Frequency of Adequate Sleep and Different Forms of Delinquency Across Adolescence: Evidence From Nationally Representative Samples of Youth

Abstract

An emerging body of research documents a relationship between sleep quantity and delinquency during adolescence. Absent from this line of research, however, is an evaluation of whether the associations between sleep duration and different forms of delinquency vary across periods of adolescence and sex. The current study aimed to address this gap in the body of literature by analyzing data from nationally representative samples of 8th, 10th, and 12th graders from the 2010 to 2019 Monitoring the Future (MTF) surveys. Multivariate regression models reveal that relations between getting less than 7 hours of sleep and violent and non-violent delinquent behavior vary across grade level. Specifically, associations between less sleep and forms of delinquency are stronger for males in grades 8 and 10 compared to males in grade 12 or females. Implications of the results for future research and policies designed to increase sleep quantity during adolescent development are discussed.

Keywords

inadequate sleep delinquency adolescence monitoring the future

Introduction

According to the National Sleep Foundation, over two-thirds of high school students in the United States do not meet the recommended 8 hours of sleep each night (Sleep for Teenagers, 2022). Not meeting this metric on a continual basis may have detrimental effects for one’s physical and mental wellbeing, especially during adolescence (Carskadon et al., 1990; Daly et al., 2015; Haack & Mullington, 2005; Rossa et al., 2014). Studies are beginning to show that when adolescents do not receive the recommended hours of sleep, they can experience higher levels of anxiety, anger, and confusion (Baum et al., 2014). Insufficient sleep during this developmental timeframe can also adversely affect brain development, which can contribute to increased impulsivity, susceptibly to peer pressure, and lower levels of self-control (Telzer et al., 2015). Other studies report that less sleep is related to higher levels of delinquent behavior (Clinkinbeard et al., 2011; Meldrum & Restivo, 2014) with some providing evidence of a non-linear relation (Mears et al., 2022).

While there is a growing body of research focused on examining the links between sleep duration and delinquency (Mears et al., 2022; Meldrum et al., 2015; Peach & Gaultney, 2013; Vazsonyi et al., 2018), much remains unknown about whether sleep is uniquely, or commonly, associated with different forms of delinquency during adolescence, and whether these associations vary by sex. Given that sleep needs and delinquent involvement vary across age—even during the short period of adolescence (Meijer et al., 2010; Meldrum et al., 2018)—and sex (Farrington, 1986; Le Blanc, 2020), it is important to explore if these associations change during adolescent development to create tailored programs and policies aimed at helping adolescents at specific ages to improve their sleep. This can reduce the risk of delinquent involvement. The current study aims to begin to address these important gaps in the existing body of research by analyzing data from multiple nationally representative samples of youth from the United States.

Inadequate Sleep Quantity and Delinquency

Previous studies examining the link between sleep quantity and delinquency have often measured delinquent offending with a variety measure of delinquency, including items capturing the prevalence and/or frequency of a combination of violent and nonviolent delinquent acts (Kamphuis et al., 2012; Lin & Yi, 2015; Mears et al., 2022; Vazsonyi et al., 2018). The results from these studies reveal that less sleep is related to higher levels of overall delinquency. For example, Lin and Yi (2015) report that adolescents from Taiwan who slept less than 6 hours a night reported higher levels of conduct problems—which the authors defined as a combination of substance abuse, fighting, and skipping class—than their peers who reported getting more than 6 hours of sleep a night. Vazsonyi et al. (2018) combined nonviolent and violent delinquent behaviors (i.e., vandalism and assault) into a broad measure of deviance and found that adolescents who slept for shorter periods of time reported more deviant behavior. Mears et al. (2022) investigated the non-linear association between sleep duration and delinquent behavior and found that sleep duration and delinquency demonstrated a U-shaped curve, such that getting less than 6 hours or more than ten hours of sleep was associated with more delinquency, compared to getting 9 hours of sleep. Other factors associated with increased delinquency in these past studies include being male, higher impulsivity, low self-control, identifying as a member of a minority racial group, poorer academic performance, low parental education, and low socioeconomic status.

While combining violent and nonviolent behaviors into a variety index of offending can help researchers evaluate the overall strength of association between sleep and delinquency, it makes it difficult to understand if violent versus nonviolent delinquency is more strongly associated with less sleep during adolescence. Despite research suggesting that adolescents who commit violent acts tend to be generalists, such that they also commit nonviolent delinquent offenses (Capaldi & Patterson, 1993; MacDonald et al., 2009), a large proportion of delinquent adolescents only engage in nonviolent delinquency, not violent delinquency (Henry et al., 1996; Lai et al., 2016; MacDonald et al., 2009). If sleep duration varies between these two groups of adolescent offenders and between those who engage in delinquency and those who do not, then the nature of the relationship between sleep and delinquency may be fundamentally different.

Another reason to suspect that sleep quantity might be differentially associated with violent and nonviolent delinquency is because of the effect of sleep on the adolescent brain. Specifically, insufficient sleep is associated with deficiencies in frontal lobe functioning (Gosselin et al., 2005). The frontal lobe is responsible for various cognitive processes including decision-making, reasoning, and self-control (Chayer & Freedman, 2001). Thus, when adolescents do not receive sufficient sleep, they can experience lower self-control (Connolly et al., 2022; Meldrum et al., 2015) and increased impulsivity (Forzano et al., 2011). Violent delinquent behaviors are often committed out of impulse resulting from low self-control (Scarpa & Raine, 2000) whereas many nonviolent delinquent acts, such as theft or property damage, are more likely to require some level of planning. Therefore, it is possible that insufficient sleep might have a stronger effect on violent delinquency, relative to nonviolent delinquency.

To date, there have only been a few studies that have assessed the association between sleep quantity and specific measures of violence in adolescent samples. These studies find that adolescents who do not get enough sleep are more likely to engage in physical fights and carry weapons (Hildenbrand et al., 2013; Meijer et al., 2010; Meldrum & Restivo, 2014). Although these studies assess the association between sleep quantity and violence, the specific forms of violent delinquent behavior are limited to fighting and weapon carrying. Carrying a weapon, which is officially defined as a violent offence when in the hands of an adolescent, might not actually result in weapon use, or the perpetration of other forms of violence, that could injure another person. Based on this limited area of empirical inquiry, additional research on more forms of violence is needed to better understand the link between sleep quantity and the perpetration of violent delinquent behavior.

Similar to research focused on violent delinquency, few studies have specifically examined the relationship between sleep quantity and nonviolent delinquency in adolescent populations. Clinkinbeard et al. (2011) found that youth who typically slept seven or fewer hours a night, had higher levels of self-reported property delinquency, including stealing and damaging property, relative to those who slept an average of eight or more hours a night. In contrast, Barnes and Meldrum (2015) found that getting fewer hours of sleep was not associated with nonviolent delinquency, such as stealing and lying. Given the limited amount of research and considering the mixed results, additional research is needed to better understand the link between sleep quantity and nonviolent delinquency as well.

Sleep Duration and Violent and Nonviolent Delinquency Across Adolescence and by Sex

Another gap in this body of literature is that much remains unknown about whether the relationships between sleep quantity and different forms of delinquent behavior change throughout adolescence. Previous studies have either combined multiple years of age together (Jackson et al., 2020; Lin & Yi, 2015; Mears et al., 2022; Meldrum & Restivo, 2014), or only included participants who are the same age (Meldrum et al., 2015). The problem with these methods is that both sleep patterns (Meldrum et al., 2018) and delinquency (Farrington, 1986; Le Blanc, 2020) change during adolescence. More specifically, sleep quantity tends to decline from early to late adolescence and delinquency tends to rise until age 17 years and then decline precipitously. The association between sleep and delinquent behavior may also range in magnitude across adolescence. One reason why sleep may be differentially associated with delinquency at different time points throughout adolescence is that the opportunities to engage in delinquency generally increase as children grow older. Relative to younger adolescents, older adolescents tend to spend less time at home with their parents and more time with their peers, decreasing adult supervision and potentially increasing opportunities for delinquent offending.

