The Relationship Between Academic Achievement and Likelihood of Police Arrest Among Delinquents

Abstract

Drawing upon a recent study on the association between low self-control and differential responses from the criminal justice system, this study examined whether academic performance, a construct linked to self-control, was also associated with the probability of police arrest. The result indicated that academic performance did have a statistically significant inverse association with the likelihood of police arrest, net of low self-control and delinquency.

Keywords

academic performance intelligence executive functions differential police response

Since the fundamental tenet of the juvenile justice system is “for the best interests” of children and youth, the police can intervene for a broader range of deviant behaviors committed by juveniles than committed by adults (Sanborn & Salerno, 2005). Law enforcement agents also have considerable discretion in handling juvenile delinquency, delivering justice in an array of different forms ranging from informal warnings to official arrests.

Obviously, the most salient factors that determine whether a delinquent is arrested or not during a police encounter are legally pertinent factors, such as the type and seriousness of the offense, prior records of the delinquent, and the degree of probable cause. A body of research, however, reveals that officers’ decision is also influenced by a range of extralegal factors. For instance, existing evidence suggests that officers’ decisions are often affected by the offender’s gender, race, social class, and officers’ psychological attributes (Goldstein, 1959; Sanborn & Salerno, 2005; Terrill & Reisig, 2003; Van Maanen, 1974; Werthman & Piliavin, 1967). In addition, it appears that police organizational factors and neighborhood factors partially affect how police officers react to delinquent adolescents (Warner & Coomer, 2003; Worden, 1989).

In particular, offenders’ demeanor—such as being belligerent, disrespectful, and obdurate—has been most frequently studied as a potential extralegal predictor of punitive actions taken by the police (Black, 1976; Piliavin & Briar, 1964; Terrill & Reisig, 2003; Worden & Shepard, 1996). Although there exists a small number of scholars who doubt the effects of offender demeanors on police punitive reactions,¹ several decades of research, both quantitative and qualitative, largely endorse that the likelihood of police arrest or other punitive actions rises when offenders display a disrespectful and hostile demeanor toward the police (Black, 1976; Mastrofski, Reisig, & McCluskey, 2002; Piliavin & Briar, 1964; Sanborn & Salerno, 2005; Terrill & Reisig, 2003; Worden & Shepard, 1996). In this regard, some scholars even assert that “the most consistent predictor of a penal police response has been the demeanor of the citizen” (Mastrofski et al., 2002, p. 523). Police officers are mandated to keep calm and administer justice in an objective manner (Mastrofski et al., 2002). Being a part of the human population, however, they may not be impervious to the disrespect, threats, or aggression displayed by the juvenile offenders they daily encounter.

Recently, Beaver and colleagues presented empirical evidence using Add Health data that delinquents’ low self-control also plays a significant role in rendering police officers to take more punitive actions (Beaver, DeLisi, Mears, & Stewart, 2009). Given the behavioral and attitudinal characteristics of individuals with low self-control (Gottfredson & Hirschi, 1990), Beaver et al.’s finding indicates that low self-control is essentially what underlies the disrespectful and impulsive demeanors that eventually lead them to receive more punitive treatment from the criminal justice system. Low self-control has been a mainstay predictor of crime and delinquency in criminology. Beaver et al.’s (2009) work has contributed importantly to the field by showing that low self-control not only predicts crime and delinquency but also predicts differential responses from the criminal justice system.

In the present study, we argue that another predictor of delinquency—namely, school performance—may also relate to differentiated responses from police officers or otherwise help delinquents to elude arrest. We make a case for such a relationship based on the following grounds.

First, students with higher self-control tend to exhibit better school grades than their counterparts with lower self-control (Duckworth & Seligman, 2005, 2006). As self-control is essentially the ability to keep one’s desire for immediate gratification in check in the expectation of later rewards, students with high self-control are likely to apply themselves diligently to academic tasks by overcoming distractions for immediate and often immaterial gratification. Given Beaver et al.’s study (2009) demonstrating that high self-controlled youths are less likely to be arrested by the police because they tend to present respectful demeanors while keeping their anger and frustration in check during police encounters, it is plausible to assume that delinquents with higher school grades are also more likely to display similar agreeable demeanors to police officers than delinquents with lower school grades. Agreeable and respectful demeanors of delinquents reduce the odds of arrest (Terrill & Reisig, 2003; Werthman & Piliavin, 1967).

Second, a recent study (Yun & Lee, 2012) revealed that verbal intelligence also influences the probability of whether a delinquent will be arrested. Analyzing the Add Health data, this study demonstrated that delinquents with higher verbal intelligence scores were significantly less likely to be arrested even controlling for the level of self-control. In light of the fact that verbal intelligence is a salient correlate of academic performance in certain subjects, it can be surmised that delinquent youths with higher academic grades will be less likely to be arrested.

Third, a brain-imaging study showed that intelligence and self-control are derived from the same neural mechanisms situated in the anterior prefrontal cortex (aPFC) in the brain (Shamosh et al., 2008). Furthermore, few would dispute the fact that academic performance is associated with intelligence (Noftle & Robins, 2007). Then, part of the covariance between self-control and academic performance can be explained by the same neural mechanisms in the brain, which in turn suggests that the tendency of self-control to reduce the odds of arrest is also manifested by high academic performance.

Finally, some delinquents are better at avoiding detection by the police than other delinquents. One of the reasons that this may occur is because they are better at cautiously assessing the likelihood of detection and consequently engage in more well-planned, delinquent acts that are least likely to be detected. Such a goal-directed act that involves planning, evaluating, and coordinating is a part of what is termed the executive function of the brain. Good academic performance also requires well-functioning executive functions (Barkley, 1997; Biederman et al., 2004; Pintrich, 2000; Spinella & Miley, 2003). It can then reasonably inferred that delinquents with higher grade point average (GPAs) will generally be more skilled at evading police detection than their academically poor counterparts due to their better planned and coordinated delinquent undertakings.

Based on the aforementioned research findings and theoretical inferences, we seek to extend recent work of Beaver et al. (2009) by examining the effects of delinquents’ academic performance on the likelihood of arrest. Considering conceptual linkages between academic performance and self-control, the relative predictive power of the two variables is also assessed. Doing so shall illuminate whether academic performance is merely another marker of self-control or it exerts independent effects over and beyond the effects of low self-control. Criminological literature suggests that, while individuals with low self-control may display deviant behaviors regardless of the situation, it is often the case that the effect of low self-control is more pronounced when paired with certain particular conditions (Grasmick, Tittle, Bursik, & Arneklev, 1993; Longshore, 1998). By the same token, the effect of self-control on the odds of arrest can interact with school performance and vice versa. Thus, we further examine the extent to which the effect of academic performance is conditioned by self-control.

Before proceeding, a word of caution is warranted. Although examining potential extralegal factors that influence police decision and behaviors is an important research agenda, no existing study has ever examined academic performance as such an extralegal factor. Thus, extant literature offers very little theoretical foundation upon which we can build to occasion our own theoretical framework to account for the possible association between school performance and odds of arrest. Indeed, the linkage between the two does not seem distinctly obvious at first glance. It is understandable that only after Beaver et al.’s (2009) publication on the effects of low self-control on police punitive actions did we become suspicious of the link between school performance and police arrests. For these reasons, the theoretical rationale we offer here should be regarded as tentative. The present study, thus, is exploratory in its nature investigating the extent and nature of the association, rather than the causation, between academic performance and arrests.

