A Motivational Intervention for African American Boys Labeled as Aggressive

Abstract

A 12-week, 32-lesson afterschool intervention was conducted with third- to fifth-grade urban African American boys classified as aggressive. Grounded in attribution theory and organized around the construct of perceived responsibility in self and others, the intervention focused on increasing both social skills and academic motivation. Participants (N = 64) were randomly assigned to an intervention or no-treatment control group. Boys in the intervention group showed an increase in social skills and academic motivation skills and were rated by their teachers as more cooperative and academically persistent. The implications of the findings for the design of interventions in urban schools are discussed.

Keywords

Black males minority academic success motivation urban education academic achievement elementary school intervention

Social behavior problems and academic motivation problems often go hand in hand. Children who disengage from school frequently have histories of poor relationships with peers, characterized by aggression against others (see Coie & Dodge, 2006). Children who aggress against others, in turn, often have histories of low achievement motivation, characterized by failing grades, chronic truancy, and school disengagement in general (Loeber et al., 2003; Risi, Gerhardstein, & Kistner, 2003). In this article, we describe an urban school-based intervention for African American third- to fifth-grade boys labeled as aggressive that was based on motivational change—decreasing the motivation to aggress and increasing the motivation to achieve—as pathways to improving social and academic outcomes.

Why focus on African American boys in elementary school? The association between poor academic achievement and behavior problems is especially troubling when we consider that low-income African American boys living in an urban setting and attending an urban neighborhood school are more likely to be perceived as aggressive by teachers and peers and are more likely to experience poor overall school adjustment, including a disproportionate number of school suspensions and greater than average rates of school dropout compared with their peers (Howard, 2014; Noguera, 2009; Roderick, 2003; Rumberger, 2011). Moreover, adjustment problems and aggressive behavior in elementary school are more likely to predict negative life outcomes and early involvement with the juvenile justice system for African American youth compared with other racial/ethnic groups (Bradshaw, Schaeffer, Petras, & Ialongo, 2010). Beginning in elementary school, on just about every indicator of educational attainment, Black males fare more poorly than other racial/ethnic groups stratified by gender (National Center for Educational Statistics [NCES], 2013).

Theoretical Framework: An Attributional Analysis of Perceived Responsibility

The motivation intervention described in this article is grounded in attribution theory, one of the dominant theories of motivation relevant to achievement strivings (Graham & Weiner, 2011; Graham & Williams, 2009). Within attribution theory, the organizing theme for our intervention is the causal construct of perceived responsibility. We consider whether peers are perceived as responsible for negative events, which has implications for reducing the motivation to aggress against those peers; and we examine the degree to which individuals perceive themselves as responsible for their academic outcomes, which has implications for increasing their own motivation to achieve.

Attributions are answers to “why” questions such as “Why do other kids pick on me?” or “Why did I fail the exam?” (Weiner, 1995, 2005). Because specific attributions will vary greatly across domains and between individuals, attribution theorists have focused on the underlying dimensions or properties of causes in addition to specific causes per se. Three such dimensions have been empirically identified. These include locus, or whether a cause is internal or external to a person; stability, which designates a cause as constant or varying over time; and controllability, or whether a cause is subject to volitional influence. Controllability connotes responsibility and intentionality (Weiner, 1995). Outcomes attributed to controllable causes, such as lack of effort, lead to the inference that the person is responsible and the behavior was intended. In contrast, outcomes attributed to an uncontrollable cause, such as low aptitude, lead to an inference of non-responsibility and unintended behavior. Individuals make responsibility attributions about other people (e.g., “they did it on purpose”) and about the self (e.g., “I didn’t try hard enough”) and both of these causal inferences have consequences for subsequent feelings and behavior.

Perceived Responsibility in Others and Aggression

One robust finding in the peer aggression literature is that aggressive children have a tendency (bias) to overattribute hostile intent to peers, particularly in ambiguous situations (Coie & Dodge, 2006). For example, imagine a situation where two students are standing in a crowded lunch line and one pushes the other, but it is unclear why the push occurred. Aggressive youth are more likely to infer that the push was intended (“on purpose”) and to retaliate against the perceived perpetrator. Being quick to assign blame following an ambiguously caused negative outcome often is an impulsive reaction that can be modified with training in how to more accurately infer another’s intentions. In an earlier intervention guided by attribution principles, Hudley and Graham (1993) created a 12-lesson curriculum titled Brain Power that resulted in significant reductions in children’s inappropriate aggression by changing their tendency to attribute hostile intentions to peers (Hudley, 2008).

The present study elaborates and builds upon those promising findings by also considering whether aggressive children are aware of the causal inferences that peers make about them. As actors, or potential provocateurs, aggressive youth have a tendency to engage in dysfunctional behavior that can arouse anger and retaliatory hostility in others. And they may lack the social skills that allow them to manage the impressions that others have of them. A good example of impression management skills occurs in the domain of account giving. Accounts are explanations or reasons for social transgressions and they include apology (confession), excuses, justifications, and denials (Scott & Lyman, 1968). Everyday expressions such as “Excuse me, but . . . ” or “What I meant was . . . ” are examples of accounts that have the potential to remediate social predicaments. Effective account giving following a transgression protects relationships and helps individuals present themselves in a more favorable light (Young & Thompson, 2011). By shifting causal responsibility away from the self, accounts have the potential to reduce anger and hostility from others. Furthermore, even when guilt and self-blame for a transgression are evident, transgressors who acknowledge responsibility and apologize for their misdeeds are more likely to evoke forgiveness (rather than anger) on the part of the offended person than are individuals who deny or minimize wrongdoing (Davis & Gold, 2011).

There is evidence to suggest that aggressive youth have poor understanding of accounts and their impression management functions. For example, Graham, Weiner, & Benesh-Weiner (1995) reported that aggressive boys were less likely than nonaggressive boys to accept (honor) an apology from a remorseful peer or to recognize that perceivers are less angry when excuses for transgressions are uncontrollable. An important component of the intervention therefore was to teach participants to understand the characteristics of different kinds of accounts, and what they imply about personal responsibility. We wanted participants to learn the adaptiveness of accepting responsibility for their own misdeeds and of honoring the accounts of others by displaying greater forgiveness toward others who apologize for their transgressions.

Self-Responsibility for Achievement and Academic Motivation

The academic component of the intervention focuses on taking personal responsibility for academic outcomes. Our basic assumption is that within an achievement context, such as many social contexts, an individual is faced with the option (decision) to ascribe responsibility for outcomes to the self (e.g., to lack of effort following failure) or to factors for which the individual cannot be held responsible (e.g., low aptitude, poor teaching). It is further assumed that self-responsibility is the more adaptive motivational state because it is more likely to result in high expectancy, positive affect, and sustained effort (Graham and Williams, 2009).

Four sets of principles from motivation theory and research indicate that students’ self-responsibility is enhanced if they choose tasks of intermediate difficulty, set proximal (short-term) as opposed to distal (long-term) goals, are mastery rather than performance oriented, and attribute failure to controllable causes such as lack of effort (Graham & Weiner, 2011). First, the classic studies of Atkinson (1964) with individuals who differ in the achievement motive revealed that preference for intermediate difficulty is associated with feeling the most pride after success and greater persistence after failure. Intermediate risk taking fosters self-responsibility, inasmuch as failure at a very difficult task (i.e., taking too high a risk) and success at a very easy task (i.e., taking little or no risk) can be attributed to factors outside of oneself.

