therapy for ADHD and NUPS among US secondary school students. This cross-sectional study used survey data collected between 2005 and 2020 as part of the Monitoring the Future study (data collected annually via self-administered survey in schools from independent cohorts). Participants were from a nationally representative sample of 3284 US secondary schools. The mean (SD) response rates were 89.5% (1.3%) for 8th-grade students, 87.4% (1.1%) for 10th-grade students, and 81.5% (1.8%) for 12th-grade students. Statistical analysis was performed from July to September 2022. Past-year NUPS. The 3284 schools contained 231 141 US 8th-, 10th-, and 12th-grade students (111 864 [50.8%, weighted] female; 27 234 [11.8%, weighted] Black, 37 400 [16.2%, weighted] Hispanic, 122 661 [53.1%, weighted] White, 43 846 [19.0%, weighted] other race and ethnicity). Across US secondary schools, the past-year prevalence of NUPS ranged from 0% to more than 25%. The adjusted odds of an individual engaging in past-year NUPS were higher at secondary schools with higher proportions of students who reported stimulant therapy for ADHD, after controlling for other individual-level and school-level covariates. Students attending schools with the highest rates of prescription stimulant therapy for ADHD had approximately 36% increased odds of past-year NUPS compared with students attending schools with no medical use of prescription stimulants (adjusted odds ratio, 1.36; 95% CI, 1.20-1.55). Other significant school-level risk factors included schools in more recent cohorts (2015-2020), schools with higher proportions of parents with higher levels of education, schools located in non-Northeastern regions, schools located in suburban areas, schools with higher proportion of White students, and schools with medium levels of binge drinking. In this cross-sectional study of US secondary schools, the prevalence of past-year NUPS varied widely, highlighting the need for schools to assess their own students rather than relying solely on regional, state, or national results. The study offered new evidence of an association between a greater proportion of the student body that uses stimulant therapy and a greater risk for NUPS in schools. The association between greater school-level stimulant therapy for ADHD and other school-level risk factors suggests valuable targets for monitoring, risk-reduction strategies, and preventive efforts to reduce NUPS.

School greenspace is associated with enhanced benefits of academic interventions on annual reading improvement for children of color in California

The restorative benefits of nature on children's cognition suggest the potential for nature exposure to further enhance the impacts of instruction on academic achievement. However, prior research has focused primarily on examining main effects of greenspace on academic performance, and have seldom examined interactions (i.e., moderation) between greenspace and other school variables on academic performance. We examined the statistical interaction of school greenspace (based on 1 m by 1 m land-use classification) and a standard literacy enrichment program on reading improvement over a one-year period for a large sample of ethnic minority (95%) elementary school children (n = 6080, k-6th grade, estimated mean age = 7.9 years) attending predominantly low-income schools (n = 85) throughout the state of California. Our analyses employed multi-level modeling to account for random effects associated with school and neighborhood environments, and incorporated extensive individual, school, and community level covariates. Results from the hypothesized interaction effects demonstrate school greenspace to accentuate the positive association between literacy interventions and improved test scores. Moreover, specific components of greenspace, namely herbaceous and tree cover each enhanced the effect of academic instruction on annual reading improvement. However, no direct association between greenspace cover and academic improvement was found. In addition to several methodological strengths, this study is one of the first to show the interaction between nature and school instruction affecting academic learning.

The impact of the COVID-19 pandemic on adolescent mental health: a natural experiment.