Differential Responses by the Police

Law enforcement agents are permitted to exercise a degree of discretion in their duties. Within legal parameters, they can choose among different courses of actions. Police discretion has greater implications as regards juvenile delinquents than adult criminals as the juvenile justice system is guided by the precept of leniency when delinquents appear amenable to rehabilitation (Leiber & Mack, 2003). Police discretion holds further implications because juveniles are subjected to police intervention for a wider range of behaviors, including noncriminal matters such as status offenses. In most jurisdictions, status offenses and other forms of minor delinquency can be subjected to police arrest. It is also noteworthy that, except for the small number of life-course-persistent offenders, minor forms of delinquent acts are age-normative (Moffitt, 1993), meaning that the majority of adolescents, often including those who are academically capable, engage in deviant behaviors.

Research on police discretion has typically focused on whether demographic characteristics (e.g., race and gender) of the offender influence police responses (Fagan & Davies, 2000; Lundman, 1998; Skogan, 1994). Underlying some of these efforts has been the assumption that police decision making might be biased due to racial and gender stereotyping. However, empirical support for this hypothesis has been inconsistent. A line of research, on the other hand, has consistently revealed that police discretion is partially a function of offenders’ demeanor. That is, officers are more likely to take punitive actions, such as arrest, against delinquents who they perceive as disrespectful, unapologetic, belligerent, or impulsive (Black, 1976; Mastrofski et al., 2002; Piliavin & Briar, 1964; Sanborn & Salerno, 2005; Terrill & Reisig, 2003). Piliavin and Briar (1964) note, in their classic study of patrol officers:

The cues used by police to assess demeanor were fairly simple. Juveniles who were contrite about their infractions, respectful to officers, and fearful of the sanction that might be employed against them tended to be viewed by patrolmen as basically law-abiding or at least “salvageable . . . ” [I]n contrast, youthful offenders who were fractious, obdurate, or who appeared nonchalant in their encounters with patrolmen were likely to be viewed as “would-be tough guys” or “punks” who fully deserved the most severe sanction: arrest. (pp. 210-211)

Police officers view themselves as community’s moral authority figures. When they perceive their moral authority is threatened by the disrespect and incivility displayed by youthful offenders, they tend to reciprocate by taking more punitive actions (Mastrofski et al., 2002). In addition, researchers have noted that certain characteristics of juveniles’ demeanor that invoke harsh reactions from the police, such as acting impulsively and being belligerent and disrespectful, are largely isomorphic with the characteristics of low self-control (Beaver et al., 2009; Mastrofski et al., 2002). According to Gottfredson and Hirschi (1990), people with low self-control are, first and foremost, impulsive and lack the ability to delay gratification. They prefer simple physical activities to complex and mental ones, are risky and self-centered, and are ill tempered. Based on the assumption that low self-control drives certain youthful offenders to behave impulsively and disrespectfully, Beaver et al. (2009) examined the extent to which low self-control predicted the probability of arrest and conviction among a nationally representative sample of youths. Their analysis showed that low self-control positively predicted the risk of being arrested and convicted. Their study is particularly meaningful in that it showed that the widely researched criminogenic construct of self-control not only predicts crime and delinquency but also predicts differential responses from the agents of the criminal justice system. In this study, we argue that delinquents’ academic performance can also be associated with differential police responses.

Intelligence and Differential Detection Hypothesis

Different scholars define intelligence in many different ways (Nisbett et al., 2012). The biased and cruel criminal justice policies of the past associated with intelligence (Dugdale, 1877; Goddard, 1914) warn present criminologists to be vigilant of misusing the concept of intelligence. Nonetheless, intelligence, as measured by common IQ tests, is a consistent correlate of crime and delinquency. The magnitude of the correlation is as strong as other correlates, such as social class and race (Hirschi & Hindelang, 1977), and the IQ differences between delinquents and nondelinquents are typically about 8 points, or a half standard deviation (Hirschi & Hindelang, 1977; Neisser et al., 1996). Scholarly debates on the precise magnitudes of the association and why such an association exists continue, however. Herrnstein and Murray (1994) went so far as to suggest in their widely debated The Bell Curve that IQ scores are the cause of crime and delinquency excluding all other salient sociological variables. Although their argument has been repeatedly disputed by future researchers (e.g., Cullen, Gendreau, Jarjoura, & Wright, 1997), criminologists at large do not discount the need to consider intelligence in the criminological research endeavor in search of the causes and correlates of crime and delinquency. A stream of research has also revealed that the causal nature of the relationship between intelligence and delinquency is not a direct one; rather, it is an indirect relationship mediated by academic performance (Hirschi & Hindelang, 1977; Lynam, Moffitt, & Stouthamer-Loeber, 1993; McGloin & Pratt, 2003), suggesting a close linkage among intelligence, school performance, and delinquency.

Despite the evidence linking intelligence and delinquency, some early scholars argued that the relationship is essentially spurious (Murchison, 1926; Stark, 1975; Sutherland, 1931). Termed the differential detection hypothesis, this argument posits that delinquents with lower IQs are more likely to be detected by the police compared with delinquents with higher IQs. In other words, there is no real difference between high and low IQ delinquents in the frequency and seriousness of their delinquent acts. Rather, lower IQ juvenile are more likely to get caught either because they engage in delinquency in a manner that is easily detected, or because they evoke harsher responses, such as arrests, during their encounters with the police (Feldman, 1977; Hirschi & Hindelang, 1977; Moffitt & Silva, 1988).

With the advent of the self-report method of delinquency measurement (Short & Nye, 1957), it became possible to assess the veracity of the differential detection hypothesis because self-reported measures enabled researchers to gauge the level of delinquency not tainted by police detection bias. Most of the studies that used self-report measures of delinquency showed that the IQ scores of arrested delinquents were not significantly lower than those of nonarrested delinquents, indicating that the IQ differences between delinquents and nondelinquents were not the result of differential police detection (Hirschi & Hindelang, 1977, for a review). As converging evidence rejected the hypothesis, the differential detection hypothesis appeared to have been dropped from the contemporary researchers’ research agenda.

Nonetheless, it should be pointed out that the general stance taken by early researchers who examined the differential detection hypothesis was an either-or position: That is, they viewed that the IQ differentials between delinquents and nondelinquents were generated either because low intelligent youths committed more delinquency or because low intelligent delinquents were more detected. Yet, these two seemingly mutually exclusive positions may not suitably describe what actually transpires in the real world. Quite the contrary, it is quite plausible that low IQ youths commit more delinquency and, concomitantly, are more likely to get caught.

If this is the case, then how can one explain the self-report research results that showed no significant IQ differences between detected delinquents and nondetected delinquents? Note that standard self-report studies typically fail to include the most serious delinquents or high school dropouts (Thornberry & Krohn, 2000). Thus, it is reasonable to assume that the most serious, but least intelligent, delinquents have been inadequately represented in the self-report studies, leading to the null findings of the self-report studies. Another issue that should be taken into account is that IQ scores are typically a composite measure consisting of verbal and nonverbal intelligence. In comparing the IQ scores of arrested and nonarrested delinquents, the full IQ scores have been typically used (Hirschi & Hindelang, 1977). However, literature from neuroscience consistently points out that the aspect of intelligence that is linked to crime and delinquency is the verbal rather than the behavioral dimension of intelligence (Luria, 1961; Lynam et al., 1993; Moffitt, Caspi, Silva, & Stouthamer-Loeber, 1995). To extend these findings to the differential detection hypothesis, differentials in verbal IQ, rather than the global IQ, may be what partially distinguishes detected delinquents from nondetected delinquents. In support of such a suspicion, Yun and Lee’s (2012) recent study analyzing the Add Health data showed that verbal intelligence as measured by Peabody Picture Vocabulary Test scores significantly predicted the likelihood of police arrests among persistent delinquents even controlling for self-control and delinquency. Given these research findings and our aforementioned rationale, we posit that intelligence does have an impact on the odds of arrest. In this study, we explore whether and to what extent academic performance, whose sizable portion of variance is accounted for by intelligence, is linked to police arrests.