Related to intermediate risk taking is the motivational skill of goal setting. Self-efficacy research has documented the motivation enhancing qualities of setting proximal goals in contrast to distal goals (Bandura, 1997). Proximal goals involve mentally decomposing a task into component parts so that it becomes more manageable. As those goals are attained, individuals have a basis for judging how well they are doing and for making the necessary adjustments to their behavior to achieve desired outcomes. Distal goals, in contrast, often are too far removed in time to effectively mobilize effort.

Intermediate risk taking and proximal goal setting are more likely to take place when students are mastery oriented rather than performance oriented (e.g., Elliot, 2005). A mastery orientation is one where the student’s intent is to master the task or to acquire new skills, in contrast to a performance orientation where one’s primary goal is to demonstrate high ability relative to others or to conceal low ability. A mastery orientation promotes self-responsibility because the student determines success and failure by comparison with self-standards rather than the normative standard created by comparison with others.

Finally, students are more likely to persist in the face of academic difficulty if they attribute failure to controllable factors, such as lack of effort or poor choice of strategy, than to uncontrollable causes, such as low ability, bad luck, or task difficulty (see Weiner, 1986). That is because failure ascribed to a factor that is controllable and for which one is responsible implies that the same outcome need not occur again. It is as if the student says to himself or herself: “I failed because I did not put forth enough effort, but I can try harder the next time” as opposed to “I failed because the work is too hard and there is really nothing I can do to change that.” Consistent with this principle, a large literature on attribution retraining in achievement contexts documents that students perform better and persist longer when they are taught to attribute failure to controllable causes (Wilson, Damiani, & Shelton, 2002).

In summary, self-responsibility can affect students’ choice of academic activities, how much effort they expend, their focus on mastery, and how long they will persist in the face of challenge. The academic component of the intervention was organized around these motivation principles.

Biased Perceptions of African American Males

The target population for the intervention is African American males labeled as aggressive by teachers and peers. We acknowledge that this label is based on both reputation in the peer group and interpretations of observed behavior. For some youth, this labeling may be accurate, reflecting actual behavior intended to harm others. For other youth, the label may be biased, the residue of deep-seated stereotypes in this country associating being Black and male of any age with aggression and violence (e.g., Whitley & Kite, 2006). Whether biased or not, once put in place, these labels have a profound effect on Black males’ educational opportunities and experiences. For example, as early as preschool, teachers rate Black males’ play behavior as more aggressive and threatening (McLoyd, 1985). By elementary school Black males labeled as aggressive (troublemakers) often receive the most negative treatment by teachers, including harsher classroom disciplinary practice, more criticism, and more social isolation (Ferguson, 2003; Howard, 2014). Rather than tackling the root causes of racial bias that can result in particular labels or reputations, our starting point is African American males who have indeed been classified as aggressive. Our goal was to teach these boys social and motivation skills that may help them to be successful in school, despite (or not directly addressing) societal and educational injustice.

Although not developed specifically for African American youth, attribution theory is an appropriate framework for the intervention presented here. The theory emphasizes how individuals’ causal interpretations of perceived failure (i.e., who is responsible?) can be adaptive or maladaptive and it provides guidelines for altering maladaptive thoughts to foster a sense of agency for the future. Our application of the theory is grounded in the everyday school experiences of Black male students labeled as aggressive, acknowledging that some of these experiences involve social and academic failure. With our focus on perceived responsibility, we seek to shift causal thoughts and feelings about these experiences away from the maladaptive and toward those that may be empowering.

Method

Selection of Participants

Participants were third- to fifth-grade African American boys labeled as aggressive who were attending a K-5 urban elementary school that offered an afterschool recreation program. The ethnic composition of the student body was approximately 50% African American and 50% Latino. The school was located in an economically depressed community of metropolitan Los Angeles and qualified for Title I compensatory education funds with 93% of students eligible for the district free/reduced lunch program. Thus, by available indicators, the participants comprised a low-SES (socioeconomic status) sample.

Selection of boys labeled as aggressive involved an extensive screening procedure that used well-established methods in the peer aggression literature (Coie & Dodge, 2006). First, parent permission forms describing the study were sent home with all third-to fifth-grade children in regular education classrooms (n = 769). Only students who returned signed consent forms participated in the screening phase of the study. With a consent rate of 78%, approximately 600 third to fifth graders distributed across 25 classrooms participated in this phase. Next, children were identified as aggressive based on teacher perceptions and their reputation among their peers.

Teacher ratings

Teachers completed a nine-item questionnaire for each participating student in their homeroom. We used the eight item aggression subscale of the Teacher Checklist developed by Coie & Dodge (1988). The items described behaviors that are characteristic of physical and verbal aggression in childhood (e.g., starts fights, gets mad easily, says mean things to others). On 5-point scales, teachers rated the extent to which they perceived each behavior described on the subscale as being characteristic of the particular student (1 = not at all, 5 = very much). Scores ranged from 8 to 40, with higher scores indicating more perceived aggression (α = .92). An additional item assessed the extent to which the student was perceived as academically motivated (“tries hard in class”). That question was embedded in the aggression subscale and was rated on the same 5-point scale.

Peer nominations

Each participating student in the same 25 classrooms received a class roster that contained the names of all the students in their class, listed alphabetically and by gender. With that roster, children were asked to circle the names of three students in their class who fit each of five behavioral descriptions. Three of these descriptions portrayed physically aggressive behavior (starts fights, disrupts the group, has a short temper). Two filler items described prosocial characteristics (works well with others, is helpful to classmates). The nominations that each child received on the three aggression items were summed. These totals were then standardized within classroom and again within the total sample.

African American boys were eligible for the intervention if they satisfied three selection criteria: (a) their teacher ratings on aggression were above the class mean; (b) their teacher rating on the motivation item was below the class mean; and (c) they were above the 60th percentile on peer-nominated aggression. These selection criteria were significantly interrelated in that teachers and peers agreed about which students were aggressive (r = .62), and the most aggressive students were generally perceived by teachers as low in motivation to work hard in school (r = −.39, both ps < .001). Of 74 third-, fourth-, and fifth-grade African American males who fit the selection criteria, we recruited 66 boys whose custodial parent(s) or guardian provided written informed consent. Thirty-one boys (M age = 9.48 years) were randomly assigned to the intervention and 35 boys (M = 9.87 years) were assigned to a no-treatment control group (a few parents only agreed to allow their son to participate if he was a control subject). The treatment and control groups did not differ significantly on any of the selection criteria: teacher ratings of aggression (Ms = 25.58 vs. 29.20); teacher-rated motivation (Ms = 2.58 vs. 2.95); peer-nominated aggression (zs = 2.08 vs. 1.72). The 219 noneligible boys in the screening sample were significantly less aggressive than both groups of participants: teacher-rated M = 16.99, peer-nominated z = .09 (ps < .001).

Intervention Curriculum: Best Foot Forward

The intervention, called Best Foot Forward, consisted of 32 lessons delivered as an afterschool program over 12 consecutive weeks. Half of the lessons focused on social skills and half targeted academic motivation skills. Table 1 presents an overview of the curriculum, including how many lessons were devoted to each topic, the specific goals of each topic, and how the corresponding lessons were designed.

Table 1.

Overview of the Best Foot Forward Curriculum.

