Journal of School Psychology 53 (2015) 179–194
Contents lists available at ScienceDirect
Journal of School Psychology journal homepage: www.elsevier.com/locate/jschpsyc
A qualitative analysis of factors influencing middle school students' use of skills taught by a violence prevention curriculum☆,☆☆ Albert D. Farrell ⁎, Krista R. Mehari 1, Alison M. Kramer-Kuhn 2, Sally A. Mays 3, Terri N. Sullivan 4 Virginia Commonwealth University, United States
a r t i c l e
i n f o
Article history: Received 25 August 2014 Received in revised form 2 February 2015 Accepted 1 March 2015 Available online xxxx Keywords: Violence prevention Qualitative analysis Adolescence Social–emotional learning Aggression Bullying prevention
a b s t r a c t This study examined factors that influenced the use of skills taught in a school-based universal violence prevention program. Interviews were conducted with 91 students from two urban schools (83% were African American and 12% multiracial) and 50 students from a nearby county school (52% were White, 32% African American, and 12% multiracial). About half the sample (54%) was male. All had been in sixth grade classrooms where the Second Step (Committee for Children, 1997b) violence prevention curriculum had been implemented earlier in the school year or in the preceding school year. Qualitative analysis of interview transcripts suggested that participants' use of intervention skills was influenced by their beliefs and values, perceived relevance and effectiveness of the skill, issues related to enacting the behavior, and contextual factors. These findings highlight the need for a more intensive and comprehensive effort to address barriers and supports that influence the relevance and impact of school-based violence prevention programs. © 2015 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved.
1. Introduction Schools are a frequent venue for prevention efforts that focus on reducing youth involvement in problem behaviors such as aggression (Farrell & Camou, 2006; Gottfredson & Gottfredson, 2001). Several factors make schools a natural setting for such efforts. Schools provide a primary setting in which peer groups form and conflicts can develop. This is especially true during middle school, when students may first encounter and be particularly susceptible to the influence of delinquent peers (Moffitt, 1993). Factors within the school climate may also increase adolescents' risk for aggression. Within some schools, informal norms may support aggression as ☆ This study was funded by the National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, CDC Cooperative Agreements 5U49CE000730 and 5U01CE001956. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. ☆☆ The authors appreciate the contributions of Amie Bettencourt, Anh-Thuy Le, Denicia Titchner, and Kelly Zbojovsky who contributed to the study through coding transcripts, and refining the theme definitions and framework. ⁎ Corresponding author at: Department of Psychology, Virginia Commonwealth University, P.O. Box 842018, Richmond, VA 23284–2018, United States. Tel.: +1 804 828 8796; fax: +1 804 827 1511. E-mail addresses:
[email protected] (A.D. Farrell),
[email protected] (K.R. Mehari),
[email protected] (A.M. Kramer-Kuhn),
[email protected] (S.A. Mays),
[email protected] (T.N. Sullivan). Action Editor: Michelle K. Demaray 1 Present address: Virginia Treatment Center for Children, 515 N. 10th St., Richmond, VA 23298, United States. 2 Present address: Child Abuse Program, Children's Hospital of The King's Daughters, 935 Redgate Avenue, Norfolk, VA 23507, United States. 3 Present address: Loyola University, 4501 N. Charles St., Baltimore, MD, 21210, United States. 4 Department of Psychology, Virginia Commonwealth University, P.O. Box 842018, Richmond, VA 23284–2018, United States.
http://dx.doi.org/10.1016/j.jsp.2015.03.001 0022-4405/© 2015 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved.
180
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
a means of achieving social status and correcting perceived wrongs (Fagan & Wilkinson, 1998; Henry, Farrell, Schoeny, Tolan, & Dymnicki, 2011). Conversely, schools can provide positive experiences for youth. Teachers and other staff members can implement programs designed to develop students' social and emotional skills, model these skills, and support their use (Gottfredson & Gottfredson, 2001). Schools can also institute policies and practices to discourage aggression and encourage prosocial behavior. Finally, at a practical level, schools provide access to the majority of school-aged youth. Although universal school-based violence prevention programs have produced some promising results, further work is needed to improve their effectiveness (Centers for Disease Control & Prevention [CDC], 2007). Meta-analyses have concluded that the majority of youth violence prevention programs have limited benefits. A recent meta-review indicated that about half of meta-analyses have found only moderate intervention effects, one-quarter reported weak effects, and only one-tenth found strong effects (Matjasko et al., 2012). A meta-analysis that focused specifically on universal school-based programs indicated that their overall effect was a 15% reduction in violent behaviors (CDC, 2007). Compared with programs targeting other age groups, effects for middle school students were more modest (i.e., a 7% reduction). Other reviewers have concluded that even when initial intervention effects have been found, they have typically not been sustained over time (Mytton, DiGuiseppi, Gough, Taylor, & Logan, 2002; Wilson, Lipsey, & Derzon, 2003). This variability in intervention effects may partially reflect the influence of individual-, school-, and communitylevel moderators (Farrell, Henry, & Bettencourt, 2013). Thus, although school-based prevention programs clearly have promise, further work is needed to improve their relevance and impact for specific groups of adolescents. The majority of school-based violence prevention programs are based on social–emotional learning (SEL) principles; that is, they are designed to improve adolescents' cognitive, social, and emotional skills such as empathy, problem-solving, and anger management (Boxer & Dubow, 2002). This approach is based on a social information-processing model that views aggressive behavior as a function of deficits in these skills along with beliefs and values that support aggression (Crick & Dodge, 1994; Huesmann, 1998). The assumption is that addressing skill deficits and altering the database that influences decisions about how to respond in a specific situation will reduce youth involvement in aggression. Ladd and Mize (1983) identified three training objectives for SEL programs: enhancing skill concepts, promoting skill performance, and fostering skill maintenance and generalization. The third goal, maintenance and generalization, identifies the need for children to transition successfully from practicing the skill during lessons to using the skill in real life. It is difficult to measure actual generalization of SEL skills directly. Instead, investigators tend to measure distal outcomes such as aggression or prosocial behavior, and simply assume that the mechanism of change was the use of social and emotional skills in daily interactions (e.g., Bierman et al., 2010). A clearer understanding of the extent to which participants in SEL interventions acquire the targeted skills and generalize them to situations they encounter in their everyday lives could guide efforts to improve their impact. This social–emotional focus is reflected in Second Step, a universal school-based violence prevention program with curricula for elementary (Committee for Children, 1997a) and middle school students (Committee for Children, 1997b). The middle school foundational curriculum consists of 15 lessons that address skills related to empathy and perspective-taking, social problem solving, and anger management, and the application of these skills to dealing with bullying, diffusing a fight, addressing peer and gang pressure, and using assertiveness to make a complaint (Committee for Children, 1997b). The U.S. Department of Education, the U.S. Department of Health and Human Services, and Communities that Care have each designated Second Step as a “best practice” or “model” program, and it has been widely implemented in schools across the U.S. (Cooke et al., 2007). Support for the efficacy of the Second Step middle school curriculum has been somewhat mixed. Some studies have not found reductions in problem behaviors among participants (e.g., McMahon & Washburn, 2003; Orpinas et al., 2000), suggesting problems with skill mastery or generalization. Others have found initial positive outcomes that were not maintained over time, suggesting problems with skill maintenance (e.g., Orpinas, Parcel, McAlister, & Frankowski, 1995). One of the more rigorous evaluations of the Second Step Middle School curriculum was conducted by Espelage, Low, Polanin, and Brown (2013), who randomized 18 matched pairs of middle schools to intervention and control conditions. Although they found significant intervention effects at an initial posttest on selfreported measures of physical aggression, they did not find effects on other key outcomes including verbal and relational aggression, peer victimization, homophobic teasing, and sexual violence. Effects have also been found to vary across subgroups of students. Sullivan, Sutherland, Farrell, and Taylor (in press) evaluated the impact of Second Step in a study that randomized 28 classrooms in two urban middle schools and a school in a neighboring county to intervention and control conditions. Although analyses of pretest-to-posttest changes did not indicate any main effects for the intervention, some effects were found for subgroups defined by disability status and gender. A major goal of Second Step and similar school-based interventions is for participants to not only master the skills taught, but generalize them by using them in relevant, real-life situations. Surprisingly little research has tested this assumption. Edwards, Hunt, Meyers, Grogg, and Jarrett (2005) interviewed 113 fourth and fifth grade students in an urban school district who participated in Second Step. Nearly all of the students (96%) provided an example of when they used one of the specific skills taught by the intervention. The majority of these examples (66%) involved the use of anger management skills. Fewer students described situations in which they used other targeted skills such as empathy (14%) and problem solving (6%). Grumm, Hein, and Fingerle (2012) examined the social validity of a German adaptation of Second Step through interviews with 117 fourth grade students. Students were asked how they liked the intervention, how useful they found it, the extent to which they used the skills addressed by the intervention, and which aspects of the intervention they liked and disliked. Half the participants found the intervention useful and identified situations in which they used what they learned, 30% found it useful but were not able to provide specific examples, and 20% did not find it useful. Farrell, Mehari, Mays, Sullivan, and Le (in press) investigated participants' perceptions of the Second Step middle school curriculum in a mixed-methods study in which they interviewed 141 students from three middle schools who had participated in Second Step as part of a randomized trial. Interviewers asked participants about their overall evaluation of the intervention, the extent to
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
181