Low Self-Control and School Performance

Self-control can be regarded as a capacity to override one’s desires for immediate gratification and to refrain from acting on negative behavioral tendencies (Gottfredson & Hirschi, 1990; Tangney, Baumeister, & Boone, 2004). Evidence shows that high self-control produces a broad range of positive outcomes in life. Among youths, for instance, those with high self-control tend to abstain from abusing substance or alcohol and eat healthy (Cook, Young, Taylor, & Bedford, 1998), save more money (Romal & Kaplan, 1995), and engage in less delinquency than those with low self-control (Pratt & Cullen, 2000).

Pertaining to the research topic in this study, it is noteworthy that self-control also directly contributes to more positive interactions with others. High self-control allows individuals to refrain from uttering hurtful things on impulse when frustrated or otherwise irate (Tangney et al., 2004). Poor self-control, in contrast, may lead to outbursts of anger or aggressive acts in such context. A longitudinal study of Shoda, Mischel, and Peake (1990) found that the capacity to delay gratification at age 4 significantly predicted good interpersonal relationships in adulthood. High self-control of children, reported by parents and teachers, also predicted socially competent interactions with peers (Fabes et al., 1999). Children with low self-control engaged in more angry conflicts with peers and responded in more unfriendly manners to others when frustrated (Murphy & Eisenberg, 1997). Given these research findings, and consistent with Beaver et al. (2009), it is expected that delinquents with low self-control are more likely to act on their feelings of frustration or anger during their encounters with police, while their high self-control counterparts are better at keeping their frustration and anger in check for later rewards.

Academic performance is a sort of a superordinate concept affected by a whole host of various factors, including cognitive, psychosocial, and demographic factors (McKenzie & Schweitzer, 2001). Research suggests that academic performance is also linked to self-control. Self-control is associated with what is called crystallized intelligence, which is the “individual’s store of knowledge about the nature of the world and learned operations such as arithmetical ones which can be drawn on in solving problems” (Nisbett et al., 2012, p. 3). While crystallized intelligence is essential for adequate school grades, students with low self-control tend to use impulsive and poor problem-solving techniques in complex questions. When an answer is not within their immediate grasp, they tend to give the first answer that comes to their mind (Fink & McCown, 1993). From this perspective, impulsivity, a facet of low self-control, appears linked to low GPAs, a position supported by recent research (Duckworth & Seligman, 2005, 2006).

Intelligence and Low Self-Control

School grades are not equivalent to intelligence. Nonetheless, GPAs covary with intelligence, and IQ accounts for about 25% of the variance of academic performance (Neisser et al., 1996). In one study, the correlation between IQ scores and the SAT reached a surprising .82 (Frey & Detterman, 2004). Good school performance is a function of intelligence and the conscientious application of that intelligence to numerous academic tasks. Of course, it is difficult for persons lacking self-control to diligently apply themselves to academic tasks regardless of their IQs. If academic performance is related to intelligence and to self-control, it is plausible to assume that self-control is in turn linked to intelligence.

Recently, a meta-analysis was conducted on the association between intelligence and self-control by Shamosh and Gray (2008). The 24 eligible studies reviewed by these researchers revealed an average correlation of .23, indicating that more intelligent people tend to have higher levels of self-control. By extension, one might argue that intelligent individuals will be less likely to say or do things that might evoke negative repercussions during police encounters than their less intelligent counterparts.

This link between intelligence and self-control had long been suspected, but the precise mechanisms underlying the association were matters of conjecture prior to the advent of recent brain-imaging techniques. Recently, Shamosh and colleagues (2008) subjected 103 healthy adults to a series of hypothetical financial rewards tests where participants were to choose between small and immediate rewards and larger but later rewards. These tests were conducted in conjunction with an array of intelligence tests, whereas neural activity was assessed through functional magnetic resonance imaging (fMRI). The study revealed that high IQ participants exhibited the greatest activation in the left part of brain region known as aPFC during intelligence tests. Similar activation in the same brain region also appeared among those who were better capable of delaying immediate gratification in the hypothetical financial tests (Shamosh et al., 2008). This region of the brain has been known to be flexibly involved in various forms of higher cognition and to integrate multiple and concurrent tasks promoting a unified goal (Koechlin & Hyafil, 2007). Activation of the same aPFC brain region indicates that the aPFC is at least partially responsible for intelligence and self-control, thus accounting for the correlation between the two. In addition, this finding that the brain is implicated in self-control accords well with recent behavioral genetic studies (Beaver, Wright, & DeLisi, 2007; Beaver, Wright, DeLisi, & Vaughn, 2008). These studies convincingly demonstrate that, contrary to the parental socialization thesis of Gottfredson and Hirschi’s (1990) on the origin of self-control, self-control is under considerable genetic influences.

Individual differences of neural activities in the aPFC partially explain the tendency of higher IQ people to resist satisfying immediate urges. The same brain area may explain the strong covariance between intelligence and school performance. The left aPFC appears to play a role in integration during mathematical problem solving (De Pisapia, Slomski, & Braver, 2007), episodic memory (Reynolds, McDermott, & Braver, 2006), and matrix reasoning (Christoff et al., 2001). Episodic memory allows people to learn new information in an organized fashion, while matrix reasoning is associated with nonverbal problem solving and inductive reasoning. Considering these functions performed by aPFC that are crucial in learning, it is no wonder that GPA is closely connected to intelligence and self-control.

Although academic performance has been widely studied as an independent variable for crime and delinquency, virtually no study has ever considered it as a correlate of differential criminal justice system responses. In this study, we examine the effects of delinquents’ academic performance on the likelihood of arrest controlling for their participation in delinquent behaviors. Considering the theoretical and conceptual overlap between academic performance and self-control, we also assess the relative predictive power of the two variables. We further examine the extent to which the effect of academic performance is conditioned by self-control.

Method

Data Sources and Sample

Data for this study come from the National Longitudinal Study of Adolescent Health (Add Health), a longitudinal and nationally representative sample of American youth (Udry, 2003). Initial data collection began in 1994-1995, when the respondents were enrolled in 7th through 12th grades. A total of 132 schools across the nation were sampled through multistage stratified sampling techniques. In the beginning, an in-school questionnaire was administered to students attending these schools, and more than 90,000 students completed the survey instrument.

To obtain more detailed information from the respondents, a stratified subsample was selected and reinterviewed at their home. In all, 20,745 adolescents and 17,700 of their primary caregivers were interviewed in this in-home survey. Adolescents were queried on involvement in delinquent behaviors, police contacts, and a host of other items related to adolescent development in general (Harris et al., 2003). The second wave of data was collected from 14,738 of the respondents in 1996. Because the lapse of time between Wave I and Wave II was relatively short, most of the items in Wave I were also included in Wave II interviews. The third wave of data was collected in 2001-2002, when the majority of the participants reached the ages of 18 and 26. As a consequence, the items in the survey instruments were redesigned to include more age-appropriate questions, such as their lifetime contact with the criminal justice system and marital status, along with other items pertaining to young adults. Overall, 15,197 participants were interviewed successfully.