While much is unknown about how the association between sleep quantity and delinquency might change from year to year during adolescence, a few studies have begun to explore how this association changes from childhood to adulthood. Peach and Gaultney (2013) analyzed a sample of students in the United States and found that decreased sleep duration was directly associated with increased delinquent behavior. However, the relationship was no longer statistically significant once participants reached early adulthood. Raine and Venables (2017) found that 15-year-olds who reported inadequate sleep were more likely to engage in violent and non-violent delinquent behaviors at ages 15 and 29 years compared to their peers who reported getting more sleep at age 15 years. Taken together, it appears that the relationship between sleep and delinquency, including violent delinquency, might not remain static throughout the life-course.

There is also reason to suspect that the association between sleep quantity and delinquency may vary across sex. Results from past research indicate that young males and females experience different sleep patterns and rates of delinquency. Females are more likely to report experiencing sleep problems—including lower sleep quantity—than males (Mong & Cusmano, 2016). Females are also less likely to engage in both violent and nonviolent delinquent behavior than males (Junger-Tas et al., 2004). While these sex differences might be indicative of differential relationships between sleep quantity and delinquency between males and females, this area of inquiry has not been appropriately assessed by previous research. In one study, Sosnowski et al. (2016) found that sleep problems mediated the effect of victimization on nonviolent delinquency in a sample of seventh graders, but that the strength of this relationship did not differ by sex. Though no sex differences were found, the main independent variable in this study was victimization, not sleep, and the sample was not nationally representative, potentially indicating community-specific patterns. In another study, Mears et al. (2022) found that the association between sleep and delinquency is similar for males and females. However, violent and nonviolent delinquent behaviors were pooled together to create the measure of delinquency and the sample consisted of both middle school and high school students together (aged 11–21 years) which may have confounded sex differences.

The Current Study

The current study seeks to begin to address the above-mentioned gaps in the existing body of literature on sleep and delinquency by examining the relations between the frequency of adequate amounts of sleep, violent, and nonviolent delinquent behavior in nationally representative samples of 8th, 10th, and 12th graders from the United States. Specifically, the current study aims to: (1) assess whether and to what extent the frequency of adequate amounts of sleep is differentially or commonly associated with violent and nonviolent delinquency; (2) examine if associations between the frequency of adequate amounts of sleep and different forms of delinquency vary across grades; and (3) evaluate whether observed associations vary by sex. Based on prior research showing that fewer adolescents engage in violent delinquent behavior relative to nonviolent behavior, we hypothesize that smaller frequencies of adequate amounts of sleep will be more strongly associated with violent delinquent behavior, compared to nonviolent delinquent behavior. We also hypothesize that the relationship between smaller frequencies of adequate amounts of sleep and both forms of delinquent offending will become stronger as grade level increases. Lastly, we hypothesize that the relations between frequency of adequate amounts of sleep and both forms of delinquency will be stronger for males than females, since males are less likely to experience problems with sleep quantity as a side effect of biological development during adolescence and are more likely to engage in delinquent behaviors.

Methods

Data

The current study used data from Monitoring the Future (MTF) surveys, which assess the behaviors, opinions, and attitudes of samples of nationally representative 8th, 10th, and 12th graders in the United States. Every year, approximately 420 public and private schools across the United States are randomly chosen to participate in the survey. Within those schools, about 18,000 8th grade students, 17,000 10th grade students, and 16,000 12th grade students are randomly selected to participate in the survey. Every student at the selected school is given a survey and time during class to complete it. To reduce the length of the surveys, MTF has four possible questionnaires for 8th and 10th grade students and 6th for 12th grade students. These questionnaires each have the same demographic questions, but differ in what set of questions they ask (e.g., participation in extracurricular activities, mental health questions). Within each school, students are randomly assigned one of the four or six possible questionnaires.

The MTF surveys from 2010 to 2019 for all three grade levels were used. These years of data were selected based on the premise that the responses would reflect views and behaviors of contemporary adolescents who are growing up in environments characterized by widespread technology and social media usage. We did not include responses from 2020 and 2021 survey respondents because the COVID-19 pandemic could have profoundly impacted sleep patterns and involvement in delinquent behaviors. As this study examined the relations between sleep quantity and perpetration of delinquent behaviors, we only included participants who filled out questionnaire form number two, which asks about both topics. We combined responses from each of our 10 waves of data for each grade level. In total, over the study period of 2010 to 2019, we have data on 51,895 8th graders, 49,378 10th graders, and 23,106 12th graders.

Participants had the option of answering all or some of the questions. The analytic sample consisted of only participants with valid data on all variables. Missing data patterns were examined with preliminary analyses. Results from chi-square tests showed that missing data on sleep quantity did not vary based on delinquency (ps > .05 for all types and all grades), but did vary based on sex (ps < .001 for all grades), race (ps < .001 for all grades), mother education level (ps < .001 for all grades), and father education level (ps < .001 for all grades). Missing sleep data was more prevalent among males, participants who identified as White (compared to participants who identified as Black), and participants with parents who had lower education levels. Since data was not consistently categorized as missing completely at random, we controlled for sex, race, and parental education in all analyses. The final sample consisted of 24,624 8th graders, 29,355 10th graders, and 13,746 12th graders.

Measures

Frequency of Adequate Amounts of Sleep

This variable was measured by responses to the following question: “How often do you get at least 7 hours of sleep?” Respondents were asked to answer the question with one of six possible choices: 1 = never, 2 = seldom, 3 = sometimes, 4 = most days, 5 = nearly every day, or 6 = every day.

Violent Delinquency

Violent delinquency was measured by responses to three questions. The questions asked participants “During the last 12 months, how often have you: (1) taken part in a fight where a group of your friends were against another group, (2) gotten into a serious fight in school or at work, and (3) hurt someone badly enough to need bandages or a doctor.” Participants had five possible answer choices: 1 = not at all, 2 = once, 3 = twice, 4 = 3 or 4 times, and 5 = 5 or more times. Reliability coefficients for violent delinquent behavior were α = .76 for 8th graders, α = .77 for 10th graders, and α = .71 for 12th graders. All three scores were summed together to create a measure of violent delinquency with scores ranging from 3 to 15.

Nonviolent Delinquency

Nonviolent delinquency was measured by responses to four questions. The questions asked participants “During the last 12 months, how often have you: (1) taken something not belonging to you worth under $50, (2) taken something not belonging to you worth over $50, (3) gone into some house or building when you weren’t supposed to be there, and (4) damaged school property on purpose”. Participants had five possible answer choices: 1 = not at all, 2 = once, 3 = twice, 4 = 3 or 4 times, and 5 = 5 or more times. All four scores were summed together to create a measure of nonviolent delinquency with scores ranging from 4 to 20. Reliability coefficients for nonviolent delinquent behavior were α = .74 for 8th graders, α = .72 for 10th graders, and α = .67 for 12th graders.