which it changed their behavior, the degree to which they used the eight specific skills targeted by the intervention to deal with problem situations, and whether they found the skills effective. Most participants described changes in at least one aspect of their behavior that they attributed to the intervention (e.g., controlling anger or improving interpersonal relationships). Although participants' responses supported the relevance of most of the skills addressed by the intervention, the percentages of students who reported attempting to use specific skills varied greatly. Most students reported using anger management (82%) and empathy (70%) at least once since completing the intervention. In contrast, the majority of students (87%) never used skills for resisting gang pressure; half (50%) never used skills for dealing with bullying; and almost half (46%) never used skills for dealing with peer pressure. Interventions such as Second Step will have limited effects unless participants not only master the skills targeted by the intervention, but actually continue to use them in their everyday lives (Ferguson, San Miguel, Kilburn, & Sanchez, 2007). Previous qualitative studies have suggested that there is considerable variability in the extent to which individuals report using the skills addressed by Second Step (Edwards et al., 2005; Farrell et al., in press; Grumm et al., 2012). This underscores the need for further research to identify factors that influence participants' use and maintenance of skills that are taught in Second Step and similar school-based interventions. Qualitative methods appear particularly well suited for addressing these issues by providing a means to examine adolescents' perspectives on the factors that support or impede their use of intervention skills in their daily lives. The purpose of this study was to identify factors that make adolescents more or less likely to generalize the social and emotional skills taught in Second Step by using them when they encounter problem situations. More specifically, qualitative analyses were conducted on segments of the interviews Farrell et al. (in press) conducted with adolescents who were in sixth grade classrooms where the Second Step intervention was implemented. Whereas Farrell et al. examined participants' overall reactions to the intervention and how frequently they used each intervention skill, the present study analyzed segments of the interview specifically designed to identify factors that would influence participants' generalization of each skill. A clearer understanding of these factors could inform efforts to strengthen supports and reduce barriers for using social and emotional skills taught by Second Step and similar violence prevention programs, and thereby improve their impact. 2. Method 2.1. Settings and participants Interviews were conducted with students from three middle schools who had previously participated in the Sullivan et al. (in press) outcome study. Two were in the public school system of a large city in the Southeastern U.S. The third was in an adjacent county in an area classified as “Rural Fringe” by the Census Bureau. Sullivan et al. obtained parental consent and student assent to participate in their outcome study and in the present study from 82% of the eligible students in the two urban schools and 75% of those in the county school. All participants in the current study were in classrooms assigned to the intervention condition and had attended class on at least half the days the Second Step curriculum was implemented. The university's institutional review board approved all procedures. The majority of students (83%) in the two urban schools were eligible for the federally-subsidized school lunch program. Interviews were conducted with 90 of 102 eligible students from these schools (41 in one school and 49 in the other). This did not include 23 students who left the school or were absent on more than half the days the intervention was implemented. The final sample of 90 participants represented 88% of those eligible. Most (i.e., 76%) were in class on at least 80% of the days the intervention was implemented. There were slightly more boys (54%) than girls and the average age was 11.8 (SD = 0.58). Most (83%) identified themselves as Black or African American; 12% endorsed more than one race. Only one participant self-identified as Hispanic or Latino. Participants most frequently indicated that they lived with both parents (24%); 20% lived with a single mother without other adults; 19% lived with a mother and stepfather; 17% lived with a single mother and another adult; and 6% lived with grandparents without a father or mother. The county school served a more diverse student population, with 22% of students eligible for the federally-subsidized school lunch program. Because of the larger number of eligible students, participants were recruited from a random sample of 66 of the 96 students in intervention classrooms. This did not include eight students who were absent on more than half the days the intervention was implemented in their classrooms. The final sample of 50 represented 76% of those eligible. Nearly all (i.e., 96%) were in class on at least 80% of the days the intervention was implemented. There were slightly more boys (54%) than girls, and the average age was 11.9 years old (SD = 0.78). Over half (52%) identified themselves as White, 32% as Black or African American, and 12% as multiracial. None indicated they were Hispanic or Latino. A majority (58%) lived with both parents, 19% lived with a single mother without another adult, and 10% lived with a mother and stepfather. 2.2. Study design The design of the study incorporated key elements of consensual qualitative research (Hill, Thompson, & Williams, 1997). These included an interview protocol that used open-ended questions, a focus on the words participants used to describe their experiences and their context, an inductive approach to identify emerging themes, and coding by a team of researchers using an iterative coding process to determine the degree to which emerging themes accurately represented the data. The approach to data collection was also consistent with consensual qualitative research, which involves identifying the sample and collecting all of the data prior to initiating coding. One limitation of this strategy is that it requires determining the sample size in advance rather than continuing data collection until saturation (i.e., the point at which no new information emerges) is reached. A fairly large number of adolescents was sampled to
182
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
ensure that sufficient data were collected. This made it possible not only to identify factors that represented experiences common to all participants, but factors reflecting the experiences of subgroups of adolescents. This was driven by the findings of a recent review (Farrell et al., 2013) that documented variability in adolescents' responses to school-based violence prevention programs such as Second Step.
2.3. Intervention Second Step is a universal school-based violence prevention curriculum designed for middle and junior high school students, with the goal of reducing impulsive and aggressive behavior by building participants' social and emotional competencies (Committee for Children, 1997b). Fifteen lessons address dealing with interpersonal conflicts through the use of skills such as empathy and perspective-taking, anger management, and problem-solving. Students learn to apply these skills to making a complaint (i.e., assertively talking to someone that bothered or hurt them), resisting peer and gang pressure, dealing with bullying, and diffusing a fight (Committee for Children, 1997b). Sullivan et al. (in press) also implemented an introductory session and incorporated role plays based on problem situations derived from previous qualitative research with urban middle school students (Farrell et al., 2007). The introductory session focused on prosocial behaviors and social networks. The main goals were for students to be able to define and understand the terms prosocial behavior and social networks, practice using prosocial behaviors toward their classmates, and define their own social network of people who supported them (e.g., who they felt close to and connected with). Sullivan et al. (in press) randomly assigned 12 physical education classes within the county school, and 12 physical education and four enrichment classes in the urban schools, to intervention and control conditions. Students in the county school were taught physical education throughout the school year. Second Step was implemented in these classes once or twice a week over a 9 to 12 week period from December 2008 to March 2009. Within each of the two urban schools most students took physical education during either the first or second semester. Students not enrolled in physical education were placed in enrichment classes. The intervention was implemented in fall classes between October and December, 2008, and in spring classes between February and April of 2009. Within physical education classes, it was generally implemented twice per week over 8 to 9 weeks. Within enrichment classes, it was completed over 6 weeks with two to three lessons taught each week. The Second Step intervention protocol was followed for the lessons. Implementation fidelity based on observers' ratings was high (Sullivan et al., in press). For each lesson, in-class activities or role plays were included to provide chances for modeling, rehearsal, and feedback. Interventionists were trained in best teaching practices (e.g., creating opportunities to respond, providing support as needed during role plays and group activities, and specific praise and supporting positive behavior).
2.4. Interview procedure For the 42 students in the urban school classrooms where the intervention was completed in December, interviews were conducted 7 to 17 weeks later in the same school year (M = 11.4, SD = 3.11). All other students were interviewed the following school year, between 28 and 49 weeks after the intervention was completed (M = 38.0, SD = 5.46). Interviews were conducted at the schools, usually in a corner of the media center or in an unoccupied classroom. The current study is based on segments of the interview that focused on participants' experiences with the eight specific skills that were the primary focus of the intervention (e.g., empathy, problem-solving, anger management). For each skill, interviewers showed participants a handout from the curriculum and asked a series of questions about students' experiences attempting to use that skill and the factors that did or would influence its use. Questions and follow-up probes for each skill differed depending on whether or not the student said they had used that skill since the intervention. Students who reported using or attempting to use a specific skill were asked to identify supports, or factors that made it easy for them to use the skill, and barriers, or factors that made it hard for them to use the skill. Interviewers followed up with questions about participants' other experiences using the skill, the types of situations in which they would not use it and why, and the reasons they may or may not use it in the future. Students who reported that they did not use a particular skill were asked to explain why they had not attempted to use the skill and the factors that would make them more and less likely to use it. Interviewers were trained using group meetings and individual supervision. Training addressed basic interviewing skills such as the use of open-ended questions and neutral responses to comments, and included modeling of the interview procedures. Trainers provided interviewers with written feedback after reviewing practice interviews that were recorded between the first and second training days. Trainers discussed common issues and provided further practice during the second training session. They also gave interviewers written and verbal feedback based on recordings of at least two additional practice interviews. Interviewers practiced with their trainers and were not allowed to conduct interviews with participants until their trainers considered them proficient. Trainers provided interviewers with written feedback on their first three interviews with research participants, and provided additional training if needed. Except for an undergraduate senior in the university's honors program, all of the 13 interviewers had bachelor's or master's degrees. African American women interviewed over half (56%) the participants; European American women interviewed 26% of participants; European American men interviewed 13% of participants; and a Latino American woman interviewed 3% of participants. Female interviewers interviewed boys and girls; male interviewers only interviewed boys. Three of the 141 interviews could not be transcribed due to faulty recording equipment. A researcher reviewed a random sample of 36% of transcripts from the spring of 2009 and all transcripts from the 2009–2010 school year to ensure accurate transcription.