Given the purpose of this research, this study utilizes only those who have committed a delinquent act that could, if detected by the police, result in an arrest. Specifically, the study involved only the adolescents who reported having committed at least one delinquent act during the entire three waves of the Add Health data collection period. To that end, delinquency scales were computed for each wave by summing delinquency items: Wave I and II delinquency scales consisted of 14 items,² whereas delinquency III scale consisted of 11 items. Youths who endorsed at least one delinquency item on one of the three delinquency scales were included in this study. With this selection criterion in place and after removing a sizable number of cases with missing data on GPAs, in conjunction with the listwise deletion method, the statistical models were based on an analytical sample of 3,914 delinquent youths.³

Measures

Police Arrest

To measure whether the delinquents in our sample have ever been arrested by the police, a single item at Wave III—Have you ever been arrested or taken into custody by the police?—was used. Table 1 shows that, out of the 3,914 delinquents, 12% reported having been arrested by the police.

Table 1.

Descriptive Statistics (N = 3,914).

Variable	M (SD)	Minimum	Maximum
Dependent variable
Arrest	0.12 (0.32)	0	1
Independent variables
Self-control	53.47 (7.82)	20	98
Academic performance	2.84 (0.67)	1	4
Control variables
Age	15.07 (1.44)	12	20
Gender	0.50 (0.50)	0	1
Race	0.33 (0.47)	0	1
Household income	47.44 (53.24)	0	999
Wave I delinquency	3.82 (4.46)	1	42
Wave II delinquency	2.02 (3.45)	0	40
Wave III delinquency	1.04 (2.33)	0	33

Academic Performance

To measure academic performance, respondents were asked during Wave I interviews to report their most recent report card grades in English or language arts, mathematics, history, and science. Responses were coded as follows: 1 = D, 2 = C, 3 = B, and 4 = A. These four items were summed and divided by four to create the Wave I GPA measure (α = .75). The same GPA measure was also created for Wave II academic performance (α = .73). Finally, the mean value of the two GPA measures was used as a composite measure of academic performance. Higher scores represent better academic performance.

Low Self-Control

Extant literature suggests that self-control covaries with academic performance. Given that offenders’ low self-control affects police responses (Beaver et al., 2009), academic performance may also influence police responses to delinquent youths.

Gottfredson and Hirschi (1990) argued that low self-control is characterized by variations in factors such as impulsivity, a preference for simple tasks, an avoidance of mental activities, a preference for physical activities, and self-centeredness. Although the scale developed by Grasmick et al. (1993) has been most widely used as a measure of self-control, the Add Health data do not contain this measure. Yet, some studies showed that the Grasmick et al.’s (1993) scale was not particularly more valid and reliable than other measures of low self-control, and using a different measure of self-control does not bias the results (DeLisi, Hochstetler, & Murphy, 2003; Pratt & Cullen, 2000). Thus, we used a 23-item self-control scale developed by Beaver and colleagues (2009) based on the Add Health data. Specifically, adolescents and their mothers were asked 23 questions during Wave I interviews that can tap individual variation in levels of self-control. For instance, mothers were asked whether their child had a bad temper, whether they could trust their child, and whether their child got along well with other children. Adolescents were asked to indicate whether they usually went with their “gut feeling” when making a decision, whether they worked hard to get what they wanted, and whether they solved problems systematically and analytically. Responses to the 23 items were summed to create a low self-control scale, where high scores indicated lower level of self-control. The validity and reliability of the scale were supported by Beaver et al.’s (2009) psychometric analyses. The Cronbach’s alpha for the scale in this study was satisfactory (.73).⁴ A complete list of the items is presented in the appendix.

Control Variables

Previous research suggests that girls are less likely than boys to be arrested (Chesney-Lind, 1973), and minority youths are more likely than Whites to behave disrespectfully to the police (Taylor, Turner, Esbensen, & Winfree, 2001). To help ensure that any relationships revealed were not the result of spuriousness, a set of demographic control variables were used. The control variables are age (in years), sex (0 = female, 1 = male), race (0 = White, 1 = non-White), and household income. Household income was created by a parent’s report of the total household income before taxes in 1994, which ranged from US $0 to US $999,000.

When predicting the probability of arrest, the most legally relevant variable is involvement in delinquency itself. To avoid estimating a mis-specified model, thus, adolescents’ delinquent involvements were accounted for, by including delinquency scales at Waves I, II, and III as additional statistical controls.⁵ These delinquency scales were the same scales that were used to cull the analytical sample of this study. The Add Health contains a range of items measuring delinquent acts. For instance, youths were asked at Wave I to self-report how many times in the past 12 months they had hurt someone badly enough to need medical attention, deliberately damaged property, sold drugs, taken something from a store without paying, painted graffiti on someone else’s property, driven other’s car without permission, and gotten into a physical fight. Response categories were mostly 0 = never, 1 = once or twice, 2 = 3 or 4 times, and 3 = 5 or more times. Delinquency items at Wave II are quite similar because most participants were attending schools and were still in their adolescence at Wave II. At Wave III, however, most respondents had reached young adulthood. Accordingly, some of the delinquency questions were dropped and others were added to reflect age-appropriate topics. For example, respondents were asked whether they had deliberately written a bad check or used someone else’s credit card without permission instead joyriding and painting graffiti. Delinquency scales were computed for each wave by summing the delinquency items: Wave I and II delinquency scales consisted of 14 items,⁶ while delinquency III scale consisted of 11 items. All of the delinquency scales exhibited acceptable reliabilities of .80, .80, and .77, respectively.

Plan of Analysis

The analyses for this study proceeded in two related steps. First, multivariate analyses using logistic regression were conducted to estimate main effect models predicting the probability of arrest, net of control variables. In this stage of the analyses, hierarchical models were estimated, where low self-control and academic performance were, first, sequentially and, then, simultaneously entered into the model. At the second stage, interactive effects of low self-control and academic performance in predicting the odds of police arrest were examined.

To test for an interaction effect, we used a split-sample method. The multiplicative interaction term typically used in linear models can be problematic in nonlinear models because the magnitude of the interaction effect does not equal the marginal effect of the interaction term and can be of opposite signs (Ai & Norton, 2003). To use a split-sample method, we dichotomized the low self-control scale by splitting it at the mean, and then classified delinquents with a score below the mean as the high self-control group and those with a score above the mean as the low self-control group.

With split-sample models, an interaction effect exists if the logistic coefficient for one sample is substantially different from the coefficient for the other sample. The split-sample method has been used by delinquency researchers for detecting interactions among predictors in the etiology of deviant behaviors (Beaver, Wright, & DeLisi, 2008; Turner, Hartman, & Bishop, 2007).

The Add Health uses a stratified, multistage cluster design, whose clusters include region, urbanicity, and schools. Such a sampling design inevitably introduces nested observations, which violate the regression assumption of independence in observations. Resulting heteroscedastic error terms artificially deflate standard errors and tend to bias tests of statistical significance for the coefficients. Our approach to address this problem was to estimate robust standard errors by adopting Huber/White/Sandwich estimator of the variance (White, 1980). Robust standard errors typically take into account minor violations of regression assumptions, including correcting the covariance matrix of the estimates for heteroscedasticity. Compared with traditional standard error models, when robust standard errors are used, coefficient estimates typically remain the same but standard errors become larger, making it harder to reject the null hypothesis (White, 1980).

Results

The analysis began by estimating multivariate logistic main effects models, where the dichotomous arrest variable was used as the dependent variable. Table 2 shows the results. The maximum likelihood estimates are presented with robust standard errors corrected for clustering in the data. Consistent with past research (Beaver et al., 2009), Model 1 shows that low self-control predicts the odds of arrest significantly (p = .002), net of controls. As Beaver et al.’s (2009) study also used the Add Health data, their estimates pertaining to low self-control are essentially the same as ours presented in Model 1.⁷

Table 2.

Logistic Regression Equations Examining the Effects of Low Self-Control and Academic Performance on Arrest.