Topic	Primary goals/objectives	How achieved
Introduction (1)	Introduce program	As a group deciding on guidelines for the program
	Develop group guidelines	As a group deciding on guidelines for the program
	Introduce trainers to students	Completing student questionnaire
Account giving (7)	Recognize the use of various accounts and their emotional and interpersonal consequences	Generating accounts, analyzing their emotional responses, considering the pros and cons of various accounts, practicing using and honoring through role-play and modeling throughout program
Inferring intent (8)	Accurately infer intentionality/responsibility in others	Recognizing different intents, analyzing stories/video scenes for intent, recognizing various intent “cues,” role-playing adaptive behavioral responses to provocation
Intermediate risk (4)	Encourage intermediate risk taking	Identifying own intermediate level and practicing monitoring own progress
Goal setting (4)	Learn to set realistic proximal goals	Defining and breaking down own goals, look at the pros and cons of decisions that affect your goals, learn how to gather more information before making a decision
Mastery focus (3)	Focus on improvement rather than performance	Rewarding effort on achievement tasks, role-playing good and bad help-seeking, practicing help-seeking with challenging tasks
Mastery focus (3)	Be effective help-seekers (demonstrating effort)
Attribution retraining (5)	Recognize control in effecting change and the role of effort in achieving change	Determining areas requiring effort to achieve goals, identify reasons for failure and the adaptiveness of focusing on effort, practicing persistence
Attribution retraining (5)	Attribute outcomes to effort

Social skills training

The social skills component was grouped around two themes. The first theme, account giving, was presented in seven lessons. These lessons were sequenced to teach students a variety of accounts (i.e., apology, excuse, denial) that are typically given to explain personal transgression. Each of the seven lessons incrementally developed students’ understanding of the connection between the use of particular types of accounts and managing the impressions of others. For example, students learned that apology following a social misdeed was more effective than excuses or denial because it elicits forgiveness and willingness to repair the relationship (Weiner, 1995). Lessons typically presented vignettes of social predicaments that might occur in real life (e.g., not returning a borrowed item, skipping a planned get-together), and students discussed appropriate responses if confronted by an offended peer.

The second theme in our social skills component, attributional bias, was adapted from the Brain Power program (Hudley, 2008). The eight lessons in this section taught students to more accurately infer the intention of peers in ambiguous negative situations. Lessons were sequenced to develop students’ ability to (a) interpret available social cues in interactions with peers to distinguish between intended and unintended outcomes, (b) attribute ambiguous negative situations to accidental or uncontrollable causes given uncertainty about causation, and (c) generate behaviors appropriate to these more adaptive attributions. For our African American participants living in high-risk neighborhoods, we recognize that in situations of clear threat, being quick to assign blame and preemptive responding can be adaptive survival strategies. It is ambiguous situations—wherein the intentions of others are unclear—that inferring hostile intent was judged to be maladaptive.

Academic motivation training

The academic motivation component of the curriculum was divided into four sections that focused on risk taking, proximal goals, mastery orientation, and attribution retraining. The section on intermediate risk taking taught students how to determine what makes a problem easy, medium, or hard and to recognize the benefits of working on tasks of intermediate difficulty. For example, boys competed in a weekly spelling game where they could learn new words that were easy, medium, or hard (cf. DeCharms, 1972). Although more points exchangeable for prizes could be earned by spelling harder words, intervention boys learned that the best strategy over the long term was to concentrate on medium difficulty words in which the probability of success was greater. Similarly, while playing a ring-toss game, boys were taught to adjust their level of aspiration upward after success and downward after failure (Atkinson, 1964). This helped them to recognize the motivational advantages of continuously revising their level of aspiration in the direction of intermediate difficulty.

Using concrete everyday examples where goal setting is likely to be instrumental to success, the section on goal setting taught participants about the importance of setting proximal or short-term goals rather than (and in addition to) distal or long-term goals. Over the course of the intervention, students kept a weekly log where they entered daily, weekly, and monthly goals, both academic and nonacademic. They were also given strategies for monitoring their behavior directed toward achieving those goals and for revising their goals in response to success or failure.

Several activities were designed to promote a mastery orientation to achievement as opposed to a performance orientation. Participants worked on a number of achievement tasks where success or failure could be manipulated (e.g., digit symbol substitution, anagrams). Here they were taught to focus on improvement rather than absolute performance, to reward themselves accordingly, and to avoid comparing themselves with others.

Finally, the attribution retraining section was designed to promote adaptive explanations for achievement failure. Participants read hypothetical failure scenarios and generated possible causes for the outcome. That was used as a context for discussing the characteristics of different causes and why some explanations might be more adaptive than others. Participants then worked on several achievement tasks that required persistence (e.g., origami puzzles). Here they were taught the utility of identifying factors within their control, such as lack of effort, and to avoid the endorsement of factors outside of their control, such as low ability and bad luck when faced with academic difficulty.

Dependent Variables

We included a variety of measures, both attitudinal and behavioral, that came from multiple informants (self-report, behavioral observations, teacher report, and searches of school records). Some measures were based on existing instruments, whereas others were created to test specific skills targeted in this intervention. Except where indicated, all data were gathered at pretest (1-2 weeks before the intervention) and at posttest (1-2 weeks after the intervention) at a single testing session.

Attitudes about social behavior and academic motivation

We adapted well-established self-report instruments to measure changes in participants’ general attitudes about aggression and their academic motivation. Fourteen items from the 18-item Aggressive Beliefs Questionnaire (Slaby & Guerra, 1988) assessed students’ attitudes about the legitimacy of aggression (e.g., “It’s okay to hit someone if you don’t like them”). Respondents rated each item on a 6-point scale using an agree–disagree format (1 = definitely disagree and 6 = definitely agree). Item scores were summed and averaged for each participant (α = .75).

To measure beliefs relevant to academic motivation, participants completed the 60-item School Attitude Measure (SAM; American College Testing [ACT], 1993). The SAM includes five 12-item subscales that assess motivation for schooling, academic self-concept, sense of control, mastery, and perceived teacher evaluation. Items were rated on 6-point scales (1 = definitely disagree and 6 = definitely agree). A 6-point score was calculated by summing and averaging subscales (α = .80).

Using and honoring accounts

To measure the use of accounts, participants read a hypothetical story where they imagined that they transgressed against a peer (e.g., bumped into them, did not return a borrowed possession on time). Four possible responses were provided (“things you could say”), these included apology (“I’m sorry”), legitimate excuse (“excuse me, but . . . ”), denial (“I didn’t do it”), and indifference (“so what”). Boys were told to choose the account “that they would say first” and to rank the remaining accounts of preference.

To assess the honoring of others’ accounts, participants read hypothetical vignettes describing a peer who committed a transgression against them and who offered one of three account types (apology, excuse, denial). Following each account type, boys rated on 6-point scales the hypothetical transgressor’s intent (1 = definitely did not mean to do it and 6 = definitely did mean to do it), how sorry they believed the peer to be for the transgression (1 = not sorry at all; 6 = very sorry), and whether or not they would forgive him or her (1 = definitely would not forgive; 6 = definitely would forgive).

Attributions for failure

To measure attributional change, participants were asked to recall the last time they did poorly on a test and to rate four attributions as possible explanations for their failure. These attributions were low ability (“you are not smart enough”), lack of effort (“you did not try hard enough”), a bad teacher (“you don’t have a good teacher”), and bad luck (“you were unlucky; the teacher asked you things you hadn’t studied”). Each attribution was rated on a 7-point scale (1 = definitely not a reason and 7 = definitely a reason).

Teacher ratings of behavior

Students’ social behavior and school motivation were rated by teachers on three subscales from the Social Skills Rating System for Teachers (Gresham & Elliot, 1990). The Self-Control subscale contained nine items that measured students’ ability to regulate their behavior (e.g., controls temper with peers, compromises during conflict, takes criticism well). The Externalizing subscale contained six items that measured negative student behaviors (e.g., gets angry easily, argues a lot, talks back to adults when corrected). Because these two scales represented opposite ends of a social behavior continuum (r = −.83), we reversed scored the items on the Self-Control subscale and combined them with the Externalizing subscale to create one negative social behavior scale (α = .92).