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
183
2.5. Coding of interview transcripts Qualitative analysis of the interview transcripts was conducted by a team of eight researchers to identify themes within the following domains: barriers, supports, and relevance for each intervention skill. Transcripts were coded using NVivo 7 software using an iterative process in which themes were generated and then redefined based on multiple passes through the transcripts. Coders used NVivo to view transcripts and to mark specific segments of the transcripts that corresponded to the emerging themes. NVivo was also used to summarize the data (e.g., number of participants who identified each theme by segment), and to refine the coding by allowing the investigators to retrieve and review segments of the interviews that were assigned the same code. The first step involved open coding wherein two investigators proposed an initial set of themes and definitions based on their independent review of ten transcripts from the first wave of interviews. Themes were then refined based on consensus coding of all 42 Wave 1 transcripts by the same two investigators. During a third wave of coding, three other investigators reviewed the initial themes to determine their consistency with the data and to establish whether additional themes were needed within each category. A preliminary coding manual was then created with definitions and examples based on coding, discussion, and triangulation with existing literature. During a fourth wave of coding, three investigators coded all 138 transcripts based on the preliminary coding scheme and definitions. Themes were grouped into subsets and a pair of coders was assigned to each subset. As part of their training, coders independently coded ten randomly selected transcripts and then met to discuss and reach consensus on any instances where there was not initial agreement. In some cases, this led to further revision and clarification of definitions and incorporation of additional examples into the coding manual. Codes for three transcripts were compared midway through coding to address potential coder drift for all coding categories. Inter-coder reliability was assessed on a random sample of 28 interviews (i.e., 20% of the transcripts). Agreement for all but four of the 26 themes was above 80%, ranging from 71% to 99%. However, kappa coefficients, which took into account the low base rate of most of these codes, were below standards (i.e., .40) recommended by Fleiss (1981) for the majority of themes representing barriers
Fig. 1. Structure of themes within categories representing barriers and supports to the use of skills taught in the Second Step Violence Prevention Curriculum. aIndicates a minor theme mentioned by less than 25% of participants.
184
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
and supports. The coding manual was therefore revised based on an analysis of discrepancies, and a new pair of coders independently coded all transcripts using the revised definitions. Inter-coder reliability based on the revised definitions improved, with percentage agreement ranging from 79% to 99% (average = 91%) and kappa averaging .49. Coding was finalized by having the two coders meet with a third coder to review all cases where their codes disagreed to assign a final code based on a consensus. Themes representing the relevance of skills to participants' daily lives were initially coded into fairly broad categories (e.g., relevance supporting use of skills). Although inter-coder reliability for these categories was generally high, the content within these categories was fairly broad and heterogeneous. These themes were therefore reviewed to identify themes representing more narrowly-defined homogenous factors. The themes that emerged from this process were then defined and used to recode responses. Inter-coder agreement for the revised themes was quite high, with percentage agreement above 95% for all four codes, and all but one of the kappas above .70 (the outlier was .49). As with barriers and supports themes, any discrepancies were reviewed with a third coder, and final codes were determined by consensus. 3. Results Themes representing barriers to the use of intervention skills that emerged from the qualitative analysis were grouped into five categories (see Fig. 1): Beliefs and values not supporting use, concerns about relevance and effectiveness, behavioral enactment issues, other internal barriers, and contextual barriers. Themes representing supports for use of skills were also grouped into five categories: Beliefs and values supporting use, perceived relevance and effectiveness, behavioral competence, other internal supports, and contextual supports. The total number and percentage of participants who mentioned each theme, as well as the number of participants who mentioned each theme in reference to each of the eight intervention skills are reported in Table 1 for barriers and Table 2 for supports. Themes mentioned in reference to one or more of the skills by less than 25% of the participants were considered minor themes. The following sections describe the themes included within each of these categories. 3.1. Beliefs and values not supporting use Participants identified beliefs and values that would make them less likely to use the skills taught in the intervention. Themes included Normative Beliefs About Aggression and Beliefs and Values Not Supportive of Prosocial Behavior. 3.1.1. Normative beliefs about aggression Nearly two-thirds of participants described normative beliefs supporting aggression that discourage the use of one or more of the intervention skills. These included beliefs that aggression would lead to better outcomes and beliefs that others' provocative behavior warranted or deserved an aggressive response. This was most frequently mentioned in reference to anger management. One girl from an urban school said that it would be hard for someone to use anger management “when somebody just keep messing with the
Table 1 Number of participants who identified themes reflecting barriers for use of the intervention skills by skill and overall. DIFF
TOTALb
9 5
27 11
89 (64%) 58 (42%)
4 5 20 8
13 2 18 11
15 6 35 14
53 (38%) 45 (33%) 97 (70%) 55 (40%)
5 8 7 1 0
3 4 3 2 1
5 1 6 1 0
2 9 3 3 0
44 (32%) 46 (33%) 24 (17%) 23 (17%) 53 (38%)
5 7 7 69 38
14 9 9 47 23
12 4 2 95 31
15 11 10 45 52
64 (46%) 51 (37%) 40 (29%) 124 (90%) 101 (73%)
Theme
ANGR
EMP
PSOL
COMP
BULL
PEER
Beliefs and values not supporting use Normative beliefs about aggression Beliefs and values not supportive of prosocial behavior
52 6
8 24
13 12
9 23
32 7
16 12
Concerns about relevance and effectiveness Personal relevance not supporting use Having their own way of handling things Perceived ineffectiveness of skill Uncertainty of consequences or situation
20 8 25 9
1 4 21 15
2 12 33 10
5 11 36 16
16 9 34 12
Behavioral enactment issues Lack of self-efficacy for use of skill Too angry Too scared or nervousa Not remembering to use skilla Other internal barriers
10 30 3 4 2
12 1 1 9 6
10 8 3 8 6
11 3 9 1 3
Contextual barriers Closeness or history with person not supporting use Location and environment not supporting use Peers not supporting use Situation relevance not supporting use Other contextual factors not supporting use
8 7 14 8 23
15 6 1 16 51
16 10 10 23 31
11 15 5 19 32
GANG
Note. N = 138. Themes mentioned by more than 10% of the sample in reference to a specific skill are in boldface. ANGR = Anger management, EMP = Empathy, PSOL = Problem solving. COMP = Making a complaint, BULL = dealing with bullying, PEER = dealing with peer pressure, GANG = Dealing with gang pressure, DIFF = Diffusing a fight, Total = number of participants mentioning that theme for one or more skill. a Minor theme mentioned by less than 25% of participants. b Number of participants (% of participants).
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
185
Table 2 Number of participants who identified themes reflecting supports for use of the intervention skills by skill and overall. Theme
ANGR
EMP
PSOL
COMP
BULL
PEER
GANG
DIFF
TOTALb
Beliefs and values supporting use Negative beliefs and values about aggression and antisocial behavior Beliefs and values supporting prosocial behavior
18 32
7 41
10 28
5 18
8 19
28 24
26 13
24 23
72 (52%) 101 (73%)
Perceived relevance and effectiveness Perceived effectiveness of skill Negative outcome expectancies for aggression or antisocial behavior Personal relevance supporting usea
64 44 12
52 1 6
55 10 2
53 10 3
54 10 0
34 39 1
14 27 0
38 28 1
116 (84%) 89 (64%) 25 (18%)
Behavioral competence Remembering to use a skill Self-efficacy for use of skill Having experience and appraisal of situation Other internal supports
22 20 5 17
8 13 25 5
14 15 7 10
5 19 12 15
5 14 4 8
12 16 5 9
3 9 2 5
5 19 13 9
47 (34%) 70 (51%) 55 (40%) 57 (41%)
Contextual supports Closeness or history with person supporting use Location and environment supporting use Situation relevance supporting use Other contextual factors supporting use Peers supporting usea
6 9 22 20 6
30 6 19 30 1
24 7 18 21 1
19 5 15 20 2
3 6 19 20 4
11 6 16 30 8
2 1 6 17 3
29 8 23 22 7
80 (58%) 35 (25%) 73 (53%) 86 (62%) 27 (20%)
Note. N = 138. Themes mentioned by more than 10% of the sample in reference to a specific skill are in boldface. ANGR = Anger management, EMP = Empathy, PSOL = Problem solving. COMP = Making a complaint, BULL = dealing with bullying, PEER = dealing with peer pressure, GANG = Dealing with gang pressure, DIFF = Diffusing a fight, Total = number of participants mentioning that theme for one or more skill. a Minor theme mentioned by less than 25% of participants. b Number of participants (% of participants).