	Model 1			Model 2			Model 3
Variable	B	SE	Odds ratio	B	SE	Odds ratio	B	SE	Odds ratio
Low self-control	0.02*	.01	1.02		-		0.02*	.01	1.02
Academic performance		-		−0.26*	.08	0.77	−0.22*	.09	0.81
Control variables
Age	0.01	.04	1.00	0.01	.04	1.00	0.01	.04	1.00
Sex	1.35*	.12	3.87	1.28*	.12	3.58	1.32*	.12	3.74
Non-White	−0.12	.10	0.88	−0.19	.12	0.82	−0.18	.12	0.83
Household income	0.99	.00	0.99	0.00	.00	1.00	0.01	.01	1.00
Wave I delinquency	0.03*	.01	1.03	0.04*	.01	1.04	0.03*	.01	1.03
Wave II delinquency	0.07*	.01	1.07	0.07*	.01	1.07	0.07*	.01	1.07
Wave III delinquency	0.18*	.02	1.20	0.19*	.02	1.21	0.19*	.02	1.20

Note: Huber/White standard errors are presented.

p < .05, two-tailed.

Among control variables, the predictive power of sex and three delinquency scales appears statistically meaningful. Model 2 replaces low self-control with academic performance. The result of Model 2 provides support for the main hypothesis of this study that academic performance of delinquents predicts the odds of arrest (p = .002). To assess the relative predictive power of low self-control and academic performance, both predictors are entered into Model 3. Surprisingly, academic performance maintains its statistical significance in the face of the presence of low self-control, and the magnitude of its predictive power attenuates only slightly (p = .011, note the change of odds ratio (OR) from .77 in Model 2 to .81 in Model 3).

Although statistical significance and direction are established through a reading of estimates in Model 3, direct interpretation of the estimate itself from a logistic regression can be challenging. For this reason, we used the prchange algorithm developed by Long and Freese (2005) to examine the discrete change in the probability of the dependent variable concerning low self-control and academic performance. The results provided strong support for our hypothesis. When the effect of academic performance was compared with the effect of self-control, the effect of the former is quite comparable with the effect of the latter. Specifically, the change in the probability of arrest was 13.6% when low self-control scores change one standard deviation unit while holding all other variables at their mean. The same value corresponding to one standard deviation unit change in academic performance was 14.9%.

After estimating the main effects models, we examined the extent to which academic performance was conditioned by self-control in predicting the probability of arrest by using the split-sample method. The full sample was split at the mean of the low self-control scale. The first model in Table 3 estimated a logistic equation for the high self-control group, whereas the model on the right-hand side estimated for the low self-control group. In both models, control variables took almost the similar patterns as in the previous full sample models. Regarding academic performance, the key variable in this study, a different pattern, emerged, however. Academic performance did not exhibit a statistically significant association with arrest in the high self-control group (p = .196), while it did show a significant relationship in the low self-control group (p = .009). These findings indicate that academic performance interacts with self-control in predicting the likelihood of arrest. Nonetheless, there is some disagreement with using split-ample approaches to studying interactions. To corroborate, thus, we additionally tested the equality of coefficients across the two models in Table 3. A difference-in-coefficients z-test, following the equation suggested by Paternoster, Brame, Mazerolle, and Piquero (1998), showed that the effect sizes for academic performance were in fact not statistically different between the two samples (p = .86).

Table 3.

Effect of Academic Performance on Arrest by Self-Control Classification.

	High self-control group			Low self-control group
Variable	B	SE	Odds ratio	B	SE	Odds ratio
Academic performance	−0.14	.11	0.87	−0.33*	.13	0.72
Control variables
Age	−0.02	.06	0.98	0.03	.05	1.03
Sex	1.16*	.17	3.19	1.57*	.22	4.82
Non-White	−0.26	.17	0.77	−0.30	.16	0.74
Household income	0.01	.01	1.00	−0.01	.01	0.99
Wave I delinquency	0.04*	.02	1.03	0.02	.02	1.02
Wave II delinquency	0.06*	.02	1.06	0.08*	.02	1.08
Wave III delinquency	0.22*	.03	1.24	0.16*	.03	1.17

Thus far, our analyses revealed an independent effect of academic performance on the odds of arrest, net of the effects of low self-control and other controls. Given that academic performance is largely a function of intelligence, one may argue that the effects of academic performance on arrests simply reflect the effects of intelligence on arrests. To test such a possibility, we conducted an auxiliary regression analysis using a vocabulary intelligence index consisting of Wave I and Wave III Peabody Picture Vocabulary Test scores available in the Add Health data (results not presented here). When the index was added to Model 3 in Table 2 as an additional variable, the effects of academic performance and low self-control remained virtually unchanged, and the influence of the vocabulary intelligence index was negligible (OR = .96, p = .6), indicating that academic performance exerts statistically meaningful effects on arrests independent of low self-control and verbal intelligence.

Discussion

Police discretion has particular implications pertaining to juvenile delinquency. Juveniles engage in a disproportionately large number of deviant behaviors relative to their percentage of the population (Moffitt, 1993). Nonetheless, the relative proportion of delinquents who receive official sanctions from the criminal justice system is small compared with their adult counterparts in part due to police discretion.

A long line of research into extralegal factors that influence police discretion has focused on an array of factors, including offenders’ demographic factors, demeanor, and neighborhoods and police organizational factors (Black, 1976; Leiber & Mack, 2003; Sanborn & Salerno, 2005; Terrill & Reisig, 2003; Warner & Coomer, 2003; Worden & Shepard, 1996). Beaver et al.’s (2009) study recently added to this body of research by showing that low self-control, a mainstay predictor of crime and delinquency, is also a significant predictor of police discretion. Drawing on this research, we explored whether another predictor of delinquency—namely, academic performance—is also linked to police discretion. In testing the hypothesis, three major findings emerged. First, low self-control was significantly associated with police arrests, even controlling for delinquency involvement transpiring during the entire waves of data collection. Second academic performance also exerted significant and comparable effects on arrest even in the presence of low self-control and a measure of verbal intelligence, indicating that academic performance’s influences on arrests are unique and independent. Third, although the effect of academic performance appeared to be moderated by self-control in split-sample models, such an interaction effect disappeared when a difference-in-coefficients z-test was conducted. The initial interaction effect observed through the split-sample method may have been a statistical artifact of a large sample size, which reminds future researchers of exercising caution when examining interaction effects.⁸

The overall findings revealed in this study provide support for the hypothesis regarding the inverse association between academic performance and the likelihood of arrest of delinquents. Revealing such an association seems contributory to the field, given that academic performance has largely been considered only as a predictor of delinquency. The precise causal mechanism underlying the association, however, remains somewhat unclear. As noted, a recent brain-imaging study showed that intelligence and self-control are correlated functions occurring in the same region of the brain (Shamosh et al., 2008). If one considers academic performance is partially a function of intelligence (Frey & Detterman, 2004; Neisser et al., 1996) and self-control (Duckworth & Seligman, 2005, 2006), it is reasonable to assume that delinquent youths with higher GPAs are less likely to display the ill-boding characteristics symptomatic of low self-control during police encounters. In other words, self-control is a latent property of academic performance.

Yet, our findings suggest that the inverse association between academic performance and arrest cannot be explained away by the overlap of academic performance and self-control alone. The only slight reduction of logistic coefficient for academic performance accompanying the induction of self-control into the model (from −.26 to −.22) shown in Table 2 indicates that some other dimensions of academic performance are also at work. Simply put, academic performance predicts arrest far more than what is predicted by low self-control. If low self-control is linked to arrest due to the fact that it contributes to adverse self-presentation and demeanor, as Beaver et al. (2009) maintain, then the remaining predictive capacity of academic performance should pertain to something other than offenders’ self-presentation and demeanor.