The Cooperation/Motivation subscale (α = .88) was comprised of six items that assessed students’ general motivation toward schoolwork (e.g., produces correct schoolwork, ignores peer distractions when doing work). Each item was rated on a 5-point scale (1 = never and 5 = almost always). We also created five new items that tapped other specific motivation skills included in the intervention (e.g., gives up easily; prefers tasks that are too hard or too easy). We considered these five items to be a measure of teacher-rated persistence (α = .74).

When completing their ratings, teachers remained blind to the treatment group of their students. That was possible because the school held a number of afterschool recreation activities in which many students participated (i.e., it was common for students to stay after school). Thus, teachers were unaware of whether the boys in this study who stayed after school were in the intervention, the control group, or another afterschool activity.

Laboratory maze task

In addition to student self-reports, we wanted one type of dependent measure that could capture social skills and academic motivation training in the context of actual behavior. Adapted from Hudley and Graham (1993), about 1 month after the intervention ended, all boys participated in a problem-solving task that supposedly was unrelated to the intervention. The task required each boy to communicate with an unseen peer who was seated on the other side of a barrier. Using simple grid maps, the unseen peer was to give directions to the participant so that he could complete a maze, with the goal of winning a prize. However, the task was designed to actually block goal attainment. Unbeknownst to either child, the peer’s map was different from the participant’s. Thus, incorrect solutions were necessarily given, the maze was not completed, and no prize was awarded. After the first trial, when it was clear that the participant had not completed the maze, he was asked a set of questions about the outcome. Embedded in those questions were two attributions about the unseen peer’s intent (e.g., “Do you think your partner meant to give you bad directions?”). Answers were reported on 7-point scales (1 = definitely no and 7 = definitely yes).

The task was also designed to assess the motivational strategies of intermediate risk taking and mastery orientation. Before the first trial, the participant was given the opportunity to choose the maze he would like to attempt from among 10 possible choices that ostensibly varied in difficulty. Mazes 1 to 3 were described as easy (“Almost everyone solves them correctly”), Mazes 4 to 7 were portrayed as of medium difficulty (“You can solve them, but you will have to think and try”), and Mazes 8 to 10 were described as hard (“Hardly anyone gets these right, not even big kids in high school”). After the first trial where “failure” was manipulated, the participant was asked to choose the next maze that he would like to attempt. Here we were able to examine whether respondents appropriately modified their aspiration level after a failure experience (i.e., chose an “easier” maze). Finally, a confederate observer, who was blind to both the hypotheses and children’s group assignments, unobtrusively wrote down everything that the participant said during the first trial where failure was manipulated. Those verbatim statements were open-coded and analyzed for evidence of mastery orientation.

Semester grades and teacher comments

At the end of the semester following the intervention, we examined each participant’s cumulative folder for that semester. Grade equivalents in language arts and math were calculated. In addition homeroom teachers’ open-ended comments about each student’s progress were coded for indicators of improving or deteriorating performance in the domains of social behavior and academic achievement.

Curriculum Implementation

The intervention was scheduled 3 days a week for 12 weeks in afterschool group sessions of four to six boys that lasted 1 hr each. Two African American female graduate students who participated in the development of the curriculum served as teachers. Five cycles of the intervention were completed over the course of 18 months.

Sample attrition

Across the five iterations, 9 of 31 boys in the treatment group did not complete all phases of the intervention: 1 had an extended hospital stay, 1 was asked to leave the program because of his continued disruptive behavior, and 7 were not included in the analysis due to irregular attendance. Among the controls, 10 of the initial 35 did not complete both pretesting and posttesting: 1 boy moved away, 2 were expelled from school, 3 voluntarily dropped out of the project, and 4 were chronically absent from school during posttesting. Thus, the final sample consisted of 47 third- to fifth-grade African American boys, 22 in the intervention and 25 in the control group. There was no significant difference in attrition rate as a function of treatment group and iteration, χ²(2) = 0.99, or on any of the selection criteria between participating boys and those who were dropped from the study.

Intervention fidelity

All 32 intervention lessons, including a detailed teaching guide and accompanying activities for each, were assembled in a predetermined order and placed in a color-coded binder (available from the first author). Based on extensive pilot testing, we designed the activities so that one lesson would be completed each day, with lessons from social and academic components interspersed to maximize participant interest. Both trainers were always present for each session and they shared the teaching responsibilities. At the end of each session, the trainers debriefed on the day’s activities and made adjustments where necessary (e.g., occasionally it was not possible to complete an entire lesson in one session). In addition, during the first two iterations, the principal investigators or an advanced graduate student independently observed 25% of the lessons and provided objective feedback about intervention fidelity. Across iterations there was no variation in the number of activities and lessons taught and little variation in the timing of each lesson.

Results

Overview

There were three parts to the data analysis. First, student self-report and teacher rating data collected at pretest and posttest were analyzed with a series of 2 × 2 (Treatment Group × Time) repeated-measures ANOVA. Intervention effects were indicated by a significant Group × Time interaction. Because we were primarily interested in treatment effects (i.e., change from pre- to posttest as a function of the intervention), repeated-measures ANOVA was the appropriate statistical test (Tabachnick & Fidell, 2013). Effect sizes using Cohen’s d are reported for significant interactions. With values of d ranging from 0 to 3, an effect size of .2 is judged as small, .5 as medium, and .8 or higher as large (Cohen, 1988). In the second part of the analysis, results are presented from the laboratory task that simulated ambiguous peer provocation and risk taking. In the third and final part, school grades and teacher comments from participants’ cumulative folders were analyzed.

Self-Report and Teacher Ratings

General attitudes about aggression and academic motivation

First, we examined whether or not there were significant changes in participants’ general attitudes about the legitimacy of aggression and academic motivation. The ANOVA on beliefs about the legitimacy of aggression yielded the predicted Group × Time interaction: F(1, 40) = 6.07, p < .05. After the intervention, boys who participated were significantly less likely to believe that aggression against others is legitimate: pretest M = 3.5, posttest M = 2.8, t(19) = 2.42, p < .05, d = .70. The difference between pre- and posttest scores for control group boys was not significant: Ms = 3.3 and 3.5 (t < 1). For the SAM, there were no differences in intervention or control group boys’ self-reported motivation before or after the intervention (F < 1 for the Group × Time interaction).

Strategic account giving

Using accounts

Recall that participants read a hypothetical story where they imagined that they transgressed against a peer and could then choose one of four possible accounts. Each respondent’s selected account at pretest and posttest received a score reflecting the adaptiveness of that choice (i.e., apology = 4, excuse = 3, denial = 2, indifference = 1). Assigned scores were based on the assumption that the most adaptive first choice would be to offer an apology with remorse, and the least adaptive strategy would be to acknowledge the transgression but with indifference and no remorse. The choice data were then examined in a 2 × 2 ANOVA. The hypothesized Group × Time interaction approached significance: F(1, 44) = 3.59, p = .07. Planned contrasts revealed that intervention boys showed a significant increase in the endorsement of adaptive accounts from pretest to posttest: t(20) = 2.06, p < .05, d = .65. For control group boys, however, the change in the pattern of account giving from pre- to posttest was not significant (t < 1).