person.” Participants often stated that fighting would be justified in self-defense. This was reflected in the explanation of how a boy from an urban school said he would deal with bullying: “when they ready to fight you, they right up in your face, and you gotta defend yourself…if you don't wanna get hurt.” Other participants stated that certain behaviors by others warranted an aggressive response. As one girl from the county school noted, “Family insults are what hurt the most. And I…definitely don't think I'd be able to use this anger management because family's very personal.” Some participants mentioned beliefs reflecting parental values supporting aggressive responses in certain situations. One boy from an urban school explained, “My momma says if they try to hurt me or anything, she allows me to use self-defense.” 3.1.2. Beliefs and values not supportive of prosocial behavior Many participants (40%) stated they would not use a skill because it was not consistent with their beliefs or values. This was most often mentioned in reference to empathy and making a complaint. Some participants stated that they would not help someone they thought was getting what they deserved. For example, a boy in an urban school explained that he would not help another student being bullied if the student had bullied others in the past, “because you shouldn't be bullying other kids in the first place.” Others indicated that using a skill would make them feel weak. A boy from an urban school explained he would not seek help from a teacher if he was being bullied because I'm not gon' be a lil whimp and go tell on the teacher. …if I keep doin', like, snitchin', I'll feel like I'm in preschool or in the first grade and I'm a lil boy that's scared and hides in the corner for the rest of his life. Another boy from an urban school explained that it was difficult to resist peer pressure to steal jewelry for his girlfriend because “she was my girl so…I should be able to get her what she wants and stuff.” Some participants indicated that using a particular skill was inconsistent with their sense of self. One girl from an urban school explained that she tended not to use empathy “'cause I don't like getting in people's business unless they actually let me know what's going on.” 3.2. Beliefs and values supporting use Participants also identified beliefs and values that supported their use of intervention skills. Themes included Negative Beliefs and Values About Aggression and Antisocial Behavior, and Beliefs and Values Supporting Prosocial Behavior. 3.2.1. Negative beliefs and values about aggression and antisocial behavior More than half the participants (52%) described beliefs or values related to the use of aggression or antisocial behavior that would support their use of a skill. This included participants' beliefs that aggression and other delinquent behaviors were inconsistent with their goals, values, or sense of self, and that aggression was morally wrong. This was most frequently mentioned as a factor that would support skills for dealing with peer pressure, resisting gang pressure, and diffusing a fight. For example, a boy from an urban school said he was able to resist peer pressure to drink alcohol because “it was a bad thing to do.” Similarly, a girl at an urban school explained
186
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
that she was able to resist peer pressure to smoke because “my momma don't want nobody smoking in her house. And she don't like it that my brother smokes, so I'm not going to smoke.” A boy at the county school explained that he was able to use skills for diffusing a fight “because I'm pretty reasonable and don't think fighting is the best choice.” 3.2.2. Beliefs and values supporting prosocial behavior Nearly three-quarters of participants identified prosocial beliefs or values that would make it easier for them to use a skill. This was most often mentioned for empathy. This theme included general statements about wanting to use a skill and more specific statements indicating that use of a skill was consistent with a participant's sense of self, knowledge of right and wrong, parental messages, or life goals. A girl from the county school reported that she planned to use what she learned about dealing with bullying to help someone who was being bullied because “I would hope someone would do that if I was getting bullied.” Others stated that using a skill was consistent with their personalities. A girl from an urban school explained that it was easy for her to use empathy because “when I see people so sad or mad or something…my reaction is just to ask them what's wrong and see, tell them something that they could do to make them feel better.” A boy from an urban school explained that it was easy for him to use skills for dealing with gang pressure because “I'm going to college and I'm gon' make somethin' of my life.” 3.3. Concerns about relevance and effectiveness Four themes represented participants' concerns about the relevance or relative effectiveness of the skills addressed in the intervention: Personal Relevance Not Supporting Use, Having Their Own Way of Handling Things, Perceived Ineffectiveness of Skill, and Uncertainty of Consequences or Situation. 3.3.1. Personal relevance not supporting use Over one-third (38%) of participants indicated that they did not or would not use a skill because they did not find it personally relevant. This was most often mentioned in reference to anger management, dealing with bullying, resisting gang pressure, and diffusing a fight. A boy from an urban school explained that he used anger management the least “because you have to do something really, really bad to get on my nerves.” A student in an urban school explained that she never had to use the skills for dealing with bullying because “everybody was my friend, and I was a likeable person.” 3.3.2. Having their own way of handling things Participants also indicated that some skills were not relevant to them because they had their own way of handling things, such as avoiding problem situations or seeking adult intervention. This was mentioned by one-third of participants for at least one skill. A girl from the county school indicated that instead of using problem-solving with a friend, she would “rather stop bein' their friend if there's gonna be a problem like that.” A boy from an urban school said he would not attempt to diffuse a fight because “when fights start I usually I just stay away from them.” 3.3.3. Perceived ineffectiveness of skill The majority of participants (70%) indicated that they might not use a skill because they felt it would be ineffective. This included instances where they thought the skill would be either ineffective for dealing with a problem or it would result in negative outcomes. This was mentioned by at least 13% of the sample for each of the skills and by over 20% of the sample for half of the skills. One girl from an urban school explained that she thought strategies for dealing with bullying would not work “because if I'm staying calm and…taking breaths or something . . . then the bully will probably still keep saying stuff to me.” Another girl from an urban school said she would not attempt to diffuse a fight in situations involving “somebody who is tougher than [another student] 'cause they will… get me, like, somehow. They're crafty like that.” Some participants stated that they did not believe a skill would work in the future based on prior experiences. One girl from an urban school said that she was not likely to use making a complaint in the future “'cause when I have made a complaint it didn't work very well.” 3.3.4. Uncertainty of consequences or situation Many participants (40%) reported that they would be less likely to use a skill because of uncertainty about the nature of the situation, the likely outcomes, or other people's reactions to their use of the skill. One girl from an urban school explained that it would be hard for her to make a complaint because the other person “might get an attitude or…it might not go the way you wanted…And I was kind of saying, ‘Is this going to turn out right or is this going to start something else?’” A boy from the county school explained that it was hard for him to diffuse a fight because he could not predict how the other student would react: “I didn't really know him…I didn't know what he was like or what he does.” 3.4. Perceived relevance and effectiveness Participants identified three themes related to relevance and effectiveness that supported the use of intervention skills. These themes included Perceived Effectiveness of Skill, Negative Outcome Expectancies for Aggression or Antisocial Behavior, and Personal Relevance Supporting Use.
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
187
3.4.1. Perceived effectiveness of skill The most frequently and consistently mentioned support for use of a skill was the belief that it would result in positive outcomes or achieve participants' goals in the situation. At least 25% of participants mentioned this as a factor for every skill except dealing with gang pressure. Many participants made general statements that a skill worked for them in the past. In some instances, they indicated that using a skill would help them avoid a negative outcome. For example, a boy from an urban school said that he was able to use problem-solving skills because “if I use them, then I won't be pretty tense or mad or get into a fight.” Others described specific positive outcomes from using a skill. One girl from an urban school explained that she planned to use empathy in the future because “it helped me out to understand what people was going through, and…if I help somebody when they going through something, when I go through something they can actually help me.” 3.4.2. Negative outcome expectancies for aggression or antisocial behavior Nearly two-thirds of the participants indicated that they would be more likely to use a skill because they anticipated that alternatives such as aggressive or antisocial responses would lead to more negative outcomes. This came up often in participants' discussions of factors that would make them more likely to use anger management, skills for dealing with peer and gang pressure, and skills for diffusing a fight. A boy from an urban school explained that he would use anger management in the future “'cause I don't want to get in no more fights.” A girl from an urban school explained that she was able to use anger management “because I didn't want to get in trouble or get suspended or anything. Because I know how my mouth is. I'll, like, catch an attitude, then get loud, then get in the person's face, and then get to fighting.” A girl from the county school explained that she planned to use strategies for resisting peer pressure to use substances because “I don't want to die young. I'd rather die old, old.” 3.4.3. Personal relevance supporting use Personal Relevance Supporting Use was a minor theme mentioned by 18% of participants. This was most frequently mentioned in reference to anger management. One boy from an urban school stated that he used anger management frequently “because I get mad quick, and I don't have patience.” A girl in the county school reported that she used the skills for diffusing a fight most frequently “because I tend to get in quite a few arguments with people.” 3.5. Behavioral enactment issues Participants identified factors that would make it difficult for them to enact intervention skills. Themes included Lack of SelfEfficacy for Use of Skill, Too Angry, Too Scared or Nervous, and Not Remembering to Use Skill. 3.5.1. Lack of self-efficacy for use of skill Some participants (32%) indicated that they were less likely to use an intervention skill because they did not feel capable of using it successfully. This included general statements that using a skill was “hard.” A girl from an urban school explained that resisting peer pressure was difficult because “It was just something different for me 'cause that was like my first time using peer pressure.” A girl from the county school said that using problem-solving was difficult “'cause sometimes it's hard to listen to the other person. 'Cause all you wanna do is—you think you're right all the time.” 3.5.2. Too angry One-third of the participants described being unable to use a skill because they were too angry. Their anger sometimes made it difficult to remember the skill, or it motivated them to respond aggressively. Not surprisingly, participants most frequently mentioned this when discussing anger management. A boy in an urban school explained that he would not use anger management “when I'm really angry…and I'm about to fight.” Another boy from an urban school reported that he would not use making a complaint, saying, “When you be mad, you don't be thinking about the skills.” Similarly, a girl from the county school explained, “When you deal with bullying, you can't always stay calm, you know. You bottle up your emotions for so long, and somebody's bullying you, that one day you're just gonna snap.” 3.5.3. Too scared or nervous A minor themed mentioned by 17% of participants was being too scared or nervous to use a skill. This included being scared of the consequences of using the skill, worried they would not use the skill correctly, or nervous because of the circumstances. One girl from an urban school explained that she was scared to try using the skills for dealing with bullying with a particular student, “since she was a bully and she scared me a lot…I was like, ‘Oh my god, she's going to kill me.’” A girl from the county school explained that it was difficult to use the skills for making a complaint because “I was scared that I wouldn't be able to use the skills in the way I would've liked…And I would've like whined or…like been a little baby about it.” 3.5.4. Not remembering to use skill A second minor theme that was mentioned by 17% of participants was Not Remembering to Use a Skill. In most cases this was in reference to empathy or problem solving. A girl from a county school explained that it was hard to remember to use empathy because “you don't think about things like that.” Similarly, a boy said that he did not always remember to use problem-solving: “If I ain't think about it, I ain't going to do it, see. If it popped to my mind, I'll do it.”