One plausible explanation relates to executive functions. As uniquely human capacities, executive functions are goal-directed neurocogntive processes that allow for the coordination and control of cognition and behaviors (Damasio, 1994; Luria, 1961; Moffitt, 1990). They enable individuals to set goals, self-monitor, and engage in well-planned, future-oriented behaviors, also allowing them to refrain from engaging in impulsive or otherwise inappropriate responses (Goldberg, 2001; Ishikawa & Raine, 2003; Moffitt, 1990). Although low academic performance had been typically considered reflecting students’ lack of discipline by parents and educators alike, recent advancement in neuroscience indicates a close linkage between academic performance and executive functions. A host of traits necessary to achieve good GPAs, such as setting goals, planning, monitoring progress, prioritizing options, and regulating attention, are essentially governed by executive functions (Barkley, 1997; Biederman et al., 2004; Pintrich, 2000; Spinella & Miley, 2003). The centrality of executive functions in learning is so important that Garner (2009) stated that although not all aspects of executive functions should be correlated with learning, all aspects of learning should be correlated with executive functions.

Not only essential in succeeding in schools, executive functions also enable individuals to perform tasks of everyday life successfully. In a simplistic example, optimal executive functions allow a housewife to plan menus, shop for groceries accordingly, and prioritize other housekeeping chores (Garner, 2009). By the same token, it is reasonable to assume that executive functions are implicated in successful commission of criminal or delinquent acts. It is well known in the criminology literature that even criminals engage in balancing of costs and benefits relative to the utility of other choices available (Clarke & Cornish, 1985). An observation of shoplifters shows that they engage in a relatively careful assessment of risks, target selections, and alternative strategies before committing the illegal act (Carroll & Weaver, 1986). Such a rational decision-making process involving executive functions also occurs in residential burglaries (Hakim, Rengert, & Shachmurove, 2001), auto burglaries (Michael, Hull, & Zahm, 2001), and even expressive crimes such as violence (Matsueda, Kreager, & Huizinga, 2006).⁹

Then, it might be that delinquent youths who are academically more proficient adopt risk management techniques more carefully due to their well-functioning executive functions than their academically challenged counterparts, thereby counteracting or minimizing successfully the chances of being caught by the police. They may be better at matching a strategy to a potential problem and thus be more skillful at adapting their behavior to changing environmental contingencies including the presence of the police. By contrast, their academically poor counterparts may often fail to see the connection between their delinquent acts and the likelihood of detection, thus leading to more arrests.

When seen from this perspective, the inverse association between academic performance and arrest may reflect the fact that more intelligent delinquents are more likely to commit delinquent acts successfully without getting detected rather than reflecting their demeanor during police encounters. However, given that low self-control, and undesirable demeanors by extension, also reflects faulty executive functions (Beaver et al., 2007), the inverse association found in our study would probably reflect demeanors and successful commission of delinquency.

Another potential explanation not related to either juveniles’ demeanor or executive functions that may account for the inverse association is police officers’ perceptions of the attitudes of parents of delinquent youths. High levels of parental involvement during law enforcement proceedings are regarded as crucial when officers opt to deal with a delinquent informally (Carrington, Schulenberg, Brunelle, & Pickles, 2004; Dantzker & Mitchell, 1998). Indeed, delinquent youths are more likely to be released to their parents if the parents have the emotional and material resources to provide their children (Sanborn & Salerno, 2005). By contrast, if parents are belligerent toward the police or refuse to appear at the police station, officers are more likely to take an official action against the delinquent youth (Carrington et al., 2004). A recent study (Boutwell & Beaver, 2010) provides support for the intergenerational transmission of self-control hypothesis wherein parents of adolescents with low self-control also tend to have low self-control themselves. Insofar as academic performance is influenced by self-control, one can surmise that parents of adolescents with poor academic grades are also likely to have comparatively low self-control and are therefore less willing to be involved with or provided support to their own troubled children. By contrast, parents of academically capable youths are likely to hold higher level of self-control. Greater involvement of these parents while their children are in trouble, then, will be more likely to lead to police’s informal sanctions such as warnings or reprimands. Although this interpretation is largely speculative, it merits further examinations by future researchers.

Although this study uncovered an association that had not yet appeared in the literature, it is important to touch upon the main limitations of our study. First, our theorization regarding the nature of the inverse association between school performance and arrest is, albeit plausible, unsubstantiated. Thus, our study should be regarded as exploratory in nature than explanatory. Examining the precise causal nature of the association is the task of future research. Second, it is well known that including serious delinquents in school-based self-report studies is notoriously difficult (Hindelang, Hirschi, & Weis, 1981; Thornberry & Krohn, 2000). It is possible that our analyses were influenced by the bias resulting from the omission of most serious delinquents. Thus, the association found in this study may apply only to youths who are mildly or moderately delinquent. Nonetheless, the Add Health is one of the largest and representative surveys of American youths, which includes a sizable number of serious and persistent delinquents who have been arrested and convicted. It is likely that our study to a degree overcame the limitations of small-scale voluntary surveys that often end up measuring only minor delinquency and youthful peccadilloes. Third, the Add Health data do not contain any information on the characteristics of the police officers who actually arrested the delinquents. Some officers may have been “quick on the draw” arresting delinquents at the slight sign of disrespect, while others held “turn-the-other-cheek” values. Nevertheless, there is no reason to believe that one type of officer happened to respond more often to delinquents with low GPAs or vice versa. Relatedly, due to the lack of available data in Add Health, we were not able to distinguish between arrests attributable to adverse demeanor and arrests attributable to poor planning. Finally, our study did not incorporate other factors surrounding police-delinquent encounters, such as victim request or presence of witnesses. These factors could certainly have moderated the effect of school performance and self-control on the likelihood of arrest. Despite these shortcomings, this study attempted to contribute to the field by examining an issue that the research community has not paid attention so far, and consequently showing initial evidence that police response toward delinquents are differentially influenced by delinquents’ academic performance.

Footnotes

Appendix

Items in the Low Self-Control Scale

1. All things considered, how is your child’s life going?

2. You get along well with your child.

3. You can trust your child.

4. Does your child have a bad temper?

5. You never argue with anyone.

6. When you get what you want, it’s usually because you worked hard for it.

7. You never get sad.

8. You never criticize other people.

9. You usually go out of your way to avoid having to deal with problems in your life.

10. Difficult problems make you very upset.

11. When making decisions, you usually go with your “gut feeling” without thinking too much about the consequences of each alternative.

12. When you have a problem to solve, one of the first things you do is get as many facts about the problem as possible.

13. When attempting to find a solution to a problem, you usually try to think of as many different ways to approach the problem as possible.

14. When making decisions, you generally use a systematic method for judging and comparing alternatives.

15. After carrying out a solution to a problem, you usually try to analyze what went right and what went wrong.

16. You like yourself just the way you are.

17. You feel like you are doing everything just about right.

18. You feel socially accepted.

19. Do you have trouble getting along with your teachers?

20. Do you have trouble paying attention in school?

21. Do you have trouble keeping your mind focused?

22. Do you have trouble getting your homework done?

23. Do you have trouble getting along with other students?

Note: Items 1 through 4 were asked to the mother, whereas all other questions were asked to the respondent.

Acknowledgements

This research uses data from Add Health, a program project directed by Kathleen Mullan Harris and designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris at the University of North Carolina at Chapel Hill and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development with cooperative funding from 23 other federal agencies and foundations. Special acknowledgment is due to Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. The Add Health website () provides information on how to obtain the Add Health data files. No direct support was received from Grant P01-HD31921 for this analysis.