Honoring accounts

Participants read hypothetical vignettes describing a provoking peer who offered one of three account types (apology, excuse, denial). Following each account, boys rated on 6-point scales the peer’s intent, how sorry that peer was for the transgression, and whether or not they would forgive him. Each of the three inference types was analyzed in a separate 2 × 3 (Time × Account Type) repeated-measures ANOVA. Because of the number and levels of the repeated factors relative to sample size, we chose to analyze the data separately for intervention and control group boys. The data for intervention boys are displayed in the top panels of Figure 1 and the data for control boys are shown in the bottom panels.

Figure 1.

Perceived intent, sorrow and forgiveness of transgressor as a function of account type for intervention and control group boys at pretest and posttest.

Among intervention boys, the Time × Account interaction was significant for perceived peer intent and sorrow: F(2, 38) = 3.81 and 4.06, respectively, ps < .05. The upper left panel of Figure 1 shows that at posttest more so than at pretest, hypothetical peers who apologized (d = .41) or offered an excuse (d = .19) were judged to have less hostile intent than peers who denied any wrongdoing at all (d = .21). There was a similar pattern to intervention boys’ inferences about how sorry the peer felt following the transgression (upper middle panel). Peers were judged to be more sorry for the transgression when they apologized (d = .48) or provided an excuse (d = .67) than when they denied the transgression (d = .32). However, participants’ reported willingness to forgive the peer did not change as a result of the intervention (F < 2 for the Time × Account interaction). As shown in the bottom panels of Figure 1, for control group boys, none of the effects involving time in the 2 × 3 repeated-measures ANOVAs were significant (F < 2 for all 3 interactions, ps > .05). That is, control group boys continued to be relatively unaware that particular kinds of accounts have different consequences for judgments of intent, feelings of sympathy, and forgiveness.

Attributions for achievement failure

At both pretest and posttest, participants were asked to recall the last time that they did poorly on a test and to rate on 7-point scales four possible explanations for their failure (low ability, lack of effort, a bad teacher, and bad luck). The two external attributional ratings were significantly correlated (r = .53) and therefore combined into a single index.

Figure 2 displays the pattern of failure attributions at pretest and posttest for intervention and control group boys. Each attribution was examined in a 2 × 2 (Group × Time) repeated-measures ANOVA. There were no significant effects for either of the internal attributions. All participants tended to report that failure was due to lack of effort and was not caused by low ability, and this was true at both the pretest and the posttest. For external attributions, however, there was a significant Group × Time interaction: F(1, 42) = 5.31, p < .001. As hypothesized, intervention boys were less likely to endorse external and uncontrollable causes for failure after participating in the intervention, t(19) = 3.35, p < .01, d = .95. For the control group, in contrast, there was no significant change across time for external attributions (t < 1). At the posttest, effort ratings were significantly higher for intervention boys than either ratings of low ability or external factors (ps < .01). Among control group boys, none of the posttest failure attribution ratings differed significantly from one another.

Figure 2.

Attributions for failure endorsed by intervention and control group boys at pretest and posttest.

Teacher ratings of social behavior and academic motivation

There were no effects of the intervention on teacher ratings of their students’ negative social behavior: F < 1 for the Group × Time interaction. However, for cooperation/motivation and persistence, the predicted interactions were found: for cooperation, F(1, 45) = 4.58, p < .05, d = .54; and for persistence, F(1, 45) = 10.33, p < .01, d = 1.23. Even though the two groups of boys were rated as equally cooperative at the posttest, the left panel of Figure 3 shows that teacher ratings increased significantly from pretest to posttest for intervention boys, t(21) = 1.92, p < .05, whereas there were no changes in ratings for control group boys, t(24) = 1.02, ns. Intervention boys were rated higher in persistence from pre- to posttest: t(21) = 2.60, p < .01, whereas ratings for boys in the control group actually declined: t(24) = 1.98, p < .05 (right panel).

Figure 3.

Teacher ratings of social behavior and academic behavior of intervention and control group boys at pretest and posttest.

Laboratory Analogue Task: Practicing Social Skills and Academic Motivation Skills

About 1 month after the intervention ended, all boys participated in a maze task that supposedly was unrelated to the intervention. The boys experienced goal frustration at the hands of an unseen peer partner, and there was some ambiguity about the partner’s responsibility for that outcome.

Attributions to hostile peer intent

After the first trial, when it was clear that the participant had not completed the maze, he was asked to rate on 7-point scales the unseen peer’s intent (“Do you think your partner meant to give you bad directions?” “tried to stop you from winning on purpose?”). The two intent questions were highly correlated (r = .68) and were therefore combined into a single index. Aggressive boys who participated in the intervention inferred significantly less hostile intent (M = 2.1) on the part of the unseen peer than did boys in the control group (M = 3.35): F(1, 45) = 5.11, p < .05, d = .67.

Risk taking before failure

Before the first trial, the participant was given the opportunity to choose the maze he would like to attempt from among 10 possible choices that ranged in difficulty. Adaptive choices (intermediate risk taking) would be indicated if participants initially chose one of the mazes described as moderately difficult. The number and percentage of boys in each group who initially selected an easy, medium, or difficult maze are shown in the top panel of Table 2. A chi-square test of this frequency table indicated that intervention boys were more likely to choose a maze of intermediate difficulty than were control group boys, χ²(2) = 8.22, p < .05. Eighteen of 22 intervention boys (82%) were intermediate risk takers compared with 4% who were low risk takers and 14% who were high risk takers. Among control boys, in contrast, less than half (11 of 25 boys, or 44%) chose a maze of intermediate difficulty, compared with 28% who were low risk takers and 28% who were high risk takers. In fact, the controls were fairly evenly distributed across the three difficulty levels.

Table 2.

Risk Taking Before “Failure” and Goal Setting After “Failure” on the Laboratory Task for Intervention and Control Group Boys.

	Treatment group
	Intervention n = 22	Control n = 25
	% n	% n
Risk taking (before “failure”)
Easy task	4 (1)	28 (7)
Medium task	82 (18)	44 (11)
Difficult task	14 (3)	28 (7)
Goal setting (after “failure”)
Unrealistically low (> 3 levels lower than first choice)	18 (4)	24 (6)
Realistic (within 3 levels of first choice)	64 (14)	32 (8)
Unrealistically high (higher than first choice)	18 (4)	44 (11)

Note. The data are presented as percentages within each treatment group. Numbers in parentheses are frequencies.

Risk taking after failure

Next, we examined how participants’ level of aspiration changed after their “failure” to successfully complete the first maze. Adaptive goal setting would be indicated by choosing a relatively easier puzzle (i.e., within three levels of their first choice). The bottom panel of Table 2 displays the goal setting patterns of the two groups of boys. The chi-square test of this frequency table approached significance: χ²(2) = 5.27, p = .07. Almost two thirds (64%) of intervention boys, compared with only one third (32%) of control group boys, chose a maze that was of intermediate difficulty for them in light of their prior failure. For the control group, there were higher percentages of boys who showed unrealistically low (24%) or unrealistically high (44%) shifts in their level of aspiration.

Verbal behavior during the maze task

While participants were working their way through the first “failure” trial, everything that they said was recorded by a female observer who was unobtrusively positioned close to the participant and who was blind to the hypotheses and children’s group assignment. Each phrase of these verbatim statements was open-coded by two independent graduate researchers who were also blind to the children’s group (Strauss & Corbin, 1998). The following four mutually exclusive categories emerged from the data:

Mastery focused: References to what needed to be done to complete the maze (e.g., “I can’t go up six blocks because there’s nothing there; I need another route”)

Negative self/task: negative comments about the task or one’s performance: (e.g., “this is stupid”; “this is boring”; “I don’t know what to do”)

Criticizing: Negative remarks to the unseen peer (e.g., “Are you deaf?”; “Cheater!”)