Despite widespread concern about the impact of COVID-19 on adolescent mental health, there remains limited empirical evidence that can causally attribute changes to the pandemic. The current study aimed to overcome existing methodological limitations by exploiting a serendipitously occurring natural experiment within two ongoing, multi-phase cluster randomized controlled trials. Depressive symptoms (primary outcome), externalizing difficulties and life satisfaction (secondary outcomes) were assessed at baseline (phase 1 [pre-COVID-19 group]: September – October 2018, phase 2 [COVID-19 group]: September – October 2019) and 1-year follow-up (pre-COVID-19 group: January – March 2020, COVID-19 group: February – April 2021). Participants in phase 1 (N = 6419) acted as controls. In phase 2, participants (N = 5031) were exposed to the COVID-19 pandemic between the baseline and follow-up assessments providing a natural experimental design. The primary analysis used a random intercept linear multivariable regression model with phase (exposure to the COVID-19 pandemic) included as the key predictor while controlling for baseline scores and individual and school-level covariates. Depressive symptoms were higher and life satisfaction scores lower in the group exposed to the COVID-19 pandemic. Had the COVID-19 pandemic not occurred, we estimate that there would be 6% fewer adolescents with high depressive symptoms. No effect of exposure to the pandemic on externalizing difficulties was found. Exploratory analyses to examine subgroup differences in impacts suggest that the negative impact of the COVID-19 pandemic on adolescent mental health may have been greater for females than males. Given the widespread concern over rising adolescent mental health difficulties prior to the pandemic, this paper quantifies the additional impacts of the pandemic. A properly resourced, multi-level, multi-sector public health approach for improving adolescent mental health is necessary. Following in-principle acceptance, the approved Stage 1 version of this manuscript was preregistered on the OSF at https://doi.org/10.17605/OSF.IO/B25DH. This preregistration was performed prior to data analysis.

Bayesian multilevel multidimensional item response modeling approach for multiple latent variables in a hierarchical structure

In this paper, we propose a multilevel multidimensional item response model for studying the relations among multiple abilities and covariates in a hierarchical data structure. As an example, this study is well suited to examining the scenario in which a test measures multidimensional latent traits (e.g., reading ability, cognitive ability, and computing ability) and in which students are nested within classes or schools. The new model can recover the correlations among multidimensional abilities, along with the correlation between person- and school-level covariates and abilities. A fully Gibbs sampling algorithm within the Markov chain Monte Carlo (MCMC) framework is proposed for parameter estimation. A unique form of the deviance information criterion (DIC) is used as a model comparison index. Two simulation studies show that the estimation method is suitable in recovering all model parameters.

Are closed campus policies associated with adolescent eating behaviours?

The effectiveness of school nutrition regulations may be undermined by food environments surrounding schools. Given challenges in regulating external retail, some have recommended policies that ensure students are unable to leave school property during the day (closed campus policies; CCP). We aimed to examine whether CCP are associated with student eating behaviours. We used student and school-administrator survey data from the 60 610 Grades 9 to 12 students and 134 Canadian secondary schools that participated in Year 7 (2018/19) of the COMPASS study. Multiple ordinal regression models tested school CCP as a predictor of weekday dietary behaviours (0-5 days), controlling for student-level (grade, sex, spending money, ethnicity) and school-level (urbanicity, province, area median household income, vending machines) covariates. CCP were reported by 16 schools. Students who attended CCP schools reported eating lunch purchased from fast food outlets or other restaurants and drinking sugar-sweetened beverages (SSBs; soft drinks and sports drinks; sweetened coffee or tea drinks) on fewer weekdays, but consumed snacks from school vending machines on more weekdays, relative to students at open campus schools. No significant differences were observed in student reports of eating home-packed or school cafeteria lunches or snacks purchased off-campus. CCP may help improve adolescent diets by reducing SSB and lunchtime fast food consumption on weekdays; however, students already purchasing food may shift from off-campus to within-school options, highlighting the importance of ensuring healthy school food environments and encouraging students to bring home-prepared lunches. Future studies using experimental longitudinal designs are needed to determine the effect of CCP on various health behaviours and outcomes.

Impact of Korean students’ individual learning time on math performance: differential effect of teachers’ assessment competency

As the 2015 revised national curriculum is being implementing in Korea, a teacher’s assessment competency has become a point of focus for the potential enhancement of students’ self-directed learning. Relevantly, a teacher’s competency in student assessment is correlated to his/her individual feedback practices. Although students’ self-directed learning affects academic performance and may vary based on their teachers’ assessment competency, few studies have investigated the effect of a teacher’s assessment competency, specially feedback practices, on student outcomes. Therefore, this study aims to empirically investigate the differential effect of a teacher’s assessment competency, particularly focusing on feedback practices (i.e., formative assessment feedback, homework feedback), on Korean students’ individual learning times and math achievements. For the analysis, a 3-level hierarchical linear modeling (HLM) is employed using data of middle school students (N = 2942) and math teachers (N = 426) in 207 middle schools of the Gyeonggi Education Panel Study (GEPS). The study revealed that a teacher’s formative assessment feedback has a positive differential effect on math performance when controlling for student-, teacher-, and school-level covariates. In addition, the effect of individual learning time on math achievement was significantly dependent on teachers’ homework feedback. The study provides newfound empirical evidence that teacher feedback plays a pivotal role in promoting a student’s self-directed learning and academic achievement. Based on these findings, the implications of improving a teacher’s assessment competency are discussed.