Depression

Research finds that depression is related to both sleep (Gregory & O’Connor, 2002; Roberts et al., 2002) and perpetration of violence (Fazel et al., 2015). For 8th and 10th graders, the questionnaire form that asked about sleep quantity and perpetration of violence also asked about depressive symptoms, which we include as a control. In line with past studies (Coley et al., 2019; Keyes et al., 2019), we included four measures of depressive symptoms that are asked in MTF surveys which were: (1) life often seems meaningless, (2) the future often seems hopeless, (3) I enjoy life as much as anyone, and (4) It feels good to be alive. Respondents were asked how much they agree or disagree with each of the above statements by choosing one of the following answer choices from a Likert-type scale: 1 = disagree, 2 = mostly disagree, 3 = neither, 4 = mostly agree, or 5 = agree. The last two statements were reverse coded and then all four scores were summed for each participant to create an index for depressive symptoms. Higher scores correspond to greater depression and scores ranged from 4 to 20. Reliability coefficients for depression were α = .81 for 8th graders and α = .82 for 10th graders. Although MTF does ask about depressive symptoms for 12th graders, it only does so in the questionnaire which does not ask about sleep or delinquency. As such, we did not include a measure of depression as a control variable for 12th graders.

Risk-Taking

Risk-taking is another variable that is highly correlated with delinquent behavior (Wood et al., 1993). Therefore, we included two measures of risk-taking that were asked in MTF surveys: (1) I like to test myself every now and then by doing something a little risky and (2) I get a real kick out of doing things that are a little dangerous. Respondents were asked how much they agree or disagree with each of the above statements by choosing one of the following answer choices: 1 = disagree, 2 = mostly disagree, 3 = neither, 4 = mostly agree, or 5 = agree. Both scores were summed for each participant to create an index of risk-taking. Scores ranged from 2 to 10. Higher scores corresponded to greater risk-taking tendencies. Reliability coefficients were α = .77 for 8th and 10th graders and α = .83 for 12th graders.

Peer Substance Use

As peer substance use is associated with delinquency (D’Amico et al., 2008), we identified four questions that ask about peer substance use and created an index score that we use as a control. The four measures of peer substance we used asked participants “How many of your friends would you estimate. . .” (1) Drink alcoholic beverages (liquor, beer, and wine), (2) Get drunk at least once a week, (3) Smoke marijuana (pot and weed) or hashish, and (4) Smoke cigarettes. The answer choices to choose from were 1 = none, 2 = a few, 3 = some, 4 = most, and 5 = all. All four scores were summed for each participant to create an index of peer substance use with scores ranging from 4 to 20.

Demographics

To control for differences among participants, we included respondent sex, race, school region, mother education level, father education level, academic performance, and year of survey completion as controls. Respondents were asked to designate their sex as either male or female. Sex was measured as a dichotomous variable where female = 0 and male = 1. The categories of race available in the public version of the datasets used in this study were non-Hispanic Black (= 1), Hispanic (= 2), and non-Hispanic White (= 3). The school region options were Midwest (= 1), Northeast (= 2), South (= 3), and West (= 4). Respondents were asked to give the highest level of schooling their mother and father completed. The options for both questions were: completed grade school or less (= 1), some high school (= 2), completed high school (= 3), some college (= 4), completed college (= 5), and graduate or professional school after college (= 6). Respondents were asked to provide their average grade in school so far. The categories available to choose from were D (= 1), C− (=2), C (= 3), C+ (= 4), B− (= 5), B (= 6), B+ (= 7), A− (= 8), and A (= 9). Year of survey completion was a continuous variable indicating when each respondent answered the MTF survey.

Plan of Analysis

The analysis was carried out in a series of interconnected steps. First, descriptive statistics were calculated for each sample. Second, Pearson correlation coefficients were calculated to examine the bivariate linear relationship between the frequency of adequate amounts of sleep and measures of delinquency across grades. Third, a series of ordinary least square (OLS) regression models were calculated to assess if inadequate sleep quantity is independently associated with levels of violent and nonviolent offending, while controlling for sex, race, school region, mother and father education, depression (8th and 10th graders only), risk-taking, peer substance use, and academic performance across 8th, 10th, and 12th grades. A second series of OLS regression models were calculated for each grade and included an interaction term between the frequency of adequate amounts of sleep and sex to assess if the association between inadequate sleep and violent and nonviolent delinquency was moderated by sex. Lastly, a series of equality of coefficients tests, specifically, Wald tests, were calculated to assess if the degree of association between adequate amounts of sleep and violent delinquent behavior differed from the association between adequate amounts of sleep and nonviolent delinquent behavior at each grade level. All analyses were conducted using Stata 16.1 (StataCorp, 2019) with robust standard errors to partially correct for clustering.

Results

Table 1 presents the descriptive statistics for each sample. As shown in Table 1, there were a larger number of 8th grade students in the every day sleep category than in any other sleep category; for both 10th and 12th graders the most common sleep category was sometimes, indicating that younger students reported getting more sleep, on average, than older students which is consistent with previous literature (Meldrum et al., 2018). For violent delinquency, the average response across all grade levels ranged from 3.48 to 3.77, and scores decreased slightly as grade level increased, meaning that younger adolescents engaged in more violent delinquent behavior than older adolescents. For nonviolent delinquency, average responses for each nonviolent measure across all grade levels ranged from 4.97 to 5.07. Scores increased slightly as grade level increased meaning that older adolescents engaged in more nonviolent delinquent behaviors than younger adolescents. Average depression scores were 7.67 for 8th graders and 7.94 for 10th graders. Average risk-taking scores were 5.99 for 8th graders, 5.80 for 10th graders, and 5.62 for 12th graders. Average peer substance use scores were 6.73 for 8th graders, 9.34 for 10th graders, and 10.04 for 12th graders.

Table 1.

Descriptive Statistics.