Irrelevant: Comments not focused on the task or the unseen peer (e.g., “I’m going to my friend’s house after school”)

Intervention boys reported a total of 104 codable comments compared with 70 on the part of boys in the control group. If boys in the intervention inferred less hostile intent on the part of the peer and showed adaptive motivational patterns in terms of their risk taking, then they should report fewer comments that criticized the peer and more comments that indicated a mastery focus. The data presented in Figure 4 show that this was indeed the case. A chi-square test of the different frequency patterns between the two groups was significant: χ²(3) = 14.38, p < .001. Two thirds (67%) of the intervention boys’ recorded statements were focused on mastering the task, compared with 14% that were negative comments, 16% that criticized the peer, and only 2% that could be classified as irrelevant. For control group boys, in contrast, the preference for mastery focus was not nearly as strong: 40% of comments classified as mastering the task, 33% as negative, 21% as critical of the peer, and 6% were irrelevant. Taken as a whole, intervention boys made significantly more mastery-oriented comments than did control boys and significantly fewer negative-task-related comments (critical z differences = 3.31 and 3.59, respectively, ps < .05).

Figure 4.

Percentage frequency of each comment type for intervention and control group boys on the laboratory maze task.

Cumulative Folder Data

The final source of data involved a search of participants’ end-of-semester cumulative folders for both grade equivalents and teacher comments about progress during that term. Homeroom teachers evaluated a number of skill areas according to three criteria: area of strength, shows growth, or needs improvement. We calculated a grade equivalent in math and language arts for each participant by assigning a score of 2 to any skill that was judged “an area of strength,” a 1 for “shows growth,” and a 0 for “needs improvement.” With eight skill areas in language arts, grade equivalents could therefore range from 0 to 16. The seven skill areas in math allowed a range of grade equivalents from 0 to 14. The analysis revealed no difference between intervention and control group boys in semester grade equivalents for either language arts (Ms = 7.35 vs 7.59) or Math (Ms = 5.80 vs 6.36).

We used deductive analysis to examine the open-ended comments that homeroom teachers wrote about the progress of their students at the end of the semester. Two independent graduate researchers who were blind to children’s treatment group content analyzed each phrase according to (a) whether or not it referenced the student’s social and/or academic behavior and if so, (b) whether the comment was positive or negative. Teacher comments were then rated on a 5-point scale ranging from +2 to −2. A score of +2 was given to comments that suggested positive change in both social behavior and study skills over the term immediately following the intervention; a score of +1 revealed improvement in one of these areas; a score of 0 was assigned to neutral comments (e.g., “good luck and work hard in the years to come”); −1 to deteriorating performance in either the social or academic domain; and −2 to declines in both areas. Coding discrepancies were resolved by the second author. Inter-rater reliability was high (Kappa = .95).

Table 3 shows the frequency of each comment type for the two groups of boys. The association between comment type and treatment group was significant: χ²(4) = 18.76, p < .01. What stands out most clearly here is the different pattern of responses in the two groups for very positive comments (+2, improvement in both areas) and very negative comments (−2, deterioration in both areas). Eight of 20 intervention boys (40%) received the most positive teacher comments, while no control group boys achieved this level of teacher evaluation. In contrast, 9 of 23 control boys (41%) had comments classified as most negative, whereas such disparaging remarks were detected in the cumulative folder of only 1 intervention boy.

Table 3.

Percentage of Intervention and Control Boys With Positive, Neutral, and Negative Written Teacher Comments in Their Cumulative Folders.

	Treatment group
Comment type	Intervention % n	Control % n
+2 positive social and study skills	40 (8)	0 (0)
+1 positive in one area	20 (4)	27 (6)
0 neutral	10 (2)	9 (2)
−1 negative in one area	25 (5)	23 (5)
−2 negative in both social and study skills	5 (1)	41 (9)

Note. n = 20 in the intervention group and n = 22 in the control group.

Discussion

Many school-based interventions address aggression and related conduct problems (e.g., Leff, Waasdorp, & Crick, 2010), and a number have specifically targeted African American youth in urban schools (see reviews in Hudley & Graham, 1995; Huey & Polo, 2008). However, multicomponent aggression reduction programs typically focus on academic enrichment, as exemplified by Fast Track (Conduct Problems Prevention Research Group, 2011), or teacher training as exemplified by the Metropolitan Area Child Study (Metropolitan Area Child Study Research Group, 2007; Tolan & McKay, 1996; see Lochman & Wells, 2004, for an exception). The intervention described here is unique in its specific theoretical focus on increasing students’ own self-regulation and academic motivation rather than increasing instructional time or altering teachers’ instructional strategies. Although any population of elementary age students at risk for school failure might benefit from such an intervention program, we developed our intervention for the members of the population whom we perceived to be most vulnerable: low achieving urban African American males labeled as aggressive.

To our knowledge, this is the first successful intervention with aggressive youth that blended social skills training with motivation skills training under one unifying theoretical framework. Also noteworthy is the fact that the intervention provided a protective, semester-long structured activity for academically at-risk youngsters during a portion of the afterschool hours—a time when unsupervised time with peers poses the risk of engagement in delinquent activity (Rorie, Gottfredson, Cross, Wilson, & Connell, 2011). The basic elements of the intervention were organized around the causal construct of perceived responsibility in others and the self. Because perceived responsibility is a core construct in attribution theory, our intervention both complements and extends prior intervention research based on attributional change in both the social and academic domains.

In the social domain, our earlier work using the Brain Power program documented improvements in the behavior of aggressive boys who had learned to accurately infer the intentions of others (Hudley, 2008). We replicated those intervention effects in the present study. Compared with the control group at posttest, intervention boys were less likely to infer hostile intent in an unseen peer following ambiguously caused provocation and less likely to endorse aggression as a legitimate way to deal with provocation. We also incorporated the important skills of giving and honoring accounts that shifted the focus toward helping aggressive youth change the inferences that other people make about them. A myriad of social cognitive factors have been enlisted to explain how aggressive children become entangled in a cycle of maladaptive beliefs and behaviors that operate to perpetuate poor relations with peers (Coie & Dodge, 2006). To that list, we propose to add the understanding (or lack thereof) of account giving as a strategy for managing the attributions of others and controlling their anger following social misconduct. We think that training in the appropriate use of accounts such as excuses and apologies and other impression management strategies is an important new direction for intervention research.

In the academic domain, the focus on self-responsibility for achievement outcomes complements a large literature on the positive effects of attribution retraining—that is, teaching failure-prone students to attribute their failures to factors within their control, such as lack of effort (Wilson et al., 2002). Also, we were able to incorporate training in other kinds of motivational skills that promote self-responsibility for achievement, such as intermediate risk taking, proximal goal setting, and mastery orientation. That risk taking, goal setting, and mastery focus are amenable to change is a meaningful finding, for there is very little empirical research on motivational change aside from that targeted specifically at attributions and related causal constructs (see Wagner & Szamoskozi, 2012). Because intervention research has not kept pace with theoretical development in the field of motivation, we turned to the historically dominant motivation theorists who studied phenomena relevant to our intervention. Those theorists laid the foundation for our motivation curriculum in their classic experimental studies of level of aspiration, risk taking, and the achievement motive (see Graham & Weiner, 2011). The motivation component of the intervention taught specific behaviors identified by early theorists, and the laboratory maze task provided a context where those behaviors could be practiced. Although the intervention did not yield improvements in participants’ actual grades, there were other indicators of academic improvement in teacher ratings of persistence and their open-ended comments at the end of the semester. A challenge for future intervention research will be to incorporate newly acquired motivation skills into the regular academic curriculum.