Are School Substance Use Policy Violation Disciplinary Consequences Associated with Student Engagement in Cannabis?

Schools are increasingly concerned about student cannabis use with the recent legalization in Canada; however, little is known about how to effectively intervene when students violate school substance use policies. The purpose of this study is to assess the disciplinary approaches present in secondary schools prior to cannabis legalization and examine associations with youth cannabis use. This study used Year 6 (2017/2018) data from the COMPASS (Cannabis use, Obesity, Mental Health, Physical Activity, Alcohol use, Smoking, Sedentary behavior) study including 66,434 students in grades 9 through 12 and the 122 secondary schools they attend in British Columbia, Alberta, Ontario, and Quebec. Student questionnaires assessed youth cannabis use and school administrator surveys assessed potential use of 14 cannabis use policy violation disciplinary consequences through a (“check all that apply”) question. Regression models tested the association between school disciplinary approaches and student cannabis use with student- (grade, sex, ethnicity, tobacco use, binge drinking) and school-level covariates (province, school area household median income). For first-offence violations of school cannabis policies, the vast majority of schools selected confiscating the product (93%), informing parents (93%), alerting police (80%), and suspending students from school (85%), among their disciplinary response options. Few schools indicated requiring students to help around the school (5%), issuing a fine (7%), or assigning additional class work (8%) as potential consequences. The mean number of total first-offence consequences selected by schools was 7.23 (SD = 2.14). Overall, 92% of schools reported always using a progressive disciplinary approach in which sanctions get stronger with subsequent violations. Students were less likely to report current cannabis use if they attended schools that indicated assigning additional class work (OR 0.57, 95% CI (0.38, 0.84)) or alerting the police (OR 0.81, 95% CI (0.67, 0.98)) among their potential first-offence consequences, or reported always using the progressive discipline approach (OR 0.77, 95% CI (0.62, 0.96)) for subsequent cannabis policy violations. In conclusion, results reveal the school disciplinary context in regard to cannabis policy violations in the year immediately preceding legalization. Various consequences for cannabis policy violations were being used by schools, yet negligible association resulted between the type of first-offence consequences included in a school’s range of disciplinary approaches and student cannabis use.

Evaluating the Implementation and Effectiveness of the SWITCH-MS: An Ecological, Multi-Component Adolescent Obesity Prevention Intervention.

Background: The purpose of this study was to evaluate the implementation and effectiveness of an ecological, multi-component adolescent obesity prevention intervention called School Wellness Integration Targeting Child Health–Middle School (SWITCH–MS). Methods: Following the effectiveness-implementation hybrid type 3 quasi-experimental design, seven middle schools (377 students) in Iowa, United States, were stratified into “experienced” (n = 3; 110 students) or “inexperienced” (n = 4; 267 students) groups to receive the 12-week SWITCH–MS intervention. To evaluate implementation, school informants (n = 10) responded to a survey and students completed behavioral tracking in the classroom on a website. For effectiveness evaluation, students in 6th, 7th, and 8th grades completed a validated questionnaire before and after intervention, to measure behaviors of physical activity (PA; “Do”), screen-based activity (“View”), and fruits and vegetable consumption (“Chew”). Results: The two groups of schools showed similar levels of implementation for best practices, awareness, and engagement. Behavioral tracking rate favored the experienced schools early on (47.5% vs. 11.7%), but differences leveled off in weeks 3–12 (sustained at 30.1–44.3%). Linear mixed models demonstrated significant time effects for “Do” (at school and out of school; p < 0.01) and “View” behaviors (p = 0.02), after controlling for student- and school-level covariates. Conclusions: This study demonstrates that prior experience with SWITCH–MS may not be a prominent factor for implementation and effectiveness, although greater experience is associated with favorable behavioral tracking when the intervention is first launched.