188
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
3.6. Behavioral competence Participants identified three factors that would support their ability to use a skill when facing a relevant situation. Themes included Remembering to Use a Skill, Self-Efficacy for Use of Skill, and Having Experience and Appraisal of Situation would facilitate their use of the skill. 3.6.1. Remembering to use a skill Over one-third of participants indicated that they were more likely to use a skill because it came to mind when they encountered a relevant situation. They mentioned this most often when discussing anger management (16%). A boy from the county school said he was confident about his ability to use anger management because “it's simple, easy to remember, straightforward.” Some participants indicated that it was easier to use a skill when they stopped to think. A boy from an urban school explained, “I didn't feel like getting in any trouble, so I just thought to myself like what should I do and stuff.” 3.6.2. Self-efficacy for use of skill More than half the participants noted that self-efficacy would make them more likely to use a skill. This was mentioned for nearly all of the skills. For example, one girl from an urban school explained that it was easy for her to use anger management because “if I want to do something, and I put my mind to it, I can do it.” A boy from the county school explained that what made it easy for him to use problem-solving skills was that he used them frequently: “Basically me doing it all the time in most of my problems.” Others explained that they were not afraid to use the skill. A boy from the county school reported that he was able to use strategies for making a complaint because “I was brave enough to go up to her.” 3.6.3. Having experience and appraisal of situation Participants also noted that prior knowledge about the situation or the other person involved, and previous experience in similar situations would make it easier for them to use a skill. Participants mentioned this most frequently when discussing empathy. For example, a girl from an urban school explained that she was able to use empathy to stop gossiping about another student: “Because someone had talked about me, and I didn't like it, and I thought of the girl that I had talked about, and she didn't like it either, so I have stopped.” In other cases, participants reported that understanding the others in the conflict made it easier to use a skill. A boy from an urban school explained that his knowledge of his friends gave him confidence in his ability to use humor to diffuse a fight: “I knew both of them was really goofy… so I just made 'em start laughin', and then they just forgot about all of it.” Some participants indicated that their general knowledge about the situation made it easier to use a skill. A girl from the county school reported that it was easy to resist peer pressure to smoke “because it was a cigarette and [I] knew, like, what was in it, and I knew like what would happen if I did try it, so I just said no.” 3.7. Other internal barriers Participants described a variety of other internal barriers to using intervention skills that were not captured by the other themes. These included having automatic aggressive responses, struggling with competing goals, and experiencing transient influences such as being in a “bad mood.” A girl from the county school said she would have difficulty using the skills for problem solving “if I was in a really, really, really, really, really bad mood.” A girl from an urban school reported that she would not use strategies for diffusing a fight if the dispute was over something significant: “If it's a little problem, I'm gonna agree with them…but if it's like so serious, we just gonna fight.” A girl from an urban school described a situation in which it would be difficult to resist peer pressure, saying, “If I was actually interested in doing something that they were pressuring me to do, that would only be the one thing that would make it hard.” 3.8. Other internal supports Participants also described a variety of other internal factors that would make it easier for them to use intervention skills. Many were person-specific and ranged from having low blood sugar that day, which made it difficult to fight, to being shy. Some responses included wanting to try something new or feeling that the situation was not important enough to warrant starting a fight. A boy from the county school reported that he was able to use anger management “because I wanted to try something new and see if it worked.” A girl from an urban school explained that she was confident about her ability to ignore other students' provocation because “I'm not gonna get all worked up over what somebody said to me.” 3.9. Contextual barriers Participants described contextual or situational factors that represented barriers to using intervention skills. These themes were well represented in the data, with between 40 and 101 participants mentioning each theme. Contextual factors were identified for all eight skills, though participants varied in how frequently they mentioned specific factors across skills. 3.9.1. Closeness or history with person not supporting use Nearly half the participants indicated that the extent to which they were close to or had a history with the other person involved in a situation would make them less likely to use a skill. Some participants reported that a lack of relationship or a poor relationship with
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
189
the other person would make them unlikely to use a skill. For example, a girl from an urban school described a person to whom she would not try making a complaint: …on a person that I don't like, because they probably you know, make it even worse. They would try to, you know, say stuff about me or try to make people laugh at me…it'd be a really big mess. Conversely, having a close relationship could also make it difficult for participants who reported wanting to avoid a conflict with friends. A boy from an urban school said it was hard to resist peer pressure because “saying no to your friends is kinda hard.” Others indicated that having no prior history with someone would make it difficult to use some skills. For example, one boy from an urban school reported that it was hard to use anger management and apologize because “I wouldn't know how she was, wasn't going to react. I ain't really know her.” 3.9.2. Location and environment not supporting use Participants described situations in which the location or environment in which the conflict occurred would make it difficult to use a skill. This was most often mentioned for making a complaint. For example, some participants said it was difficult to use skills where it was noisy or crowded. A boy from an urban school indicated that it was hard to use making a complaint “when somebody talking over one of us.” Participants also stated that the presence and behavior of bystanders could make it difficult to use a skill. A boy from an urban school described a time when it was difficult for him to use problem-solving: One time when I got into a fight it was hard to use that skill because, like, a whole bunch of people started gathering around. And then I tried to walk away but then somebody pushed me into the person. Some participants noted that there were places where they would not use a skill. A boy from an urban school said, “At school I use [anger management] because I ain't trying to get in trouble, and I'll get in trouble when I get home. But on the streets, I wouldn't do it.” 3.9.3. Peers not supporting use Participants reported that they were less likely to use a skill if their peers did not support its use. This was most often mentioned when discussing anger management. Some participants expressed concerns that using a skill would damage their image or reputation or could isolate them from their peers in some way. A girl from an urban school explained that it was hard for her to use anger management because “some people thought I was a punk.” A girl from the county school reported that it was hard to use strategies for dealing with bullying to defend a boy in her class because of how it might change others' perception of her: In the land of middle school, if you even talk to a guy…everybody thinks you're dating… I don't wanna say nobody really liked him, but everybody…thought he was weird…So that was another thing with me talking about or hanging out or talking to a, you know, mentally disabled person. Similarly, a boy from an urban school explained that it was hard to use skills for diffusing a fight because it might draw unwanted attention from peers: “All the students bein' around like cheerin' people on to fight, and then me, the only one that step out to be the oddball and stop the fight.” 3.9.4. Situation relevance not supporting use Nearly all participants (90%) indicated that they did not use a skill because a relevant problem situation did not occur. This was most frequently mentioned for skills that focused on dealing with specific situations such as bullying, peer pressure, gang pressure, and diffusing a fight. For example, a boy from an urban school stated that he had “never seen an actual gang.” Similarly, a girl from the county school reported that she had not used skills for dealing with bullying “because people don't want to fight you that often.” 3.9.5. Other contextual factors not supporting use Participants identified a variety of other contextual factors that were not captured by the other specific themes. These included the age or size of the other person, the number of others involved, and the potentially dangerous nature of the situation, such as when weapons were present. A boy from an urban school said that he would not use anger management if he thought he could win a fight: “if they my size, and I know I can beat them, I just fight them.” Another boy from an urban school said that it was difficult to use skills for dealing with bullying “when you have a lot of people, and they're like trying to, like, trap you, keep you there, and stuff.” A girl from a county school reported that she would not use skills for dealing with bullying “if the person has a gun or knife or anything to hurt you.” 3.10. Contextual supports Participants described five contextual or situational factors that made it easier or more likely for them to use an intervention skill. These factors were the converse of the factors identified as barriers to using skills and were generally well represented across participants, with 27 to 86 participants mentioning each factor as influencing their use of at least one skill.