Capitalizing on the regular curriculum suggests that the most effective interventions will be multicomponent with multiple lessons building upon one another over an extended time. However, a number of highly publicized recent interventions emerging from social psychology research have increased excitement about the potential of brief, even single-session treatments. These interventions, some targeted toward African American youth, utilize constructs such as stereotype threat, mind-sets, and self-affirmation to deliver short but powerful treatments that not only boost immediate achievement but also reduce the racial achievement gap (see review in Yeager & Walton, 2011). We are firm believers in theory-guided interventions and we applaud the social psychologists engaged in new intervention approaches that can better uncover the mechanisms underlying motivational change. However, we are less convinced that changing one set of beliefs—be it worries about confirming racial stereotypes, implicit theories about intelligence, or the importance of affirming personal values—will have lasting effects on motivation and achievement. Our approach targets multiple beliefs about oneself and other people and acknowledges the close intersection between students’ academic lives and their social lives. We doubt that enduring changes in multiple beliefs and behavior of truly at-risk youth can be achieved with very brief interventions no matter how powerfully they are delivered.

Limitations of the Research

Multicomponent intervention research with high-risk populations is often difficult to carry out and we acknowledge the limitations of the study reported here. First, our design did not include a placebo group (i.e., youth who participated in a comparable afterschool program that did not target social and motivation skills). Thus, we cannot rule out the possibility that some of the effects were due to the extra attention that intervention boys received, independent of curriculum content. Participant attrition was a second limitation because about 25% of the intervention boys did not complete all phases of the program. It will be important to develop more effective strategies for reducing participant attrition in an afterschool program, including working more closely with parents to help sustain their child’s commitment to the intervention.

Related to attrition, the intervention was not successful for all participating boys, and the sample was too small to investigate the heterogeneity of participants or to track baseline by treatment interactions. With our selection criteria, high-risk boys in this study were aggregated together in treatment groups. The social reinforcement that group members receive from one another for acting out sometimes functions as a kind of deviancy training that can actually result in increased problem behavior (Rorie et al., 2011). One solution is to have mixed intervention groups that include a balance of high- and low-risk boys.

The absence of follow-up is a fourth limitation. We do not know whether the intervention had any lasting effects beyond the end of the school semester following implementation. That is particularly important given our interests in the effect of the intervention on more general outcomes such as academic performance and attitudes about school. At the level of general attitudes, the intervention was more successful at improving beliefs about socially responsible behavior than beliefs about academically responsible behavior. We suspect that academic outcomes are part of more cumulative intervention effects that unfold gradually over time.

Implications for Urban Education

Best Foot Forward was carried out in an urban elementary school in one of the largest school districts in the nation. Serving K-5, the school encountered many of the challenges associated with large urban districts: a high concentration of ethnic minority students, high poverty (more than 90% of students qualified for free or reduced-price lunch), a third of the teachers not fully credentialed, and a rank in the bottom 10% of schools on statewide academic performance indicators. In other words, ours was a challenged urban school as defined in the literature (e.g., Milner, 2009). Many contemporary approaches to research on urban schooling view these difficulties as a product of institutional racism. To varying degrees, reform efforts tackle racism head on with the goal of empowering the oppressed and the marginalized, especially Black males (e.g., Ladson-Billings, 2009).

Best Foot Forward also has empowerment as a goal. We made special efforts to develop activities that complemented the learning styles and life experiences of urban African American boys. The theory-based content of the curriculum targeted the social and academic challenges faced by African American youth. For example, practicing better self-presentational and impression management skills is a valuable social competency for youth who must deal with the biases of decision makers in both the school and the juvenile justice arena. Learning to better gauge the relation between effort and outcomes, as well as realistic versus unrealistic goals, can be a useful antidote to the perceived barriers to opportunity that can have debilitating effects on ethnic minority youth in urban schools (Taylor & Graham, 2007). And Best Foot Forward might serve as a preventive intervention as youth prepare for the transition to middle school. A great deal of research indicates that the transition to middle school is accompanied by declines in motivation and school engagement, especially for ethnic minority youth in urban schools, and from which some youth never recover (Eccles & Roeser, 2011). Data on interventions that specifically buffer secondary school transitions are virtually nonexistent.

Finally, Best Foot Forward was aligned with known guidelines of good prevention science in that it was theory-guided, used random assignment, monitored treatment fidelity, and gathered multiple sources of data from multiple informants—elements that are essential for rigorous evaluation (Flay et al., 2005). We therefore blended the rigor of prevention science with an intervention that was situated in an urban context. Our intervention approach (including its strengths and limitations) highlights fruitful strategies for teachers, psychologists, and related personnel to advance students’ social and academic proficiencies and serves as a springboard for future studies aimed at promoting the academic and social skills of at-risk students in urban schools.

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: This work was supported by the John and Dora Haynes Foundation.

Author Biographies

Sandra Graham is the University of California Presidential Chair in Education and Diversity and a professor of Human Development and Psychology in the Department of Education at UCLA. Dr. Graham’s teaching interests include achievement motivation, attribution theory, social development, and risk and resiliency in African American youth. Her research interests are in the areas of cognitive approaches to motivation, the development of attributional processes, motivation in African Americans, and peer-directed aggression and victimization.

April Taylor came to the Department of Child and Adolescent Development in 2004 after receiving her doctorate in Educational Psychology from the University of California, Los Angeles and completing an AERA/IES postdoctoral fellowship hosted at the Rossier School of Education at the University of Southern California. She is a professor at California State University, Northridge. Taylor’s research focuses on the development of motivation and more specifically concerns motivation for academic achievement among urban minority youth. Her research program examines social, cultural, and social-psychological influences on achievement motivation, and the identification of theory guided motivation-enhancing practices.

Cynthia Hudley, PhD, is a professor emerita at the University of California, Santa Barbara. Hudley’s work has focused on two distinct but interrelated areas: academic achievement motivation and children’s peer-directed aggression. Much of Hudley’s scholarly activities have been dedicated to understanding children’s aggressive behaviors across samples of students that vary by social class, gender, ethnicity, age and region of residence.

References

American College Testing. (1993). School Attitude Measure. Iowa City, IA: Author.

Atkinson

(1964). An introduction to motivation. Princeton, NJ: Van Nostrand.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.

Bradshaw

C. P.

Schaeffer

C. M.

Petras

Ialongo

(2010). Predicting negative life outcomes from early aggressive-disruptive trajectories: Gender differences in maladaptation across life domains. Journal of Youth Adolescence, 39, 953-966. doi:10.1007/s10964-009-9442-8

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

Coie

Dodge

(1988). Multiple sources of data on social behavior and social status in the school: A cross-age comparison. Child Development, 59, 815-829.

Coie

Dodge

(2006). Aggression and antisocial behavior. In Eisenberg

(Ed.), Handbook of child psychology: Social, emotional, and personality development (Vol. 3. pp. 779-862). New York, NY: John Wiley.

Conduct Problems Prevention Research Group. (2011). The effects of the Fast Track preventive intervention on the development of conduct disorder across childhood. Child Development, 82, 331-345. doi:10.1111/j.1467-8624.2010.01558.x

Davis

Gold

(2011). An examination of emotional empathy, attributions of stability, and the link between perceived remorse and forgiveness. Personality and Individual Differences, 50, 392-397. doi:10.1016/j.paid.2010.10.031

10.