School-Level Prevalence and Predictors of e-Cigarette Use in 8th, 10th, and 12th Grade U.S. Youth: Results From a National Survey (2015–2016)

PurposeThe aim of this study was to examine the school-level prevalence and predictors of e-cigarette use among U.S. adolescents. MethodsThe Monitoring the Future study is an annual national cross-sectional study of secondary (middle and high) school students. This study uses 2015 and 2016 survey data from 8th, 10th, and 12th grade students (n = 38,926) attending 580 U.S. public and private secondary schools. E-cigarette use, binge drinking, cigarette smoking, marijuana use, and nonmedical prescription drug use were measured at the individual level. School-level characteristics were measured with both aggregated student-level measures (e.g., school-level e-cigarette use, school-level racial composition) and school characteristics (e.g., public vs. private, urban vs. rural). ResultsThe prevalence of past-month e-cigarette use varied considerably across 580 U.S. middle and high schools, ranging from 0% to 60% at individual schools (mean = 10.2%, standard deviation = 8.9%). Multivariable regression analyses indicated that past-month e-cigarette use was significantly higher at schools with a higher proportion of White students, schools that were located in the U.S. Southern and Western regions, and schools with higher prevalence of past-month cigarette smoking, after controlling for relevant individual- and school-level covariates. ConclusionsE-cigarette use varies considerably across schools, and this study suggests that school context plays an important role in e-cigarette use. Cigarette smoking and e-cigarette use at the school level are closely linked and are important risk factors for individual-level e-cigarette use. The wide variation in prevalence of e-cigarette use highlights the need for schools to work with professionals to obtain school-level assessment rather than relying solely on state or national results to guide prevention efforts.

Long-term effects of early childhood programs through eighth grade: Do the effects fade out or grow?

Support for policies to improve early childhood educational development and reduce disparities grew rapidly this century but recently has wavered because of findings that program effects might fade out prematurely. Two programs implemented at scale in North Carolina (Smart Start and More at Four) have been associated with academic success early in elementary school, but it is not known whether these effects fade out or are sustained in middle school. Smart Start provides state funding to support high-quality early childcare in local communities, and More at Four provides state-funded slots for a year of credentialed pre-kindergarten. Funds were allocated for each program at varying rates across counties and years. We used this variation to estimate the long-term impact of each program through eighth grade, by measuring the association between state funding allocations to each program, in each of 100 counties over each of 13 consecutive years, and later student performance. Students were matched to funding levels provided to their home county in their early childhood years and then followed through eighth grade. Analyses using county- and year-fixed-effects regression models with individual- and school-level covariates conducted on nearly 900,000 middle school students indicate significant positive impacts of funding for each program on reading and math test scores and reductions in special education placement and grade retention. These impacts do not fade out and seem instead to grow (for More at Four) as students progress through middle school. Students from economically disadvantaged backgrounds experience particularly large benefits from the More at Four Program.

A false promise? Decentralization in education systems across the globe

This paper seeks to empirically evaluate the association of decentralization of education systems and student achievement. It employs PISA (2015) data to perform multilevel analysis for each and every participating country. Decentralization effects are estimated on the imputed scores of mathematics after controlling for various studentlevel and school-level covariates. Findings reveal an important pattern across the analyzed countries. The random intercepts models suggest that decentralization has no significant effect on student achievement. The results of this paper challenge the primary notions of advocating for decentralization in education systems for improving student outcomes. Based on the empirical evidence from a multitude of countries, this paper argues that decentralization could be more of a false promise and an ineffective prescription.