Variable	8th Grade (n = 23,823)	10th Grade (n = 28,712)	12th Grade (n = 13,098)
Variable	N (%) or M (SD)	N (%) or M (SD)	N (%) or M (SD)
7 hours of sleep
Never	905 (3.80)	1,332 (4.64)	859 (6.56)
Seldom	1,997 (8.38)	3,855 (13.43)	2,232 (17.04)
Sometimes	3,590 (15.07)	6,079 (21.17)	3,324 (25.38)
Most days	4,286 (17.99)	6,051 (21.08)	2,905 (22.18)
Nearly every day	5,434 (22.81)	5,777 (20.12)	2,090 (15.96)
Every day	7,611 (31.95)	5,618 (19.57)	1,688 (12.89)
Violent delinquency	3.767 (1.76)	3.548 (1.51)	3.483 (1.37)
Nonviolent Delinquency	4.969 (2.16)	5.043 (2.23)	5.070 (2.14)
Depression Index	7.674 (3.94)	7.937 (3.89)	NA
Risk-taking	5.999 (2.63)	5.795 (2.47)	5.616 (2.27)
Peer substance use	6.728 (3.22)	9.338 (3.72)	10.039 (3.78)
Academic performance
D	395 (1.66)	619 (2.16)	100 (0.76)
C−	607 (2.55)	1,019 (3.55)	182 (1.39)
C	913 (3.83)	1,585 (5.52)	469 (3.58)
C+	1,650 (6.93)	2,446 (8.52)	844 (6.44)
B−	1,996 (8.38)	2,790 (9.72)	1,288 (9.83)
B	3,278 (13.76)	4,629 (16.12)	2,180 (16.64)
B+	4,060 (17.04)	4,855 (16.91)	2,524 (19.27)
A−	4,714 (19.79)	5,390 (18.77)	2,708 (20.68)
A	6,210 (26.07)	5,379 (18.73)	2,803 (21.40)
Mother education
Grade school or less	840 (3.53)	911 (3.17)	554 (4.23)
Some high school	1,899 (7.97)	2,090 (7.28)	949 (7.25)
High school	4,869 (20.44)	5,553 (19.34)	2,871 (21.92)
Some college	3,637 (15.27)	4,912 (17.11)	2,644 (20.19)
College	7,880 (33.08)	9,701 (33.79)	4,133 (31.55)
Graduate school	4,698 (19.72)	5,545 (19.31)	1,947 (14.87)
Father education
Grade school or less	1,079 (4.53)	1,053 (3.67)	716 (5.47)
Some high school	2,524 (10.60)	2,992 (10.42)	1,419 (10.83)
High school	6,499 (27.28)	7,351 (25.60)	3,651 (27.87)
Some college	3,047 (12.79)	4,358 (15.18)	2,199 (16.79)
College	6,747 (28.32)	7,883 (27.46)	3,206 (24.48)
Graduate school	3,927 (16.48)	5,075 (17.68)	1,907 (14.56)
School region
Midwest	5,963 (25.03)	7,629 (26.57)	3,440 (26.26)
Northeast	4,360 (18.30)	5,844 (20.35)	2,439 (18.62)
South	9,078 (38.11)	9,548 (33.25)	4,624 (35.30)
West	4,422 (18.56)	5,691 (19.82)	2,595 (19.81)
Sex
Female	12,378 (51.96)	14,701 (51.20)	6,859 (52.38)
Male	11,445 (48.04)	14,011 (48.80)	6,239 (47.63)
Race
Black	2,768 (11.62)	3,083 (10.74)	1,475 (11.26)
Hispanic	4,736 (19.88)	4,816 (16.77)	2,246 (17.15)
White	16,319 (68.50)	20,813 (72.49)	9,377 (71.59)
Year
2010	2,690 (11.29)	3,344 (11.65)	1,449 (11.06)
2011	2,826 (11.86)	3,165 (11.02)	1,482 (11.31)
2012	2,542 (10.67)	3,142 (10.94)	1,442 (11.01)
2013	2,372 (9.96)	2, 649 (9.23)	1,349 (10.30)
2014	2, 276 (9.55)	2,652 (9.24)	1,243 (9.49)
2015	2,349 (9.86)	3,138 (10.93)	1,305 (9.96)
2016	2,657 (11.15)	2,834 (9.87)	1,133 (8.65)
2017	2,300 (9.65)	2,595 (9.04)	1,230 (9.39)
2018	1,842 (7.73)	2,622 (9.13)	1,209 (9.23)
2019	1,969 (8.27)	2,571 (8.95)	1,256 (9.59)

Pearson correlation coefficients revealed that the frequency of adequate amounts of sleep was negatively correlated with violent and nonviolent behavior across all grades as the correlations ranged from −.16 to −.6 (ps < .001). This means that students who frequently reported getting at least 7 hours of sleep reported engaging in fewer acts of delinquency compared to students who reported getting 7 hours of sleep less frequently. The correlations between frequency of adequate amounts of sleep and delinquency varied across grade-level where 8th graders demonstrated the strongest association and the 12th graders demonstrated the weakest, suggesting the possibility of an age-graded effect.

After establishing that frequency of adequate amounts of sleep was correlated with violent and nonviolent delinquency, the next step in the analysis focused on examining the independent association between key variables across grade levels. Table 2 presents the regression results examining frequency of adequate amounts of sleep and delinquency among 8th graders. Model 1 presents the results from the multivariate model predicting violent delinquency in 8th grade. The results revealed that frequency of adequate amounts of sleep was not significantly associated with violent delinquency (b = .00, p > .05). Model 2 presents a second multivariate model predicting violent delinquency with an interaction term specified between frequency of adequate amounts of sleep and sex (i.e., “7 hours of Sleep × Male”). The interaction term was negative and statistically significant (b = −.06, p < .01), suggesting that males who slept less than 7 hours more regularly reported higher levels of violent delinquent behavior, compared to females who reported less than 7 hours of sleep. Model 3 presents the results from the multivariate model predicting nonviolent delinquency. The results show that frequency of adequate amounts of sleep was significantly associated was nonviolent delinquency (b = −.06, p < .001) suggesting that 8th graders who regularly sleept less than 7 hours reported higher levels of nonviolent delinquent behavior. Model 4 presents a second multivariate model predicting nonviolent delinquency with an interaction term again specified between frequency of adequate amounts of sleep and sex. The interaction term was negative and statistically significant (b = −.09, p < .001), indicating that males who slept less than 7 hours reported higher levels of nonviolent delinquent behavior, compared to females with similar habits of getting less than 7 hours of sleep.

Table 2.

OLS Regressions Predicting Violent and Nonviolent Delinquency in 8th Grade.

	Violent delinquency				Nonviolent delinquency
	Model 1		Model 2		Model 3		Model 4
	b	95% CI	b	95% CI	b	95% CI	b	95% CI
7 hours of sleep	0.00	[−0.01, 0.02]	0.03**	[0.01, 0.05]	−0.06***	[−0.09, −0.04]	−0.02	[−0.05, 0.00]
Depression	0.05***	[ 0.04, 0.06]	0.05***	[0.04, 0.06]	0.07***	[0.06, 0.08]	0.07***	[0.06, 0.08]
Father education	−0.02	[−0.04, 0.00]	−0.02	[−0.04, 0.00]	0.03**	[0.01, 0.06]	0.03**	[0.01, 0.06]
Mother education	0.01	[−0.01, 0.03]	0.01	[−0.01, 0.03]	0.02	[−0.01, 0.04]	0.02	[−0.01, 0.04]
School region
Northeast	0.01	[ −0.05, 0.06]	0.01	[−0.05, 0.06]	−0.00	[−0.07, 0.06]	−0.00	[−0.07, 0.07]
South	−0.03	[−0.09, 0.02]	−0.03	[−0.09, 0.02]	0.01	[−0.06, 0.07]	0.01	[−0.06, 0.07]
West	−0.06	[ −0.12, 0.01]	−0.06	[−0.12, 0.00]	0.05	[−0.03, 0.13]	0.05	[−0.03, 0.13]
Male	0.24***	[0.20, 0.29]	0.50***	[0.33, 0.67]	0.27***	[0.22, 0.32]	0.67***	[0.47, 0.89]
Race
Hispanic	−0.39***	[−0.49, −0.29]	−0.39***	[−0.49, −0.29]	−0.14*	[−0.25, −0.03]	−0.14*	[−0.25, −0.03]
White	−0.49***	[−0.57, −0.41]	−0.49***	[−0.57, −0.41]	−0.17***	[−0.26, −0.08]	−0.17***	[−0.26, −0.08]
Risk-taking	−0.07***	[−0.07, −0.06]	−0.07***	[−0.08, −0.06]	−0.11***	[−0.12, −0.10]	−0.12***	[−0.13, −0.10]
Peer substance use	0.16***	[0.15, 0.17]	0.16***	[0.15, 0.17]	0.21***	[0.19, 0.22]	0.21***	[0.19, 0.22]
Academic performance	−0.08***	[−0.09, −0.07]	−0.08***	[−0.09, −0.07]	−0.05***	[−0.07, −0.04]	−0.05**	[−0.07, −0.04]
Year	−0.01***	[−0.02, −0.01]	−0.01***	[−0.02, −0.01]	−0.02***	[−0.03, −0.01]	−0.02***	[−0.03, −0.01]
7 hours of sleep × male	−	−	−0.06**	[−0.09, −0.02]	−	−	−0.09***	[−0.13, −0.05]
N	23,823		23,823		23,823		23,823
R ²	.19		.19		.21		.21
Root MSE	1.58		1.58		1.92		1.92

Note. “7 Hours of Sleep” was rated on a 6-point scale, from 1 = never to 6 = every day. b = unstandardized regression coefficient; CI = confidence interval; MSE = mean square error.

p < .05. **p < .01. ***p < .001.