190
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
3.10.1. Closeness or history with person supporting Use Over half the participants described situations in which their prior relationship or history with the other person involved made it easier or more likely for them to use a skill. This was mentioned most often for empathy, problem solving, making a complaint, and diffusing a fight. Participants reported that feeling close to the other person supported using a skill in some situations, whereas having no relationship or prior history supported using the skill in others. A girl from an urban school reported that she would use the skills for diffusing a fight “if one of my friends get in a fight, and they was friends, I wouldn't let them fight.” In contrast, a boy from an urban school explained, “If I got into a fight with a person that I don't know, I'll use [empathy]. The clues like is he getting mad, look at his facial expressions, and stuff… his body language and stuff.” 3.10.2. Location and environment supporting use One-quarter of participants discussed how the location or other environmental factors could make it easier to use a skill. This factor was spread across all eight skills, with 7% or less of participants identifying it as a factor for any one specific skill. Some participants noted that having a quiet or isolated location to go to made it easier to calm down. For example, a boy from an urban school explained that it was easy to use anger management because “I just…went in a quiet place.” A student in an urban school stated that it was easy for him to use anger management because “we was in the gym so it was like a lot of space, so then I walked outside and just cooled off, walked around and stuff.” In other cases, they indicated that the presence of others made it easier to use a skill. A girl from an urban school explained that it was easier for her to resist peer pressure to fight “'cause [a counselor] was right there.” A boy in the county school stated that it was not hard for him to diffuse a fight because “I knew a principal was going to be there soon, so I just stepped in their way for a couple seconds and then it just worked out.” 3.10.3. Situation relevance supporting use More than half the participants (53%) identified a skill they had used or were likely to use because they encountered relevant situations. This was mentioned fairly consistently across all of the skills except for dealing with gang pressure. For example, a girl from the county school reported that she was likely to use diffusing a fight “'cause people try to fight people for no reason all the time.” A boy from an urban school indicated that he would use dealing with bullying in the future “because there are always going to be somebody in your life that you not gonna like.” 3.10.4. Other contextual factors supporting use Nearly two-thirds of participants described other contextual factors that would support the use of a skill, but which were not captured by the other themes. These included descriptions of situations in which they anticipated using a skill in the future, such as in college or at work. Some participants explained that they would be more likely to use a skill in a potentially dangerous situation. For example, one boy from an urban school reported that he would use strategies to deal with gang pressure “if it is a big person, and, like, if they got a weapon or something. And then you will have to use the skills 'cause you can end up getting hurt.” A boy from the county school explained that he generally would “go with the flow” but that he would use the steps involved in resisting peer pressure if his peers were “gonna be like destroy school property or go on to like try to steal something, then I'll probably have to walk away or say ‘no’ or try to list the consequences to 'em.” 3.10.5. Peers supporting use Peers Supporting Use was a minor theme mentioned by 20% of participants. Examples included friends explicitly encouraging the use of a skill and participants believing their friends would think the skill was useful or effective. One boy from the county school explained that it was easier for him to deal with peer pressure because “I thought if some people think that these will work, and I thought I'd try using it, and it really works.” 4. Discussion The overall goal of this study was to identify factors that influence adolescents' use of the social and emotional skills taught in the Second Step curriculum in their everyday lives. Although this study focused on Second Step, the factors that emerged are likely to be relevant to other SEL interventions with a similar focus. Qualitative analysis of interviews conducted with adolescents 3 to 12 months after they participated in the sixth grade curriculum identified a variety of factors that either supported or discouraged their use of the specific skills targeted by the curriculum. The vast majority of participants were able to identify both barriers and supports for using the skills. Factors identified as barriers to using a skill were often the converse of those identified as supports. At a general level, the interviews highlighted some of the challenges and successes students experienced in their efforts to use the skills targeted by the intervention. This emphasizes the complex processes surrounding adolescents' decisions to make use of what is addressed by the curriculum. Participants identified a variety of individual-level factors that would influence their likelihood of using specific social and emotional skills targeted by the intervention. These included their beliefs and values, their perceptions of the relevance and effectiveness of a skill versus alternative responses (e.g., aggression), and their capacity to use the skill (e.g., self-efficacy, emotional state). These factors are all well represented within social information-processing theories that depict aggressive behavior as a function of deficits in information processing skills and a database of beliefs, values, goals, and schemas (Crick & Dodge, 1994; Huesmann, 1998). The individual-level factors identified in this study are also consistent with previous qualitative studies conducted with other samples of adolescents. Farrell et al. (2008) interviewed sixth and seventh graders attending urban public schools to identify factors that
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
191
would influence their use of specific violent and nonviolent responses in problem situations with peers. The themes that emerged included personal resources, problem-solving skills, emotion regulation skills, and self-efficacy for enacting nonviolent responses. Many of their participants expressed the belief that nonviolent responses similar to those targeted by social-emotional interventions (e.g., talking out the problem, telling an adult, ignoring a provocation) would be ineffective in difficult situations. Other individuallevel factors identified by Farrell et al. (2008) included appraisals and attributions, beliefs about the acceptability and necessity of fighting, and anticipated outcomes of aggressive and nonviolent responses. Similarly, Quinn, Bell-Ellison, Loomis, and Tucci (2007) conducted a qualitative study of middle school youth in summer programs, and found that more than half the participants reported the belief that fighting was sometimes necessary, usually to defend oneself or in response to insults about family members. Participants in this study also identified environmental or contextual factors that would influence their use of intervention skills. These included situation-specific factors, such as their relationship with the other person in the conflict, and environmental factors, such as the location in which the situation occurs, that would influence their interpretation of the situation and how best to respond to it. These factors are also well represented within the social information-processing model and in other studies that have examined the influence of contextual factors on behavior. Huesmann and Guerra (1997) noted that an adolescent's context will influence the social cognitions that are acquired, maintained, and recalled from the database during a particular problem situation. Contextual factors may also be embedded in school environments and may create standards for what is considered an appropriate response to provocation, above and beyond individual students' beliefs (Felson, Liska, South, & McNulty, 1994). Participants identified peers as an important contextual factor that could either encourage or discourage the use of an intervention skill. The strong influence of peers was reflected in a qualitative study by Farrell et al. (2010) that identified a variety of ways peers can influence whether an adolescent will choose a prosocial or an aggressive response in a given situation. These included direct influences, whereby peers and bystanders actively encourage a prosocial or aggressive response to a specific situation while it is occurring, and more general forms of peer influence, such as concern about how a given response might influence one's reputation or social standing with peers. Quinn et al. (2007) also identified contextual factors in their qualitative study of adolescents in summer programs. In particular, some adolescents discussing the potential outcomes of walking away from a fight expressed the belief that fighting would be inevitable or that peers would harass the person who walked away. 4.1. Implications for research and practice The results of this study provide further support for focusing universal school-based programs on helping adolescents develop social and cognitive skills to reduce their use of aggression. Ironically, many of the individual-level factors participants identified as influencing their use of intervention skills (e.g., beliefs and values, self-efficacy, emotional state) are often the targets of interventions (e.g., altering beliefs and values, increasing self-efficacy, emotion regulation; Boxer & Dubow, 2002). Although the current study provides further support for focusing interventions on these factors, it also suggests that more intensive efforts are needed to ensure that the skills targeted by SEL interventions are used, mastered, generalized across different situations, and maintained over time (Ladd & Mize, 1983). Participants in this study described how their beliefs and values influenced their likelihood of using skills taught by the intervention versus alternative responses, typically those involving the use of aggression. This highlights the need for intervention strategies to directly address participants' underlying beliefs and values not only about the use of aggression, but about the use of intervention skills. For example, beliefs related to revenge and the perceived need to retaliate in response to a real or perceived challenge may trump other potential responses to a problem situation (Frey, Pearson, & Cohen, 2014). Interventions may need to provide opportunities for participants to discuss their goals and think through conflicting values. For example, during Second Step adolescents are encouraged to consider their family's reactions as part of problem solving. Adolescents who believe it is necessary to respond aggressively in reaction to an insult about their family may be encouraged to think about whether their family is harmed more by peers insulting them or by being suspended for fighting. Encouraging students to voice their beliefs and values must be pursued cautiously, however, given the strong influence of peer norms on behavior (e.g., Henry et al., 2000). Interventionists must ensure that the process successfully alters beliefs rather than simply exposing adolescents to peer norms that support aggression. Adolescents are not likely to use the skills targeted by SEL interventions unless they believe they will be more effective than the alternatives. The vast majority of participants in this study identified perceived effectiveness as a support for the use of one or more intervention skill. Conversely, many participants also indicated that they would not use a skill because they either thought it would not work or believed it would make things worse in a particular situation. Some noted that even if the skill solved the immediate problem, it might have other negative consequences, such as damage to their image or reputation. The effectiveness of skills might be increased by efforts to get teachers, school personnel and students to reinforce their use outside of intervention classrooms. For example, students in the Farrell et al. (2010) qualitative study indicated that they were unlikely to seek help from a teacher if they were having a problem with another student because teachers would not help them and peers would label them a “snitch.” This suggests the need for intervention efforts that address factors within the school and peer environment that increase the likelihood that intervention skills will be effective or that reduce the likelihood of negative repercussions for using skills. Such efforts could be informed by further research to identify factors that reduce the actual or perceived effectiveness of skills that are the focus of current SEL interventions (Farrell et al., 2008). This study also suggested the need for further efforts to ensure that students master and are able to use the skills targeted by SEL interventions. Participants indicated that although they were sometimes able to identify an appropriate response to a problem situation, they were sometimes unable to employ it because of strong emotions such as anger and fear, or because they simply did not think of using it. Emotion regulation is a key skill targeted by SEL interventions. Ironically, 22% of the participants in this study reported
192
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