Effectiveness of a gamification strategy to prevent childhood obesity in schools: a cluster-randomized controlled trial

AbstractIntroductionSchool-based interventions to prevent obesity have shown heterogeneous results. Recent school-based trials with “negative” have cast doubt on their effectiveness. In the Juntos Santiago trial, we used an innovative, theory-based gamification strategy to increase motivation and participation in healthy behavioural changes. This is the first trial, to our knowledge, to use gamification to prevent childhood obesity.ObjectiveTo examine the effectiveness of a multicomponent gamification strategy to prevent obesity in 5th and 6th-grade schoolchildren in Santiago de Chile.Materials and methodsSchool-based, parallel cluster-randomized controlled trial. 81 schools in two municipalities in Santiago, Chile, with more than 40 students in 5th and 6th grade altogether were eligible. Schools were randomized and sequentially invited to participate. The multicomponent intervention was a gamification strategy with four components: (i) healthy challenges (snacks, steps and healthy activities); (ii) gamification incentives, such as points, levels, leaderboards and badges; (iii) rewards (infrastructure and fun activity) and (iv) an online platform for parents and students to monitor the progress. Participants were followed up at 5 and 8 months. Pre-registered primary outcomes were body mass index (BMI) z-score and waist circumference (WC) at 8 months. Secondary outcomes were BMI and systolic (SBP) and diastolic blood pressure (DBP) at 8 months. Multilevel analysis adjusted for individual and school-level covariates were used.Results24 schools (9 controls) and 2333 students (709 controls) were assessed at baseline and 2264 students (689 controls) were assessed at 8 months. The mean BMI z-score was lower in the intervention group compared to the control group at 8 months (mean difference -0.132, 95% CI -0.249; -0.014), while no difference was observed for WC. Mean BMI and SBP were lower in the intervention arm compared to the control arm (mean difference for SBP -0.139; 95% CI -2.39; -0.40). No difference was observed for DBP.DiscussionThe gamification strategy appears to prevent childhood obesity and reduce systolic blood pressure in school children in Santiago. These effects are larger than the pooled effects from existing meta-analyses. Further research should identify which components of the gamification strategy were more effective.Trial registrationClinicalTrials.gov NCT03459742Funded byMayors Challenge 2016, Bloomberg Philanthropies

Elementary English Learner Classroom Composition and Academic Achievement: The Role of Classroom-Level Segregation, Number of English Proficiency Levels, and Opportunity to Learn

Using mixed methods, we investigated (a) the association of the extent of English learner (EL) classroom-level segregation (proportion EL) and number of EL English proficiency levels with elementary EL academic achievement, using 2 years of administrative data, and (b) school staff–reported opportunity to learn–related advantages and disadvantages in segregated versus integrated compositions, using 3 years of interviews. Findings were corroborative across methods. After accounting for student-, classroom-, and school-level covariates, we found that ELs in more segregated classrooms exhibited lower performance, on average, on state tests of English language arts, mathematics, and English proficiency, and little evidence that classroom number of EL English proficiency levels was related to achievement. School staff consistently detailed the instructional, academic, and socio-emotional opportunities to learn afforded by the diversity/heterogeneity of integrated classrooms.