DeCharms

(1972). Personal causation training in the schools. Journal of Applied Social Psychology, 2, 95-113. doi:10.1111/j.1559-1816.1972.tb01266.x

11.

Eccles

Roeser

(2011). Schools as developmental contexts during adolescence. Journal of Research on Adolescence, 21, 225-241.

12.

Elliot

(2005). Handbook of competence and motivation. New York, NY: Guilford Press.

13.

Ferguson

(2003). Bad boys: Public schools in the making of Black masculinity. Ann Arbor: University of Michigan Press.

14.

Flay

B. R.

Biglan

Boruch

R. F.

Castro

F. G.

Gottfredson

Kellam

. . . Ji

(2005). Standards of evidence: Criteria for efficacy, effectiveness, and dissemination. Prevention Science, 6, 151-175.

15.

Graham

Weiner

(2011). Motivation: Past, present, future. In Harris

Graham

Urdan

(Eds.), APA Educational Psychology Handbook (vol. 1, pp. 367-397). Washington, D.C.: American Psychological Association.

16.

Graham

Weiner

Benesh-Weiner

(1995). An attributional analysis of the development of excuse-giving in aggressive and nonaggressive African American boys. Developmental Psychology, 31, 274-284.

17.

Graham

Williams

(2009). An attributional approach to motivation in school. In Wentzel

Wigfield

(Eds.), Handbook of motivation at school (pp. 11-33). Hillsdale, N.J.: Lawrence Erlbaum.

18.

Gresham

Elliot

(1990). Social Skills Rating System (teacher form). Circle Pines, MN: American Guidance.

19.

Howard

(2014). Black male(d): Peril and promise in the education of African American males. New York, NY: Teachers College Press

20.

Hudley

(2008). You did that on purpose: Understanding and changing children’s aggression. New Haven, CT: Yale University Press.

21.

Hudley

Graham

(1993). An attributional intervention to reduce peer-directed aggression among African American boys. Child Development, 64, 124-138.

22.

Hudley

Graham

(1995). School-based interventions for aggressive African American boys. Applied & Preventive Psychology, 4, 185-195.

23.

Huey

Polo

(2008). Evidence-based psychosocial treatments for ethnic minority youth: A review and meta-analysis. Journal of Clinical Child & Adolescent Psychology, 37, 262-301.

24.

Ladson-Billings

(2009). Race still matters: Critical race theory in education. In Apple

Gandin

(Eds.), The Routledge international handbook of critical education (pp. 110-122). New York, NY: Taylor & Francis.

25.

Leff

S. S.

Waasdorp

T. E.

Crick

N. R.

(2010). A review of existing relational aggression programs: Strengths, limitations, and future directions. School Psychology Review, 39, 508-535.

26.

Lochman

J. E.

Wells

K. C.

(2004). The coping power program for preadolescent aggressive boys and their parents: Outcome effects at the 1-year follow-up. Journal of Consulting and Clinical Psychology, 72, 571-578. doi:10.1037/0022-006X.72.4.571

27.

Loeber

Farrington

Stouthamer-Loeber

Moffitt

Caspi

White

. . . Beyers

(2003). The development of male offending: Key findings from fourteen years of the Pittsburgh Youth Study. In Thornberry

Krohn

(Eds.), Taking stock of delinquency: An overview of findings from contemporary longitudinal studies (pp. 93-136). New York, NY: Plenum.

28.

McLoyd

(1985). Are toys (just) toys? Exploring their effects on pretend play of low income preschoolers. In Spencer

M. B.

Brookins

G. K.

Allen

W. R.

(Eds.), Beginnings: The social and affective development of Black Children (pp. 81-100). Hillsdale, NJ: Lawrence Erlbaum.

29.

Metropolitan Area Child Study Research Group. (2007). Changing the way children “think” about aggression: Social-cognitive effects of a preventive intervention. Journal of Consulting and Clinical Psychology, 75, 160-167. doi:10.1037/0022-006X.75.1.160

30.

Milner

(2009). Preparing teachers of African American students in urban schools. In Tillman

(Ed.), The SAGE handbook of African American education (pp. 123-140). Thousand Oaks, CA: SAGE.

31.

National Center for Educational Statistics. (2013). Digest of education statistics. Retrieved from http://nces.ed.gov/programs/digest/2013menu_tables.asp

32.

Noguera

(2009). The trouble with Black boys: And other reflections on race, equity, and the future of public education. San Francisco, CA: Jossey-Bass.

33.

Risi

Gerhardstein

Kistner

(2003). Children’s classroom peer relationships and subsequent educational outcomes. Journal of Clinical Child & Adolescent Psychology, 32, 351-361. doi:10.1207/S15374424JCCP3203_04

34.

Roderick

(2003). What’s happening to the boys? Early high school experiences and school outcomes among African American male adolescents in Chicago. Urban Education, 38, 538-607.

35.

Rorie

Gottfredson

Cross

Wilson

Connell

(2011). Structure and deviancy training in after-school programs. Journal of Adolescence, 34, 105-117. doi:10.1016/j.adolescence.2010.01.007

36.

Rumberger

(2011). Dropping out: Why students drop out of high school and what can be done about it. Cambridge, MA: Harvard University Press.

37.

Scott

Lyman

(1968). Accounts. American Sociological Review, 23, 46-62.

38.

Slaby

Guerra

(1988). Cognitive mediators of aggression in adolescent offenders I: Assessment. Developmental Psychology, 24, 580-588.

39.

Strauss

Corbin

(1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: SAGE.

40.

Tabachnick

Fidell

(2013). Using multivariate statistics. Needham heights, MA: Allyn & Bacon.

41.

Taylor

Graham

(2007). An examination of the relationship between achievement values and perceptions of barriers among low-SES African American and Latino students. Journal of Educational Psychology, 99(1), 52-64. doi: 10.1037/0022-0663.99.1.52

42.

Tolan

P. H.

McKay

M. M.

(1996). Preventing serious antisocial behavior in Inner-City children: An empirically based family intervention program. Family Relations, 45, 148-155. doi:10.2307/585285

43.

Wagner

Szamoskozi

(2012). Effects of direct academic motivation-enhancing intervention programs: A meta-analysis. Journal of Cognitive and Behavioral Psychotherapies, 12, 85-101.

44.

Weiner

(1986). An attributional theory of motivation and emotion. New York, NY: Springer-Verlag.

45.

Weiner

(1995). Judgments of responsibility: A foundation for a theory of social conduct. New York, NY: Guilford Press.

46.

Weiner

(2005). Motivation from an attributional perspective and the social psychology of perceived competence. In Weiner

(Ed.), Handbook of competence and motivation (pp. 73-84). New York, NY: Guilford Press.

47.

Whitley

Kite

(2006). The psychology of prejudice and discrimination. Belmont, CA: Thomson.

48.

Wilson

Damiani

Shelton

(2002). Improving the academic performance of college students with brief attributional interventions. In Wilson

Damiani

Shelton

(Eds.), Improving academic achievement: Impact of psychological factors on education (pp. 89-108). San Diego, CA: Academic Press.

49.

Yeager

D. S.

Walton

G. M.

(2011). Social-psychological interventions in education: They’re not magic. Review of Educational Research, 81, 267-301.

50.

Young

Thompson

(2011). Gender, attributions of responsibility, and negotiation of deviant labels in small groups. Deviant Behavior, 32, 626-652. doi:10.1080/01639625.2010.514203