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: This study was supported by research fund from Chosun University, 2014.

Notes

References

Norton

(2003). Interaction terms in logit and probit models. Economics Letters, 80, 123-129.

Barkley

(1997). ADHD and the nature of self-control. New York: Guilford.

Beaver

DeLisi

Mears

Stewart

(2009). Low self-control and contact with the criminal justice system in a nationally representative sample of males. Justice Quarterly, 26, 695-715.

Beaver

Wright

DeLisi

(2007). Self-control as an executive function: Reformulating Gottfredson and Hirschi’s parental socialization thesis. Criminal Justice and Behavior, 34, 1345-1361.

Beaver

Wright

DeLisi

(2008). Delinquent peer group formation: Evidence of a gene x environment correlation. Journal of Genetic Psychology, 169, 227-244.

Beaver

Wright

DeLisi

Vaughn

(2008). Genetic influences on the stability of low self-control: Results from a longitudinal sample of twins. Journal of Criminal Justice, 36, 478-485.

Biederman

Monuteaux

Doyle

Seidman

Wilens

Ferrero

Faraone

S. V.

(2004). Impact of executive function deficits and attention-deficit/hyperactivity disorder (ADHD) on academic outcomes in children. Journal of Consulting and Clinical Psychology, 72, 757-766.

Black

(1976). The behavior of law. New York, NY: Academic Press.

Boutwell

Beaver

(2010). The intergenerational transmission of low self-control. Journal of Research in Crime & Delinquency, 47, 174-209.

10.

Carrington

Schulenberg

Brunelle

Pickles

(2004). Police discretion with young offenders. Report to the Department of Justice Canada. Retrieved from http://canada.justice.gc.ca/eng/pi/yj-jj/res-rech/discre/pdf/rep-rap.pdf

11.

Carroll

Weaver

(1986). Shoplifters’ perceptions of crime opportunities: A process-tracing study. In Cornish

Clark

(Eds.), The reasoning criminal: Rational choice perspectives on offending (pp. 19-38). New York, NY: Springer-Verlag.

12.

Chesney-Lind

(1973). Judicial enforcement of the female sex role: The family court and the female delinquent. Issues in Criminology, 82, 51-69.

13.

Christoff

Prabhakaran

Dorfman

Zhao

Kroger

Holyoak

Gabrieli

(2001). Rostrolateral prefrontal cortex involvement in relational integration during reasoning. NeuroImage, 14, 1136-1149.

14.

Clarke

Cornish

(1985). Modeling offenders’ decisions: A framework for research and policy. Crime and Justice, 6, 147-185.

15.

Cook

Young

Taylor

Bedford

(1998). Personality correlates of alcohol consumption. Personality and Individual Differences, 4, 641-647.

16.

Cullen

Gendreau

Jarjoura

Wright

(1997). Crime and the bell curve: Lessons from intelligent criminology. Crime & Delinquency, 43, 387-411.

17.

Damasio

(1994). Descartes’ error: Emotion, reason, and the human brain. New York, NY: G. P. Putnam.

18.

Dantzker

Mitchell

(1998). Understanding today’s police. Scarborough, Ontario, Canada: Prentice Hall.

19.

DeLisi

Hochstetler

Murphy

(2003). Self-control behind bars: A validation study of the Grasmick et al. scale. Justice Quarterly, 20, 241-263.

20.

De Pisapia

Slomski

Braver

. (2007). Functional specializations in lateral prefrontal cortex associated with the integration and segregation of information in working memory. Cerebral Cortex, 17, 993-1006.

21.

Duckworth

Seligman

(2005). Self-discipline outdoes IQ in predicting academic performance of adolescents. Psychological Science, 16, 939-944.

22.

Duckworth

Seligman

(2006). Self-discipline gives girls the edge: Gender in self-discipline, grades, and achievement test scores. Journal of Educational Psychology, 98, 198-208.

23.

Dugdale

(1877). “The Jukes”: A study in crime, pauperism, disease, and heredity: Also further studies of criminals. New York, NY: G.P. Putnam.

24.

Fabes

Eisenberg

Jones

Smith

Guthrie

Poulin

Friedman

(1999). Regulation, emotionality, and preschoolers’ socially competent peer interactions. Child Development, 70, 432-442.

25.

Fagan

Davies

(2000). Street stops and broken windows: Terry, race, and disorder in New York City. Fordham Urban Law Journal, 28, 457-504.

26.

Feldman

(1977). Criminal behavior: A psychological analysis. New York, NY: John Wiley.

27.

Fink

McCown

(1993). Impulsivity in children and adolescents: Measurement, causes and treatment. In McCown

Shure

Johnson

(Eds.), The impulsive client, theory, research and treatment (pp. 279-308). Washington, DC: American Psychological Association.

28.

Frey

Detterman

(2004). Scholastic assessment or g? The relationship between the Scholastic Assessment Test and general cognitive ability. Psychological Science, 15, 373-378.

29.

Garner

J. K.

(2009). Conceptualizing the relations between executive functions and self-regulated learning. Journal of Psychology, 143, 405-426.

30.

Goddard

(1914). Feeble-mindedness: Its causes and consequences. New York, NY: Macmillan.

31.

Goldberg

(2001). The executive brain: Frontal lobes and the civilized mind. New York, NY: Oxford University Press.

32.

Goldstein

(1959). Police discretion not to invoke the criminal process; low-visibility decisions in the administration of justice. Yale Law Journal, 69, 543-592.

33.

Gottfredson

Hirschi

(1990). A general theory of crime. Stanford, CA: Stanford University Press.

34.

Grasmick

Tittle

Bursik

Arneklev

(1993). Testing the core empirical implications of Gottfredson and Hirschi’s general theory of crime. Journal of Research in Crime & Delinquency, 30, 5-29.

35.

Hakim

Rengert

Shachmurove

(2001). Target search of burglars: A revised economic model. Papers in Regional Science, 80, 121-137.

36.

Harris

Florey

Tabor

Bearman

Jones

Udry

(2003). The National Longitudinal Study of Adolescent Health: Research design. Retrieved from http://www.cpc.unc.edu/projects/addhealth/design

37.

Herrnstein

Murray

(1994). The bell curve: Intelligence and class structure in American life. New York, NY: Free Press.

38.

Hindelang

Hirschi

Weis

(1981). Measuring delinquency. Beverly Hills, CA: Sage.

39.

Hirschi

Hindelang

M. J.

(1977). Intelligence and delinquency: A revisionist review. American Sociological Review, 42, 571-587.

40.

Ishikawa

Raine

(2003). Prefrontal deficits and antisocial behavior: A causal model. In Lahey

Moffitt

Caspi

(Eds.), Causes of conduct disorder and juvenile delinquency (pp. 277-304). New York, NY: Guilford.

41.

Katz

(1988). Seductions of crime: Moral and sensual attractions in doing evil. New York, NY: Basic Books.

42.

Klinger

D. A.

(1994). Demeanor or crime? Why “hostile” citizens are more likely to be arrested. Criminology, 32, 475-493.

43.

Koechlin

Hyafil

(2007). Anterior prefrontal function and the limits of human decision-making. Science, 318, 594-598.

44.

Leiber

Mack

(2003). The individual and joint effects of race, gender, and family status on juvenile justice decision-making. Journal of Research in Crime & Delinquency, 40, 34-70.

45.

Long

Freese

(2005). Regression models for categorical outcomes using Stata. College Station, TX: Stata Press.

46.

Longshore

(1998). Self-control and criminal opportunity: A prospective test of the general theory of crime. Social Problems, 45, 102-113.

47.

Lundman

(1998). City police and drunk driving: Baseline data. Justice Quarterly, 15, 527-546.