Table 3 presents the regression results examining frequency of adequate amounts of sleep and delinquency for 10th graders. Model 1 presents the results from the multivariate model predicting violent delinquency in 10th grade. The results revealed that frequency of adequate amounts of sleep was not significantly associated with violent delinquency (b = −.01, p > .05). Model 2 presents a second multivariate model predicting violent delinquency with an interaction term specified between frequency of adequate amounts of sleep and sex (i.e., “7 hours of Sleep × Male”). The interaction term was negative and statistically significant (b = −.06, p < .001), showing that males who reported sleeping less than 7 hours more regularly reported higher levels of violent delinquency, compared to females with similar habits of sleeping less than 7 hours. Model 3 presents the results from a multivariate model predicting nonviolent delinquency for 10th grade respondents. The results revealed that frequency of adequate amounts of sleep was significantly associated was nonviolent delinquency (b = −.07, p < .001), indicating that 10th graders who reported sleeping less than 7 hours reported higher levels of nonviolent delinquent behavior. Lastly, Model 4 presents a second multivariate model predicting nonviolent delinquency with an interaction term specified between frequency of adequate amounts of sleep and sex. The interaction term was statistically significant (b = −.08, p < .001), suggesting that males reporting less than 7 hours of sleep also reported higher levels of nonviolent delinquent behavior, compared to females with similar levels of sleeping less than 7 hours.

Table 3.

OLS Regressions Predicting Violent and Nonviolent Delinquency in 10th Grade.

	Violent delinquency				Nonviolent delinquency
	Model 1		Model 2		Model 3		Model 4
	b	95% CI	b	95% CI	b	95% CI	b	95% CI
7 hours of sleep	−0.01	[−0.02, 0.00]	0.02**	[0.01, 0.04]	−0.07***	[−0.09, −0.05]	−0.03*	[−0.05, −0.01]
Depression	0.03***	[0.02, 0.03]	0.03***	[0.02, 0.03]	0.07***	[0.06, 0.08]	0.07***	[0.06, 0.08]
Father education	−0.04***	[−0.06, −0.03]	−0.04***	[−0.06, −0.03]	0.02*	[0.00, 0.05]	0.02*	[0.00, 0.05]
Mother education	−0.00	[−0.02, 0.01]	−0.00	[−0.023, 0.01]	0.04**	[0.02, 0.06]	0.04**	[0.02, 0.06]
School region
Northeast	0.05*	[0.00, 0.09]	0.05*	[0.00, 0.09]	−0.03	[−0.09, 0.04]	−0.03	[−0.09, −0.04]
South	0.04	[−0.01, 0.08]	0.04	[−0.01, 0.08]	0.01	[−0.05, 0.07]	0.01	[−0.05, 0.07]
West	−0.05	[−0.09, 0.00]	−0.04	[−0.09, 0.01]	0.11**	[0.04, 0.19]	0.12**	[0.04, 0.19]
Male	0.27***	[0.24, 0.31]	0.53***	[0.41, 0.64]	0.38***	[0.33, 0.43]	0.72***	[0.55, 0.88]
Race
Hispanic	−0.28***	[−0.36, −0.19]	−0.28***	[−0.36, −0.19]	−0.20***	[−0.31, −0.09]	−0.20***	[−0.31, −0.09]
White	−0.41***	[−0.48, −0.33]	−0.41***	[−0.48, −0.33]	−0.28***	[−0.37, −0.19]	−0.28***	[−0.37, −0.19]
Risk-taking	−0.06***	[−0.07, −0.05]	−0.06***	[−0.07, −0.05]	−0.14***	[−0.15, −0.13]	−0.14***	[−0.15, −0.13]
Peer substance use	0.08***	[0.07, 0.08]	0.08***	[0.07, 0.08]	0.15***	[0.14, 0.16]	0.15***	[0.14, 0.16]
Academic performance	−0.07***	[−0.08, −0.06]	−0.07***	[−0.08, −0.06]	−0.08***	[−0.09, −0.06]	−0.07***	[−0.09, −0.06]
Year	−0.01***	[−0.02, −0.01]	−0.01***	[−0.02, −0.01]	−0.01*	[−0.02, −0.00]	−0.01*	[−0.02, −0.00]
7 hours of sleep × male	−	−	−0.06**	[−0.09, −0.04]	−	−	−0.08***	[−0.12, −0.05]
N	28,712		28,712		28,712		28,712
R ²	.13		.13		.18		.18
Root MSE	1.41		1.41		2.02		2.02

Note. “7 hours of sleep” was rated on a 6-point scale, from 1 = never to 6 = every day. b = unstandardized regression coefficient; CI = confidence interval; MSE = mean square error.

p < .05. **p < .01. ***p < .001.

Table 4 presents the last series of regression results examining frequency of adequate amounts of sleep and delinquency among 12th graders. Model 1 presents the results from the first multivariate model predicting violent delinquency. The results revealed that frequency of adequate amounts of sleep was significantly associated with violent delinquency (b = −.04, p < .001), suggesting that 12th graders who reported sleeping less than 7 hours more regularly reported higher levels of violent delinquent behavior. Model 2 presents a second multivariate model predicting violent delinquency with an interaction term specified between frequency of adequate amounts of sleep and sex (i.e., “7 hours of Sleep × Male”). The interaction term was not statistically significant (b = −.02, p > .05). Model 3 presents results from the third multivariate model predicting nonviolent delinquency among 12th grade respondents. The results show that frequency of adequate amounts of sleep was negatively and significantly associated was nonviolent delinquency (b = −.08, p < .001) indicating that 12th graders who reported sleeping less than 7 hours more regularly reported higher levels of nonviolent delinquent behavior. Model 4 presents a multivariate model predicting nonviolent delinquency with an interaction term between frequency of adequate amounts of sleep and sex. As can be seen, the interaction term was not statistically significant (b = −.00, p > .05), suggesting that the association between frequency of adequate amounts of sleep and nonviolent delinquency was not moderated by respondent sex.

Table 4.

OLS Regressions Predicting Violent and Nonviolent Delinquency in 12th Grade.