not being able to use anger management because they were too angry. In other words, they were sometimes unable to use the skill when they most needed it. The finding that strong emotional responses limited the ability to use not only anger management, but other skills targeted by the intervention highlights the importance of ensuring that students adequately master this skill. Similarly, nearly one-third of participants identified low self-efficacy as a barrier to using one or more of the skills. Improving adolescents' confidence in their ability to successfully employ the skills targeted by interventions requires going beyond superficial coverage of skills by providing opportunities for practice, assessing students' mastery of skills, and providing remediation when needed. For example, Sutherland, Wright, and Sullivan (2008) suggested the use of curriculum-based measures to identify skills students have not yet mastered. Such measures might be used to set a threshold for the acquisition of specific social and emotional skills, with further efforts focused on ensuring that students meet that threshold. Interventions might also provide more opportunities for students to discuss challenges to using skills and how they might be addressed. The fact that most students were able to describe a successful use of at least one intervention skill suggests that students may be able to learn from each other. Several aspects of the Second Step intervention attempt to facilitate this. For example, as part of the intervention, students discuss possible responses to problem situations, write down gains and losses for angry reactions to situations, and engage in similar activities designed to help them adapt intervention skills to meet their goals in relevant situations. This is an excellent first step toward helping adolescents master the application of intervention strategies in difficult situations, but more intensive efforts may be needed. Environmental and contextual factors also influence students' abilities to use intervention skills. This suggests the need for an additional problem-solving lesson in which students' identify challenging situations based on these factors (e.g., specific environmental and peer contexts) and problem-solve how to deal with these situations. Adolescents may find it difficult to use new skills to address challenging situations and their initial efforts may not be met with success (Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011). Explicitly addressing this possibility may encourage students to persist in the use of social and emotional skills despite less than optimum initial outcomes. In addition to mastery of skills, a key goal of SEL interventions is generalization and maintenance (Ladd & Mize, 1983). There are several approaches SEL interventions might use to promote this goal. Including relevant examples of frequently encountered problem situations that place youth at-risk for interpersonal conflict could facilitate the generalization of problem-solving skills by providing students with opportunities to practice skills in a supportive context. For example, Sullivan et al. (in press) adapted materials in the Second Step curriculum to incorporate examples based on situations students identified in a previous qualitative study (Farrell et al., 2007). Homework assignments can also be used to facilitate the generalization and maintenance of learned skills, especially if parents become involved. Although homework assignments are included in Second Step, we could not find evaluations of Second Step for middle school students that examined homework completion as an aspect of implementation fidelity or assessed parental involvement in reviewing homework assignments with students. Additional factors that might increase maintenance include promoting awareness of skills at home and at school. For example, teachers and other school staff could provide students with feedback on their use of key Second Step skills in context including reinforcement for appropriate use of skills or discussion of alternative responses to better address peer-based conflict. This would require ensuring that teachers and staff are aware of the skills taught in Second Step and procedures for supporting them. Although this may occur informally, such strategies are not explicitly addressed in Second Step, and may be less likely when the intervention is implemented by a subset of teachers in selected classes. Other techniques could include placing posters in classrooms and hallways, and morning announcements that serve as reminders of skills, and sending letters home to parents suggesting ways they might reinforce the use of skills addressed by the intervention. Participants' responses suggested environmental changes that could reduce barriers and increase support for the use and maintenance of intervention skills. One common theme was the location and environment in which the problem situation occurred. This theme is consistent with prior research examining the connection between school violence and specific unsupervised places and times, such as cafeterias, locker rooms, and hallways during class transitions (Astor, Meyer, & Behre, 1999). Identifying times and places in which problems tend to occur and increasing adult monitoring may reduce the prevalence of problems. In addition, increased supervision and monitoring may provide the scaffolding needed for students to safely practice using the skills they have learned. Providing safe, quiet places where students can go to calm down may also increase the use of skills, as students reported that their overwhelming anger was sometimes a significant barrier to using what they learned in the intervention. 4.2. Limitations and future directions This study had several limitations that merit discussion. A key limitation concerns the manner in which the intervention was implemented in the Sullivan et al. (in press) study from which participants in this study were drawn. Sullivan et al. implemented Second Step within a randomly selected subset of sixth grade classrooms in the participating schools. Second Step includes specific curricula for sixth, seventh, and eighth grades. Adolescents may have had different responses to the intervention if they had participated across all three grade levels and if the intervention had been implemented on a school-wide basis. Moreover, the class schedules within the two urban schools necessitated implementing the intervention within a single semester rather than across the whole school year. Although this more intensive implementation schedule may have had advantages, it may also have limited the intervention's impact. Furthermore, although teachers were encouraged to be present and participate in the intervention, they were not actively involved in its implementation. The use of research staff rather than teachers to implement the intervention provided more control over implementation fidelity, but reduced teachers' level of involvement. The implementation of Second Step through health and physical education and elective classes was not unique to this study, but may have further limited opportunities to use teachers as a resource to support students' application and generalization of social and emotional skills. Other school-wide prevention efforts, such as the Olweus Bullying Prevention Program (Olweus & Limber, 2010) include training all school staff in the intervention components and
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
193
focusing on broader school-level climate change compared to individual-level skill building. An important direction for future research would be to determine the extent to which intervention effects might be enhanced by training all school staff on the key skills that were taught to students and how to help them generalize these skills in dealing with actual problem situations that arise during the school day. Although the qualitative approach used in this study provided an important perspective, it also had limitations. Participants were asked to provide retrospective reports of their experiences with the intervention skills. Consequently, their reports may have been less accurate due to the passage of time. This may be particularly true for those participants who were interviewed nearly a year after their exposure to the intervention. This approach also assumes that adolescents have clear insights into the factors that influence their behavior and are willing to share them. There may also have been interviewer effects, in that participants may have told interviewers what they believed the interviewers wanted to hear. Efforts were made to minimize these effects through a thorough interviewer training process. This appears to have been at least somewhat successful, in that most adolescents discussed some barriers to using the skills. Although the themes that emerged were generally well represented in the data, several (e.g., Not Remembering to Use Skill, Peers Supporting Use) were identified by a smaller number of participants, particularly when broken down by specific skill. Whereas these minor themes may not be representative of the experiences of the majority of participants in the intervention, they may indicate factors that limit the intervention's impact for subgroups of adolescents. Further work is needed to determine the relevance of these minor themes. Finally, although we made several efforts to enhance the quality and credibility of the data collected, including use of multiple investigators who participated in multiple iterations of coding and review, use of multiple coders with assessment of inter-coder reliability and use of consensus coding to resolve discrepancies, triangulation with previous studies, and incorporating quotes within our descriptions, we did not employ additional steps such as an independent audit or member checking (Creswell & Miller, 2000). Other limitations were related to sample characteristics. Participants were drawn from three middle schools: two in an urban area and one in an adjoining county. It is not clear how well the findings might generalize to schools in other settings. Moreover, although including schools representing two different contexts provided a broader sample, settings were not sufficiently sampled to allow cross-setting comparisons. Differences across schools in the two settings also may have influenced the extent to which students had opportunities to use the skills targeted by Second Step. Analyses of baseline data reported in the outcome study (Sullivan et al., in press) indicated that compared to students in the county school, those in the two urban schools reported higher rates of overt victimization (d = .24 to .34), and higher teacher ratings of overt (d = .48 to .65) and relational (both ds = .37) victimization. Within the current study, a higher percentage of students in the urban schools were excluded due to low attendance on days the curriculum was taught. Their low level of participation in the intervention did not provide a basis for seeking their input, which is unfortunate as they represent a group that may be in particular need of the skills targeted by the intervention. Finally, the current study did not attempt to identify the extent to which barriers and supports to the use of intervention skills may have differed across subgroups of adolescents (e.g., aggressive adolescents). Future research might employ a more purposive sampling strategy to determine the extent to which barriers and supports are related to factors that have been found to moderate the impact of school-based violence prevention programs (cf. Farrell et al., 2013). 4.3. Conclusions The substantial resources devoted to developing and implementing school-based violence prevention programs underscores the need to identify ways to increase their effectiveness. This effort requires a clear understanding of the individual and contextual factors that minimize and maintain aggressive behavior and that influence the use of effective nonviolent behaviors for a particular target population. It also requires the development of effective approaches to ensure that participants adequately master the strategies they are taught, use them appropriately, and find them sufficient to meet the challenges they encounter on a daily basis (Ferguson et al., 2007). Qualitative methods such as those used in this study and the related study by Farrell et al. (in press) can inform these efforts by improving the understanding of adolescents' experiences attempting to apply what they learn from interventions. This effort could also be informed by including measures that assess mastery and use of intervention skills in outcome studies to determine the extent to which participants acquire and use the skills targeted by the intervention, the impact on outcomes, and the extent to which these effects are sustained over time. Finally, the barriers and supports identified by participants in this study suggest that universal school-based curricula such as Second Step can play an important role in preventing violence. At the same time, they may be of limited benefit unless they are part of a more comprehensive prevention effort that addresses contextual factors that support or reduce the effectiveness of those skills (Farrell & Vulin-Reynolds, 2007). For example, improving adolescents' problem solving skills is unlikely to reduce their use of aggression if they continue to have peers who respond negatively to their use of nonviolent responses, teachers who do not provide consistent consequences for behaviors, or parents who endorse standing up for oneself by fighting (Farrell et al., 2008, 2010). This suggests the need not only for interventions that focus on individual students, but also for strategies that alter the peer culture and broader social environment to ensure that individual-level changes are supported (Farrell & Vulin-Reynolds, 2007; Gottfredson & Gottfredson, 2001). References Astor, R. A., Meyer, H. A., & Behre, W. J. (1999). Unowned places and times: Maps and interviews about violence in high schools. American Educational Research Journal, 36, 3–42. http://dx.doi.org/10.3102/00028312036001003. Bierman, K. L., Coie, J. D., Dodge, K. A., Greenberg, M. T., Lochman, J. E., McMahon, R. J., et al. (2010). The effects of a multiyear universal social–emotional learning program: The role of student and school characteristics. Journal of Consulting and Clinical Psychology, 78, 156–168. http://dx.doi.org/10.1037/a0018607.