33. Tobacco Retail Environment And Alternative Tobacco Product Use Among Teens

The rise of non-cigarette, alternative tobacco product (ATP) use among adolescents may be due to an increase in retail availability of ATPs. This is the first study to examine the relationship of tobacco retailer density and proximity to any ATP use, assess many ATPs simultaneously, and assess this relationship longitudinally. We examine whether the proximity and density of stores near students’ home is associated with a higher likelihood of initiating ATP use over time including e-cigarettes, small and large cigars, chewing or dipping tobacco or moist snuff, tobacco pipes, and hookah. This geospatial analysis of a multilevel, longitudinal data set linked survey data from 728 adolescents (ages 13-19) nested in 191 neighborhoods nested in 10 California high schools with location data for tobacco retailers from a state licensing list. At the individual level, proximity was measured as roadway distance from home to the nearest tobacco retailer. At the census tract level, tobacco retailer density measured tobacco retailers per square mile. Other tract-level and school-level covariates were obtained from the American Community Survey and the California Department of Education. To account for the fact that students who lived in the same census tracts attended different high schools, cross-classified multilevel logistic regression models were used to estimate odds ratios (OR) and 95% confidence intervals (CI) for proximity to a tobacco retailer and tract density (at baseline) with ATP initiation at follow-up. The sample was predominantly female (63.5%) and was racially and ethnically diverse. Approximately one-third of participants (32.5%) reported ever ATP use at Wave 1 and 106 (14.5%) initiated ATP use within one year (Wave 2). The mean number of tobacco retailers per square mile within a tract was 5.66 (SD=6.3) and the average distance from each participant’s residence to the nearest tobacco retailer was 0.61 miles (SD=0.4). In fully adjusted two-level multilevel models (MLM) adjusting for demographic, socioeconomic, and behavioral factors, higher odds of initiating ATP use were associated with living in tracts with higher retail density (OR=1.32, 95% CI= 1.21, 3.81) in the neighborhood-only model and 1.06 (1.03, 1.92) in the school-only model. After adjusting for covariates, greater tobacco retailer density was still associated with higher odds of ATP initiation (AOR=1.22, 95% CI=1.07, 2.12). Neighborhood-level and school-level variations were 4% and 3%, respectively. Only retail density, not proximity, was related to ATP initiation. Our findings show that tobacco retailers clustered around students’ home neighborhood may be an influence on adolescents’ ATP use. These findings suggest the importance of regulating the tobacco retail environment, in order to reduce the marketing and accessibility of ATPs, which can thereby reduce the uptake of ATPs among youth. Moreover, these findings suggest the importance regulations focused on adolescents’ school environment are expanded to also include regulations of the tobacco retail environment in adolescents’ residential neighborhoods. Policy efforts to reduce adolescent ATP use should limit and reduce the density of tobacco retailers.

Significant predictors of mathematical literacy for top-tiered countries/economies, Canada, and the United States on PISA 2012: Case for the sparse regression model.

National ranking from the triennial Programme of International Student Assessment (PISA) often serves as a barometer of national performance and human capital. Though excessive student- and school-level covariates (n>700) may prove intractable for traditional least-squares estimate procedures, shrinkage methods may be more suitable for subset selection. With a focus on the United States, this paper proposes sparse regression for PISA 2012 to discover salient student- and school-level predictor variables for mathematical literacy achievement. The sparse regression analysis was conducted on 10 top-tiered OECD countries/economies, Canada, and the United States in mathematical literacy on the 2012 PISA. Two- and three-level hierarchical regression analyses were performed on Canadian and US students (N=26,522) along with five of the ten top-tiered countries/economies (N=58,385). Using the 'least absolute shrinkage and selection operator' (LASSO) technique, the study (1) identified salient predictor variables of mathematical literacy performance for the top-tiered countries/economies, Canada, and the United States and (2) used these salient variables to perform two- and three-level hierarchical regression on data from Canada and the United States along with five top-tiered countries/economies. Weights and replicates were used to account for complex sample design. A weighted, two-level confirmatory factor analysis was performed to identify latent constructs. Missing data were handled through multiple imputation. Separate two-level hierarchical models accounted for 32-35% student-level and 58-70% school-level variance in Canada and the United States, respectively; three-level models accounted for 33% of level-one variance, 62-65% level-two variance, and 13-44% of level-three variance for the US/Canada and US/Canada/top-tiered students, respectively. Following top-tiered countries/economies, Canadian students had high levels of self-efficacy, were more likely to encounter advanced concepts in class, were less activity/small group-centred, and were more likely to consider truancy a learning hindrance. Factor analyses revealed a positive relation with rigour and class organization (teacher-centred) for top-tiered countries and Canada, though not for the United States. For all countries, there was a strong relation between rigour and self-beliefs. Compared to top performers, a less rigorous curriculum, coupled with class and school factors, may explain lag in US performance.

School start time changes in the COMPASS study: associations with youth sleep duration, physical activity, and screen time