	Violent delinquency				Nonviolent delinquency
	Model 1		Model 2		Model 3		Model 4
	b	95% CI	b	95% CI	b	95% CI	b	95% CI
7 hours of sleep	−0.04***	[−0.06, −0.02]	−0.03**	[−0.05, −0.01]	−0.08***	[−0.12, −0.05]	−0.08***	[−0.16, −0.05]
Depression	−	−	−	−	−	−	−	−
Father education	−0.02	[−0.04, 0.00]	−0.02	[−0.04, 0.00]	0.05**	[0.02, 0.08]	0.05**	[0.02, 0.08]
Mother education	−0.03*	[−0.05, −0.01]	−0.03*	[−0.05, −0.01]	−0.01	[−0.04, 0.02]	−0.01	[−0.04, 0.02]
School region
Northeast	0.03	[−0.04, 0.10]	0.03	[−0.04, 0.10]	0.12*	[0.01, 0.21]	0.11*	[0.01, 0.21]
South	0.01	[−0.05, 0.07]	0.01	[−0.05, 0.07]	0.07	[−0.02, 0.16]	0.07	[−0.02, 0.16]
West	−0.08*	[−0.14, −0.01]	−0.08*	[−0.14, −0.01]	0.09	[−0.02, 0.19]	0.09	[−0.02, 0.19]
Male	0.21***	[0.17, 0.26]	0.27**	[0.12, 0.43]	0.39***	[0.32, 0.46]	0.40***	[0.19, 0.16]
Race
Hispanic	−0.43***	[−0.55, −0.31]	−0.43***	[−0.55, −0.31]	−0.28**	[−0.43, −0.12]	−0.28**	[−0.43, −0.12]
White	−0.46***	[−0.56, −0.36]	−0.46***	[−0.56, −0.36]	−0.34***	[−0.47, −0.21]	−0.34***	[−0.47, −0.21]
Risk-taking	−0.05***	[−0.06, −0.04]	−0.05***	[−0.06, −0.04]	−0.14***	[−0.16, −0.13]	−0.14***	[−0.16, −0.13]
Peer substance use	0.05***	[0.04, 0.05]	0.05***	[0.04, 0.05]	0.13***	[0.12, 0.14]	0.13***	[0.12, 0.14]
Academic performance	−0.06***	[−0.08, −0.04]	−0.06***	[−0.07, −0.04]	−0.07***	[−0.10, −0.05]	−0.07***	[−0.10, −0.05]
Year	−0.01**	[−0.02, −0.01]	−0.01**	[−0.02, −0.01]	−0.01*	[−0.03, −0.00]	−0.01*	[−0.03, −0.00]
7 hours of sleep × male	−	−	−0.02	[−0.06, 0.01]	−	−	−0.00	[−0.05, 0.05]
N	13,098		13,098		13,098		13,098
R ²	.07		.07		.13		.13
Root MSE	1.32		1.32		2.00		2.00

Note. “7 hours of sleep” was rated on a 6-point scale, from 1 = never to 6 = every day. b = unstandardized regression coefficient; CI = confidence interval; MSE = mean square error.

p < .05. **p < .01. ***p < .001.

The results of the Wald tests revealed that the effect of adequate amounts of sleep significantly differed between violent and nonviolent delinquency at 8th, 10th, and 12th grades (ps < .01). More specifically, the effect of adequate amounts of sleep was stronger for nonviolent delinquency than it was for violent delinquency.

Discussion

A growing body of research reports that sleep quantity is negatively associated with general levels of delinquency during adolescence (Clinkinbeard et al., 2011; Mears et al., 2022; Meldrum & Restivo, 2014). While this line of research has highlighted the important role of sleep for patterns of adolescent delinquent behavior, largely missing from this line of inquiry is an examination of whether: (1) the frequency of adequate amounts of sleep is uniquely or commonly associated with different forms of delinquent behavior, (2) if the associations between the frequency of adequate amounts of sleep and different forms of delinquent behavior varies across adolescence, and (3) if the associations are conditioned by sex. The current study aimed to begin to fill these gaps in the existing body of literature by analyzing data from three nationally samples of youth. Several findings emerged from the employed analysis that warrant further discussion.

First, the findings revealed that adolescents who reported regularly sleeping less than 7 hours a night reported higher levels of nonviolent delinquent involvement across all grade levels. Additionally, 8th and 10th grade males who reported sleeping less than 7 hours more frequently reported engaging in more nonviolent delinquent behavior, compared to females who reported similar levels of sleep. With respect to violent delinquency, frequency of adequate sleep was not associated with violent delinquency among the entire sample of 8th and 10th graders, but when sex was assessed as a moderator, results suggested that males who regularly reported sleeping fewer than 7 hours a night engaged in more violent delinquency than their female counterparts. Contrary to these results, frequency of adequate sleep was associated with violent delinquency among the entire sample of 12th graders whereby those who sustained 7 hours of sleep less frequently reported engaging in more violence, but this relationship was not moderated by sex. Overall, these results do not provide support for our first hypothesis, which was that smaller frequencies of adequate amounts of sleep will be more strongly associated with violent delinquent behavior, compared to nonviolent delinquent behavior. These results were also confirmed by the Wald tests. However, our third hypothesis suggesting that the association between higher frequency of adequate amounts of sleep and delinquent behavior would be stronger for males than females was supported.

Our second hypothesis predicting that the association between the frequency of adequate amounts of sleep and both types of delinquent behavior would increase in strength as grade level increased was supported. Our results suggested that the frequency of adequate amounts of sleep was related to nonviolent delinquent behavior across all grades, with the effect sizes being slightly larger with each subsequent grade level. Additionally, the association between higher frequency of adequate amounts of sleep and violent delinquency, among males and females combined, was only significant amongst 12th graders. One possible reason for this finding could be that 12th graders might have more opportunities to engage in violent delinquent behaviors than their younger peers. At this age, adolescents can legally drive, which allows them to spend more time with their peers and less time with their parents as they do not need to conform to their parents’ schedules to get to one place to another. Future research should begin to unpack the reasons behind changes in the frequency of adequate amounts of sleep amongst adolescents and whether these reasons are related to changes in behaviors during this period of the life span.

Our results also suggested that certain control variables were associated with violent and nonviolent delinquency. More specifically, higher depression scores, being male, and higher peer substance use scores were all consistently associated with higher increased violent and nonviolent delinquency for each grade level across all models in our main analyses. Conversely, identifying as Hispanic or White (relative to identifying as Black), higher risk-taking scores, higher academic performance, and later survey years were all associated with decreased violent and nonviolent delinquency for each grade level across all models in our main analyses. Overall, these results are consistent with results of past research on this topic (Lin & Yi, 2015; Mears et al., 2022; Vazsonyi et al., 2018).

While our results provided evidence of significant associations between frequency of adequate amounts of sleep and violent and nonviolent delinquency, it is important to note that most of the effect sizes were relatively small across all models, suggesting that sleep plays a small role in explaining what drives adolescents to engage in delinquent behaviors. As can be seen in the main result tables, sex, race, peer substance use, risk-taking, and peer substance use were generally more strongly related to both forms of delinquent behavior than was frequency of adequate sleep with respect to effect size and significance. Even considering all the control variables, the models presented in the current study only accounted for a small percentage of the variances in delinquency, ranging from about 7% to 21%. Future research should consider this limitation and include other variables that might account for a higher proportion of variance of violent and nonviolent delinquency, such as mental illness, past victimization, or neighborhood characteristics.

There are some further limitations of the current study that should be discussed to advance future research. First, this study used cross-sectional, rather than longitudinal data to test hypotheses about relations across adolescent development. As such, it is not possible to draw causal conclusions between sleep duration and delinquent offending. Subsequent studies should attempt to use longitudinal data to examine the direction of effects between changes in sleep duration and delinquency over the course of adolescence. Another limitation is that the sleep measure did not indicate how much sleep (in hours) students were averaging. Students who indicated that they never get at least 7 hours of sleep could be averaging 4 hours of sleep or 6 hours of sleep per night. Since past research has indicated that there are differences in the sleep-delinquency relationship for adolescents who report sleeping 4 and 6 hours per night (Mears et al., 2022), an hourly measure would help unpack this issue. Despite this limitation, 7 hours of sleep is less than the minimum recommend nightly number for most adolescents, and most students do get 7 hours of sleep per night (Hansen et al., 2005). A third limitation of this study is that MTF did not ask any questions about sleep quality, so we were unable to assess how sleep quality and delinquency vary by grade level. As results from previous papers report that sleep quality is also associated with delinquency (Backman et al., 2015), and sometimes even more so than sleep quantity (Connolly et al., 2021), future research should begin to examine if age affects the relationship between different measures sleep quality (e.g., degree of restless sleep and time in different sleep stages) and violent and nonviolent delinquency.