48.

Luria

(1961). The role of speech in the regulation of normal and abnormal behavior. New York, NY: Basic Books.

49.

Lynam

Moffitt

Stouthamer-Loeber

(1993). Explaining the relation between IQ and delinquency: Class, race, test motivation, school failure, or self-control? Journal of Abnormal Psychology, 102, 187-196.

50.

Mastrofski

Reisig

McCluskey

(2002). Police disrespect toward the public: An encounter-based analysis. Criminology, 40, 519-552.

51.

Matsueda

Kreager

Huizinga

(2006). Deterring delinquents: A rational choice model of theft and violence. American Sociological Review, 71, 95-122.

52.

McGloin

Pratt

(2003). Cognitive ability and delinquent behavior among inner-city youth: A life-course analysis of main, mediating, and interaction effects. International Journal of Offender Therapy and Comparative Criminology, 47, 253-271.

53.

McKenzie

Schweitzer

(2001). Who succeeds at university? Factors predicting academic performance in first year Australian university students. Higher Education Research & Development, 20, 21-33.

54.

Michael

Hull

Zahm

(2001). Environmental factors influencing auto burglary: A case study. Environment & Behavior, 33, 368-388.

55.

Moffitt

(1990). The neuropsychology of juvenile delinquency: A critical review. In Tonry

Morris

(Eds.), Crime and justice: An annual review of research (Vol. 12, pp. 99-169). Chicago, IL: The University of Chicago Press.

56.

Moffitt

(1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701.

57.

Moffitt

Caspi

Silva

Stouthamer-Loeber

(1995). Individual differences in personality and intelligence are linked to crime: Cross-context evidence from nations, neighborhoods, genders, races, and age-cohorts. In Hagan

(Ed.), Current perspectives on aging and the life cycle: Delinquency and disrepute in the life course (Vol. 4, pp. 1-34). Greenwich, CT: JAI.

58.

Moffitt

Silva

(1988). IQ and delinquency: A direct test of the differential detection hypothesis. Journal of Abnormal Psychology, 97, 330-333.

59.

Murchison

(1926). Criminal intelligence. Worcester, MA: Clark University Press.

60.

Murphy

Eisenberg

(1997). Young children’s emotionality, regulation and social functioning and their responses when they are a target of a peer’s anger. Social Development, 6, 18-36.

61.

Neisser

Boodoo

Bouchard

Boykin

Brody

Ceci

Halpern

Zeller

(1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77-101.

62.

Nisbett

Aronson

Blair

Dickens

Flynn

Halpern

Turkheimer

(2012). Intelligence: New findings and theoretical developments. American Psychologist, 67, 130-159.

63.

Noftle

Robins

(2007). Personality predictors of academic outcomes: Big Five correlates of GPA and SAT scores. Journal of Personality and Social Psychology, 93, 116-130.

64.

Paternoster

Brame

Mazerolle

Piquero

(1998). Using the correct statistical test for the equality of regression coefficients. Criminology, 36, 859-866.

65.

Piliavin

Briar

(1964). Police encounters with juveniles. American Journal of Sociology, 70, 206-214.

66.

Pintrich

(2000). The role of goal orientation in self-regulated learning. In Boekaerts

Pintrich

Zeidner

(Eds.), Handbook of self-regulation (Vol. 451, pp. 451-502). San Diego, CA: Academic Press.

67.

Pratt

Cullen

(2000). The empirical status of Gottfredson and Hirschi’s general theory of crime: A meta-analysis. Criminology, 38, 931-964.

68.

Presdee

(2005). Burning issues: Fire, carnival and crime. In Soothill

Peelo

(Eds.), Questioning crime and criminology (pp. 69-82). London, England: Willan Publishing.

69.

Reynolds

McDermott

Braver

(2006). A direct comparison of anterior prefrontal cortex involvement in episodic retrieval and integration. Cerebral Cortex, 16, 519-528.

70.

Romal

Kaplan

(1995). Difference in self-control among spenders and savers. Psychology: A Quarterly Journal of Human Behavior, 32, 8-17.

71.

Sanborn

Salerno

(2005). The juvenile justice system: Law and process. Los Angeles, CA: Roxbury.

72.

Shamosh

Deyoung

Green

Reis

Johnson

Conway

Gray

J. R.

(2008). Individual differences in delay discounting: Relation to intelligence, working memory, and anterior prefrontal cortex. Psychological Science, 19, 904-912.

73.

Shamosh

Gray

(2008). Delay discounting and intelligence: A meta-analysis. Intelligence, 36, 289-305.

74.

Shoda

Mischel

Peake

P. K.

(1990). Predicting adolescent cognitive and self-regulatory competencies from preschool delay of gratification: Identifying diagnostic conditions. Developmental Psychology, 26, 978-986.

75.

Short

Nye

(1957). Reported behavior As a criterion of deviant behavior. Social Problems, 5, 207-213.

76.

Skogan

(1994). Contacts between police and public: Findings from the 1992 British Crime Survey. London, England: Her Majesty’s Stationery Office.

77.

Spinella

Miley

(2003). Impulsivity and academic achievement in college students. College Student Journal, 37, 545-549.

78.

Stark

(1975). Social problems. New York, NY: Random House.

79.

Sutherland

(1931). Mental deficiency and crime. In Young

(Ed.), Social attitudes (pp. 357-375). New York, NY: Holt, Rinehart & Winston.

80.

Tangney

Baumeister

Boone

(2004). High self-control predicts good adjustment, less pathology, better grades, and interpersonal success. Journal of Personality, 72, 271-324.

81.

Taylor

Turner

Esbensen

Winfree

(2001). “Coppin” an attitude: Attitudinal differences among juvenile toward police. Journal of Criminal Justice, 29, 295-305.

82.

Terrill

Reisig

(2003). Neighborhood context and police use of force. Journal of Research in Crime & Delinquency, 40, 291-321.

83.

Thornberry

Krohn

(2000). The self-report method for measuring delinquency and crime. In Duffee

(Ed.), Measurement and analysis of crime and justice (pp. 33-83). Washington, DC: US Department of Justice.

84.

Turner

M. G.

Hartman

J. L.

Bishop

D. M.

(2007). The effects of prenatal problems, family functioning, and neighborhood disadvantage in predicting life-course-persistent offending. Criminal Justice and Behavior, 34, 1241-1261.

85.

Udry

J. R.

(2003). The national longitudinal study of adolescent health (add health), waves I and II, 1994–1996; wave III, 2001-2002 (Machine-readable data file and docu-mentation). Chapel Hill: Carolina Population Center, University of North Carolina.

86.

Van Maanen

. (1974). Working the street: A developmental view of police behavior. In Jacob

(Ed.), The potential for reform of criminal justice (pp. 83-130). Beverly Hills, CA: Sage.

87.

Warner

Coomer

(2003). Neighborhood drug arrest rates: Are they a meaningful indicator Of drug activity? A research note. Journal of Research in Crime & Delinquency, 40, 123-138.

88.

Werthman

Piliavin

(1967). Gang members and the police. In Bordua

(Ed.), The police: six sociological essays (pp. 56-98). New York, NY: John Wiley.

89.

White

(1980). A heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. Econometrica, 48, 817-838.

90.

Worden

(1989). Situational and attitudinal explanations of police behavior: A theoretical reappraisal and empirical assessment. Law Society Review, 23, 667-711.

91.

Worden

Shepard

(1996). Demeanor, crime, and police behavior: A reexamination of the police services study data. Criminology, 34, 83-105.

92.

Yun

Lee

(2012). IQ and delinquency: The differential detection hypothesis revisited. Youth Violence and Juvenile Justice. Advance online publication.