BackgroundTo date, no longitudinal population-based studies of school start times have been conducted within Canada. School schedule changes provided an opportunity to examine start times in association with youth sleep, physical activity, and screen use over time. MethodsThis longitudinal study included grade 9–12 students attending 49 Ontario secondary schools that participated in at least two consecutive years of the COMPASS study (2012–2017). Fixed effects models tested whether differences in within-student change in self-reported sleep duration, moderate-to-vigorous physical activity, and screen time were associated with school start time changes, adjusting for student- (grade, sex, ethnicity, spending money) and school-level covariates (median income, urbanicity, geographical area). ResultsThirteen start time changes of 5–10 min were reported. Ten-minute advances at earlier clock times (8:30 AM–8:20 AM; 8:40 AM–8:30 AM) were associated with steeper sleep duration declines than schools with consistent start times but had no effect at later times (9:00 AM–8:50 AM). While sleep change did not differ with 5-min delays, 10-min delays (8:50 AM–9:00 AM) were associated with additional sleep (23.7 min). Apart from one school that shifted from 8:30 AM to 8:35 AM, in which screen time and physical activity decreased more steeply, no effect was found for screen time, and 5-min delays were associated with more physical activity (10.9 min) and advances with less activity (−8.0 min). ConclusionsResults support start time delays as a valuable strategy to help ameliorate sleep debt among youth. Interference with physical activity or increased screen time appear unlikely with modest schedule changes. Potential adverse impacts on sleep require consideration with 10-min advances.

Why Are Single-Sex Schools Successful?

We exploit two unusual policy features of academic high schools in Seoul, South Korea—random assignment of pupils to high schools within districts and conversion of some existing single-sex schools to the coeducational (coed) type over time—to identify three distinct causal parameters: the between-school effect of attending a coed (versus a single-sex) school; the within-school effect of school-type conversion, conditional on (unobserved) school characteristics; and the effect of class-level exposure to mixed-gender (versus same-sex) peers. We find robust evidence that pupils in single-sex schools outperform their counterparts in coed schools, which could be due to single-sex peers in school and classroom, or unobservable school-level covariates. Focusing on switching schools, we find that the conversion of the pupil gender type from single-sex to coed leads to worse academic outcomes for both boys and girls, conditional on school fixed effects and time-varying observables. While for boys, the negative effect is largely driven by exposure to mixed-gender peers at school-level, it is class-level exposure to mixed-gender peers that explains this disadvantage for girls.

The relationship among school safety, school satisfaction, and students' cigarette smoking: Based on a multilevel mediation model

BackgroundSchool context greatly affects students' cigarette smoking. Based on the ‘context–process–outcomes’ model, this study investigated school safety as a context variable to explore how safety (at the school level) affects students' smoking behaviours. MethodsHierarchical linear modelling was used to examine the link between school safety at the school level and students' cigarette smoking, with school satisfaction as a mediator. Data were obtained from the National Children's Study of China (2009), where 24,013 fourth- to ninth-grade students from 607 schools were recruited from 100 counties in 31 provinces (boys 52.9%, elementary school students 50.1%, middle school students 49.9%). ResultsStudent-level and school-level factors were associated with an increased likelihood of cigarette smoking. Multilevel mediation analyses showed that school safety problems increased the behaviour of cigarette smoking directly as well as indirectly through student's school satisfaction, controlling for both student- and school-level covariates; its direct effect was larger than its indirect effect. A gender difference also existed in the multilevel mediation model: for girls, school satisfaction fully mediated the relationship between school safety at the school level and cigarette smoking, but only partially for boys. There was no grade-stage difference in the multilevel mediation model. ConclusionsLimitations of the study as well as implications are discussed. School safety should receive increased attention from policymakers because of its impact on students' cigarette smoking behaviour. The construction of a safe school climate should be approached dynamically and multi-perspectivally.

Kindergarten standardized testing and reading achievement in the U.S.: Evidence from the early childhood longitudinal study

Drawing from data use theory (i.e., a theory for making data-driven educational decisions), the present study sought to understand how frequency of standardized testing is related to student learning, mediated by reading instruction, after controlling for child-level (e.g., gender, race/ethnicity) and school-level covariates (e.g., private/public, proportion of students eligible for free lunch). Using data from the Early Childhood Longitudinal Study Kindergarten Cohort of 2010–2011, the sample included 12,241 children attending 1067 kindergartens in the U.S. findings from a multilevel structural equation mediation model suggest that the frequency of state/local standardized testing in kindergarten did not have a direct effect on reading achievement near the end of kindergarten, after controlling for covariates. However, the amount and type of reading instruction mediated the relationship between the frequency of testing and reading achievement, after controlling for covariates. The implications for policy and practice on the use of standardized tests in kindergarten are discussed.