Despite these limitations, the results of this study offer important implications for school policy and activities offered for adolescents. One such implication is moving school start times to a later hour. Currently, the CDC estimates that most middle and high schools across the country start before 8:30 am (Schools and Sleep Disorders, 2020). Yet research shows that early school start times for adolescents contributes to sleep problems such as decreased sleep quantity and feeling sleepy throughout the school day (Carskadon et al., 1998; Hansen et al., 2005), which is not surprising given that people in this age group naturally wake up later compared to older and younger individuals (Crowley et al., 2007; Patel & Marianna Sockrider, 2017). Thus far, results of studies suggest that implementing later school start times improves sleep, school performance, and cognitive functioning (Dunster et al., 2018). Criminologists have recently proposed that shifting school start times to later in the morning could help reduce delinquency and substance use in adolescents (Semenza et al., 2020). In line with shifting back school start times, athletic directors and school club leaders should think about cutting back on early morning practices and organized meetings.

Another implication of this study is to implement sleep education classes in middle and high schools. One randomized controlled study of students between the ages of 10 and 19 years found that students who were assigned to attend weekly sleep education classes reported longer sleep duration throughout the week and weekend and better moods compared to the students who did not attend these classes (Chung et al., 2017). Lastly, a third implication of the current study is for adolescents to limit their use of digital media devices such as computers and smart phones that emit blue light and disrupt the circadian rhythm. Prior research has found that using these devices at high frequency and close to bedtime is associated with poor sleep and risky behaviors (Hale et al., 2019; Touitou et al., 2016). While it is not always possible for young people to eliminate screen time late at night, the use of blue light blockers and other devices and techniques could help reduce the negative effects of blue light on sleep (Van der Lely et al., 2015).

Conclusion

The results from the current study suggest that getting an adequate amount of sleep is associated with lower levels of nonviolent delinquent behavior in three nationally representative samples of 8th, 10th, and 12th grade students and lower levels of violent delinquent behavior amongst 12th grade students. While these relations vary in strength across grade level and sex, our findings are some of the first to show that students who never or rarely sleep 7 hours per night are at an increased risk of engaging in delinquent acts, compared to their same-aged peers who occasionally or regularly get seven or more hours of sleep. The results highlight the importance of adolescents achieving the recommended amount of sleep each night during adolescence. To gain a better and more detailed understanding of this relationship, future research should focus on analyzing prospective data with repeated measures to assess potential bi-directional relations between changes in the frequency of adequate amounts of sleep and violent and nonviolent delinquency.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The completion of this study was supported by the Larry Hoover Summer Research Fellowship awarded to Kristina Block in May 2021.

ORCID iD

Kristina Block

Author Biographies

Kristina Block is a PhD candidate in the Department of Criminal Justice and Criminology at Sam Houston State University. Her research interests include biopsychosocial with a focus on how health-related factors such as sleep, head injuries, and participation in athletics influence crime across the life course. Her work has appeared in journals such as Crime & Delinquency, Journal of Criminal Justice, and Punishment & Society.

Eric J. Connolly is an Associate Professor in the Department of Criminal Justice and Criminology at Sam Houston State University. His research examines biological and social factors for antisocial behavior and victimization across the life course. His work has appeared in journals such as Child Development, Child Maltreatment, Criminology, Journal of Quantitative Criminology, Journal of Adolescent Health, and Social Science & Medicine.

References

Backman

Laajasalo

Saukkonen

Salmi

Kivivuori

Aronen

E. T.

(2015). Are qualitative and quantitative sleep problems associated with delinquency when controlling for psychopathic features and parental supervision? Journal of Sleep Research, 24, 543–548.

Barnes

J. C.

Meldrum

R. C.

(2015). The impact of sleep duration on adolescent development: A genetically informed analysis of identical twin pairs. Journal of Youth and Adolescence, 44(2), 489–506.

Baum

K. T.

Desai

Field

Miller

L. E.

Rausch

Beebe

D. W.

(2014). Sleep restriction worsens mood and emotion regulation in adolescents. Journal of Child Psychology and Psychiatry, 55, 180–190.

Capaldi

D. M.

Patterson

G. R.

(1993, March 25–28). The violent adolescent male: Specialist or generalist? [Paper presentation]. Biennial meeting of the Society for Researchin Child Development, New Orleans, LA.

Carskadon

M. A.

(1990). Patterns of sleep and sleepiness in adolescents. Pediatrician, 17, 5–12.

Carskadon

M. A.

Wolfson

A. R.

Acebo

Tzischinsky

Seifer

(1998). Adolescent sleep patterns, circadian timing, and sleepiness at a transition to early school days. Sleep, 21, 871–881.

Chayer

Freedman

(2001). Frontal lobe functions. Current Neurology and Neuroscience Reports, 1(6), 547–552.

Chung

K.-F.

Chan

M.-S.

Lam

Y.-Y.

Lai

C. S.-Y.

Yeung

W.-F.

(2017). School-based sleep education programs for short sleep duration in adolescents: A systematic review and meta-analysis. Journal of School Health, 87, 401–408.

Clinkinbeard

S. S.

Simi

Evans

M. K.

Anderson

A. L.

(2011). Sleep and delinquency: Does the amount of sleep matter? Journal of Youth and Adolescence, 40, 916–930.

10.

Coley

R. L.

O’Brien

Spielvogel

(2019). Secular trends in adolescent depressive symptoms: Growing disparities between advantaged and disadvantaged schools. Journal of Youth and Adolescence, 48, 2087–2098.

11.

Connolly

E. J.

Jackson

D. B.

Semenza

D. C.

(2021). Quality over quantity? Using sibling comparisons to examine relations between sleep quality, sleep duration, and delinquency. Social Science & Medicine, 280, 114053.

12.

Connolly

E. J.

Schwartz

J. A.

Block

(2022). The role of poor sleep on the development of self-control and antisocial behavior from adolescence to adulthood. Journal of Criminal Justice, 82, 101995.

13.

Crowley

S. J.

Acebo

Carskadon

M. A.

(2007). Sleep, circadian rhythms, and delayed phase in adolescence. Sleep Medicine, 8, 602–612.

14.

D’Amico

E. J.

Edelen

M. O.

Miles

J. N.

Morral

A. R

. (2008). The longitudinal association between substance use and delinquency among high-risk youth. Drug and Alcohol Dependence, 93(1–2), 85–92.

15.

Daly

B. P.

Jameson

J. P.

Patterson

McCurdy

Kirk

Michael

K. D.

(2015). Sleep duration, mental health, and substance use among rural adolescents: Developmental correlates. Journal of Rural Mental Health, 39, 108.

16.

Dunster

G. P.

de la Iglesia

Ben-Hamo

Nave

Fleischer

J. G.

Panda

Horacio

(2018). Sleepmore in Seattle: Later school start times are associated with more sleep and better performance in high school students. Science Advances, 4, eaau6200.

17.

Farrington

D. P.

(1986). Age and crime. Crime and Justice, 7, 189–250.

18.

Fazel

Wolf

Chang

Larsson

Goodwin

G. M.

Lichtenstein

(2015). Depression and violence: A Swedish population study. The Lancet Psychiatry, 2, 224–232.

19.

Forzano

L. B.

Michels

J. L.

Carapella

R. K.

Conway

Chelonis

J. J.

(2011). Self-control and impulsivity in children: Multiple behavioral measures. The Psychological Record, 61, 425–448.

20.

Gosselin

De Koninck

Campbell

K. B

. (2005). Total sleep deprivation and novelty processing: implications for frontal lobe functioning. Clinical Neurophysiology, 116(1), 211–222.

21.

Gregory

A. M.

O’Connor

T. G.

(2002). Sleep problems in childhood: A longitudinal study of developmental change and association with behavioral problems. Journal of the American Academy of Child & Adolescent Psychiatry, 41, 964–971.