194
A.D. Farrell et al. / Journal of School Psychology 53 (2015) 179–194
Boxer, P., & Dubow, E. F. (2002). A social-cognitive information-processing model for school-based aggression reduction and prevention programs: Issues for research and practice. Applied and Preventive Psychology, 10, 177–192. http://dx.doi.org/10.5993/AJHB.24.4.3 (CDC, 2012). Centers for Disease Control and Prevention (2007). The effectiveness of universal school-based programs for the prevention of violent and aggressive behavior: A report on recommendations of the Task Force on Community Preventive Services. Morbidity and Mortality Weekly Report, 56(RR-7), 1–11. Committee for Children (1997a). Second Step: A violence prevention curriculum; Elementary school. Seattle, WA: Committee for Children. Committee for Children (1997b). Second Step: A violence prevention curriculum; Middle school/junior high. Level 1 Foundation Lessons. Seattle, WA: Committee for Children. Cooke, M. B., Ford, J., Levine, J., Bourke, C., Newell, L., & Lapidus, G. (2007). The effects of city-wide implementation of “Second Step” on elementary school students' prosocial and aggressive behaviors. The Journal of Primary Prevention, 28(2), 93–115. http://dx.doi.org/10.1007/s10935-007-0080-1. Creswell, J. W., & Miller, D. L. (2000). Determining validity in qualitative inquiry. Theory Into Practice, 39(3), 124–131 (0040–5841/2000$1.50). Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social information-processing mechanisms in children's social adjustment. Psychological Bulletin, 115, 74–101. http://dx.doi.org/10.1037/0033-2909.115.1.74. Durlak, J. A., Weissberg, R. P., Dymnicki, A. B., Taylor, R. D., & Schellinger, K. B. (2011). The impact of enhancing students' social and emotional learning: A meta‐analysis of school‐based universal interventions. Child Development, 82, 405–432. http://dx.doi.org/10.1111/j.1467-8624.2010.01564.x. Edwards, D., Hunt, M. H., Meyers, J., Grogg, K. R., & Jarrett, O. (2005). Acceptability and student outcomes of a violence prevention curriculum. The Journal of Primary Prevention, 26, 401–418. http://dx.doi.org/10.1007/s10935-005-0002-z. Espelage, D. L., Low, S., Polanin, J. R., & Brown, E. C. (2013). The impact of a middle school program to reduce aggression, victimization, and sexual violence. Journal of Adolescent Health, 53, 180–186. http://dx.doi.org/10.1016/j.jadohealth.2013.02.021. Fagan, J., & Wilkinson, D. L. (1998). Social contexts and functions of adolescent violence. In D. S. Elliott, B. A. Hamburg, & K. R. Williams (Eds.), Violence in American schools: A new perspective (pp. 55–93). Cambridge, U.K.: Cambridge University Press. Farrell, A. D., & Camou, S. (2006). School-based interventions for youth violence prevention. In J. Lutzker (Ed.), Preventing violence: Research and evidence-based intervention strategies (pp. 125–145). Washington DC: American Psychological Association. Farrell, A. D., Erwin, E. H., Allison, K., Meyer, A. L., Sullivan, T. N., Camou, S., Kliewer, W., & Esposito, L. E. (2007). Problematic situations in the lives of urban African American middle school students: A qualitative study. Journal of Research on Adolescence, 17, 413–454. Farrell, A. D., Erwin, E. H., Bettencourt, A., Mays, S., Vulin-Reynolds, M., Sullivan, T., et al. (2008). Individual factors influencing effective nonviolent behavior and fighting in peer situations: A qualitative study with urban African American adolescents. Journal of Clinical Child and Adolescent Psychology, 37, 397–411. http://dx.doi. org/10.1080/15374410801955821. Farrell, A. D., Henry, D. B., & Bettencourt, A. (2013). Methodological challenges examining subgroup differences: Examples from universal school-based youth violence prevention trials. Prevention Science, 14, 1–13. http://dx.doi.org/10.1007/s11121-011-0200-2. Farrell, A. D., Mays, S., Bettencourt, A., Erwin, E. H., Vulin-Reynolds, M., & Allison, K. W. (2010). Environmental influences on fighting versus nonviolent behavior in peer situations: A qualitative study with urban African American Adolescents. American Journal of Community Psychology, 46, 19–35. http://dx.doi.org/10.1007/s10464010-9331-z. Farrell, A. D., Mehari, K. R., Mays, S., Sullivan, T. N., & Le, A. T. (2015s). Participants' perceptions of a violence prevention curriculum for middle school students: Was it relevant and useful? Journal of Primary Prevention (in press). Farrell, A. D., & Vulin-Reynolds, M. (2007). Violent behavior and the science of prevention. In D. J. Flannery, A. T. Vazsonyi, & I. D. Waldman (Eds.), The Cambridge handbook of violent behavior and aggression. New York, NY: Cambridge University Press. http://dx.doi.org/10.1017/CBO9780511816840.041. Felson, R. B., Liska, A. E., South, S. J., & McNulty, T. L. (1994). The subculture of violence and delinquency: Individual vs. school context effects. Social Forces, 73, 155–173. http://dx.doi.org/10.2307/2579921. Ferguson, C. J., San Miguel, C., Kilburn, J. C., & Sanchez, P. (2007). The effectiveness of school-based anti-bullying programs: A meta-analytic review. Criminal Justice Review, 32, 401–414. http://dx.doi.org/10.1177/0734016807311712. Fleiss, J. L. (1981). Statistical methods for rates and proportions (2nd ed.). New York: John Wiley. Frey, K. S., Pearson, C. R., & Cohen, D. (2014). Revenge is seductive if not actually sweet: Why friends matter in bullying prevention efforts. Journal of Applied Developmental Psychology. http://dx.doi.org/10/j.appdev.2014.08.002. Gottfredson, G. D., & Gottfredson, D. C. (2001). What schools do to prevent problem behaviors and promote healthy environments. Journal of Educational and Psychological Consultation, 12, 313–344. http://dx.doi.org/10.1207/S1532768XJEPC1204_02. Grumm, M., Hein, S., & Fingerle, M. (2012). Children's subjective perceptions of an aggression prevention program. International Journal of Developmental Science, 6, 97–106. http://dx.doi.org/10.3233/DEV-2012-11071. Henry, D. B., Farrell, A. D., Schoeny, M. E., Tolan, P. H., & Dymnicki, A. B. (2011). Influence of school-level variables on aggression and associated attitudes of middle school students. Journal of School Psychology, 49, 481–503. http://dx.doi.org/10.1016/j.jsp.2011.04.007. Henry, D., Guerra, N. G., Huesmann, L. R., Tolan, P. H., VanAcker, R., & Eron, L. D. (2000). Normative influences on aggression in urban elementary school classrooms. American Journal of Community Psychology, 28, 59–81. http://dx.doi.org/10.1023/A:1005142429725. Hill, C. E., Thompson, B. J., & Williams, E. N. (1997). A guide to conducing consensual qualitative research. The Counseling Psychologist, 25, 517–572. http://dx.doi.org/10. 1177/0011000097254001. Huesmann, L. R. (1998). The role of social information processing and cognitive schema in the acquisition and maintenance of habitual aggressive behavior. In R. G. Geen, E. Donnerstein, R. G. Geen, & E. Donnerstein (Eds.), Human aggression: Theories, research, and implications for social policy (pp. 73–109). San Diego, CA: Academic. Huesmann, L. R., & Guerra, N. G. (1997). Children's normative beliefs about aggression and aggressive behavior. Journal of Personality and Social Psychology, 72, 408–419. http://dx.doi.org/10.1037/0022-3514.72.2.408. Ladd, G. W., & Mize, J. (1983). A cognitive–social learning model of social-skill training. Psychological Review, 90, 127–157. http://dx.doi.org/10.1037/0033-295X.90.2.127. Matjasko, J. L., Vivolo-Kantor, A. M., Massetti, G. M., Holland, K. M., Holt, M. K., & Dela Cruz, J. (2012). A systematic meta-review of evaluations of youth violence prevention programs: Common and divergent findings from 25 years of meta-analyses and systematic reviews. Aggression and Violent Behavior, 17, 540–552. http://dx.doi.org/10.1016/j.avb.2012.06.006. McMahon, S. D., & Washburn, J. J. (2003). Violence prevention: An evaluation of program effects with urban African American students. The Journal of Primary Prevention, 24, 43–62. http://dx.doi.org/10.1023/A:1025075617356. Moffitt, T. E. (1993). Adolescent-limited and life-course persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674–701. Mytton, J. A., DiGuiseppi, C., Gough, D. A., Taylor, R. S., & Logan, S. (2002). School-based violence prevention programs: Systematic review of secondary prevention trials. Archives of Pediatric and Adolescent Medicine, 156, 752–762. http://dx.doi.org/10.1001/archpedi.156.8.752. Olweus, D., & Limber, S. P. (2010). Bullying in school: Evaluation and dissemination of the Olweus Bullying Prevention Program. American Journal of Orthopsychiatry, 80(1), 124–134. http://dx.doi.org/10.1111/j.1939-0025.2010.01015.x. Orpinas, P., Kelder, S., Frankowski, R., Murray, N., Zhang, Q., & McAlister, A. (2000). Outcome evaluation of a multi-component violence-prevention program for middle schools: The Students for Peace project. Health Education Research, 15, 45–58. http://dx.doi.org/10.1093/her/15.1.45. Orpinas, P., Parcel, G. S., McAlister, A., & Frankowski, R. (1995). Violence prevention in middle schools: A pilot evaluation. Journal of Adolescent Health, 17, 360–371. http://dx.doi.org/10.1016/1054-139X(95)00194-W. Quinn, G. P., Bell-Ellison, B. A., Loomis, W., & Tucci, M. (2007). Adolescent perceptions of violence: Formative research findings from a social marketing campaign to reduce violence among middle school youth. Public Health, 121, 357–366. http://dx.doi.org/10.1016/j.puhe.2006.11.012. Sullivan, T. N., Sutherland, K. S., Farrell, A. D., & Taylor, K. A. (2015s). An evaluation of the Second Step Middle School Program: What are the benefits for youth with and without disabilities? Remedial and Special Education (in press). Sutherland, K. S., Wright, S., & Sullivan, T. N. (2008, November). Measuring social competence skill acquisition using curriculum-based measures: Development of measures and initial findings. Paper presented at the 32nd annual Severe Behavior Disorders of Children and Youth Conference, Tempe, AZ. Wilson, S. J., Lipsey, M. W., & Derzon, J. H. (2003). The effects of school-based intervention programs on aggressive behavior: A meta-analysis. Journal of Consulting and Clinical Psychology, 71(1), 136–149. http://dx.doi.org/10.1037/0022-006X.71.1.136.