22.

Haack

Mullington

J. M.

(2005). Sustained sleep restriction reduces emotional and physical well-being. Pain, 119, 56–64.

23.

Hale

Hartstein

L. E.

LeBourgeois

M. K.

(2019). Media use and sleep in teenagers: What do we know? Current Sleep Medicine Reports, 5, 128–134.

24.

Hansen

Janssen

Schiff

Zee

P. C.

Dubocovich

M. L.

(2005). The impact of school daily schedule on adolescent sleep. Pediatrics, 115, 1555–1561.

25.

Henry

Caspi

Moffit

T. E.

Silva

P. A.

(1996) Temperamental and familial predictors of violent and nonviolent criminal convictions: Age 3 to age 18. Developmental Psychology, 32, 614–623.

26.

Hildenbrand

A. K.

Daly

B. P.

Nicholls

Brooks-Holliday

Kloss

J. D.

(2013). Increased risk for school violence-related behaviors among adolescents with insufficient sleep. Journal of School Health, 83(6), 408–414. https://doi.org/10.1111/josh.12044

27.

Jackson

D. B.

Testa

Vaughn

M. G.

Semenza

D. C.

(2020). Police stops and sleep behaviors among at-risk youth. Sleep Health, 6(4), 435–441.

28.

Junger-Tas

Ribeaud

Cruyff

M. J.

(2004). Juvenile delinquency and gender. European Journal of Criminology, 1(3), 333–375.

29.

Kamphuis

Meerlo

Koolhaas

J. M.

Lancel

(2012). Poor sleep as a potential causal factor in aggression and violence. Sleep Medicine, 13(4), 327–334. https://doi.org/10.1016/j.sleep.2011.12.006

30.

Keyes

K. M.

Gary

O’Malley

P. M.

Hamilton

Schulenberg

(2019). Recent increases in depressive symptoms among us adolescents: Trends from 1991 to 2018. Social Psychiatry and Psychiatric Epidemiology, 54, 987–996.

31.

Lai

Zeng

Chu

C. M.

(2016). Violent and nonviolent youth offenders: Preliminary evidence on group subtypes. Youth Violence and Juvenile Justice, 14, 313–329.

32.

Le Blanc

. (2020). On the future of the individual longitudinal age-crime curve. Criminal Behaviour and Mental Health, 30(4), 183–195.

33.

Lin

W. H.

C. C.

(2015). Unhealthy sleep practices, conduct problems, and daytime functioning during adolescence. Journal of Youth and Adolescence, 44(2), 431–446. https://doi.org/10.1007/s10964-014-0169-9

34.

MacDonald

J. M.

Haviland

Morral

A. R.

(2009). Assessing the relationship between violent and nonviolent criminal activity among serious adolescent offenders. Journal of Research in Crime and Delinquency, 46, 553–580.

35.

Mears

D. P.

Tomlinson

T. A.

Turanovic

J. J.

(2022). The goldilocks rule—Too little, too much, and “just right”: Curvilinear effects of sleep duration on delinquency. Justice Quarterly, 39, 267–303. https://doi.org/10.1080/07418825.2020.1729393

36.

Meijer

A. M.

Reitz

Dekovic

Van Den Wittenboer

G. L. H.

Stoel

R. D.

(2010). Longitudinal relations between sleep quality, time in bed and adolescent problem behaviour. Journal of Child Psychology and Psychiatry, 51, 1278–1286. https://doi.org/10.1111/j.1469-7610.2010.02261.x

37.

Meldrum

R. C.

Barnes

J. C.

Hay

(2015). Sleep deprivation, low self-control, and delinquency: A test of the strength model of self-control. Journal of Youth and Adolescence, 44(2), 465–477.

38.

Meldrum

R. C.

Jackson

D. B.

Archer

Ammons-Blanfort

(2018). Perceived school safety, perceived neighborhood safety, and insufficient sleep among adolescents. Sleep Health, 4(5), 429–435.

39.

Meldrum

R. C.

Restivo

(2014). The behavioral and health consequences of sleep deprivation among US high school students: Relative deprivation matters. Preventive Medicine, 63, 24–28.

40.

Mong

J. A.

Cusmano

D. M.

(2016). Sex differences in sleep: Impact of biological sex and sex steroids. Philosophical Transactions of the Royal Society B: Biological Sciences, 371, 20150110.

41.

Patel

Marianna Sockrider

(2017). Healthy sleep in teens. American Journal of Respiratory and Critical Care Medicine, 195, 1.

42.

Peach

H. D.

Gaultney

J. F.

(2013). Sleep, impulse control, and sensation-seeking predict delinquent behavior in adolescents, emerging adults, and adults. Journal of Adolescent Health, 53, 293–299.

43.

Raine

Venables

P. H.

(2017). Adolescent daytime sleepiness as a risk factor for adult crime. Journal of Child Psychology and Psychiatry, 58, 728–735.

44.

Roberts

R. E.

Roberts

C. R.

Chen

I. G.

(2002). Impact of insomnia on future functioning of adolescents. Journal of Psychosomatic Research, 53, 561–569.

45.

Rossa

K. R.

Smith

S. S.

Allan

A. C.

Sullivan

K. A.

(2014). The effects of sleep restriction on executive inhibitory control and affect in young adults. Journal of Adolescent Health, 55, 287–292.

46.

Scarpa

Raine

(2000). Violence associated with anger and impulsivity. In Borod

(Ed.), The neuropsychology of emotion (pp. 320–339). Oxford University Press.

47.

Schools and Sleep Disorders. (2020). Centers for disease control and prevention. Retrieved June 14, 2021 from https://www.cdc.gov/sleep/features/schools-start-too-early.html

48.

Semenza

D. C.

Meldrum

R. C.

Jackson

D. B.

Vaughn

M. G.

Piquero

A. R.

(2020). School start times, delinquency, and substance use: A criminological perspective. Crime & Delinquency, 66, 163–193.

49.

Sleep for Teenagers. (2022, January). Sleep Foundation. https://www.sleepfoundation.org/teens-and-sleep

50.

Sosnowski

D. W.

Kliewer

Lepore

S. J.

(2016). The role of sleep in the relationship between victimization and externalizing problems in adolescents. Journal of Youth and Adolescence, 45, 1744–1754.

51.

StataCorp. (2019). Stata Statistical Software: Release 16. StataCorp LLC.

52.

Telzer

E. H.

Goldenberg

Fuligni

A. J.

Lieberman

M. D.

Gálvan

(2015). Sleep variability in adolescence is associated with altered brain development. Developmental Cognitive Neuroscience, 14, 16–22.

53.

Touitou

Reinberg

(2016). Disruption of adolescents’ circadian clock: The vicious circle of media use, exposure to light at night, sleep loss and risk behaviors. Journal of Physiology-Paris, 110, 467–479.

54.

Van der Lely

Frey

Garbazza

Wirz-Justice

Jenni

O. G.

Steiner

Wolf

Cajochen

Bromundt

Schmidt

. (2015). Blue blocker glasses as a countermeasure for alerting effects of evening light-emitting diode screen exposure in male teenagers. Journal of Adolescent Health, 56, 113–119.

55.

Vazsonyi

A. T.

Jiskrova

G. K.

Ksinan

A. J.

(2018). Sleep, low self-control, and deviance: Direct and indirect links across immigrant groups and socioeconomic strata. Journal of Adolescence, 68, 40–49.

56.

Wood

P. B.

Pfefferbaum

Arneklev

B. J.

(1993). Risk-taking and self-control: Social psychological correlates of delinquency. Journal of Crime and Justice, 16, 111–130.