Reducing disruptive behavior in an urban school cafeteria: An extension of the Good Behavior Game

Journal of School Psychology 47 (2009) 39 – 54

Reducing disruptive behavior in an urban school cafeteria: An extension of the Good Behavior Game Barry L. McCurdy a,⁎, Amanda L. Lannie a , Ernesto Barnabas b a

Devereux Center for Effective Schools, King of Prussia, PA, United States b Lehigh University, United States

Received 8 January 2008; received in revised form 31 July 2008; accepted 22 September 2008

Abstract Non-classroom settings are often the most violence-prone areas within a school. This study investigated the impact of an interdependent group contingency on the disruptive behaviors of students in grades K–6 in an urban school cafeteria. Nine female noontime aides and National School and Community Corps staff members implemented the Lunchroom Behavior Game (LBG), a modification of the Good Behavior Game (Barrish, Saunders, & Wolf, 1969), within a multiplebaseline design across three lunch periods. Results showed a decrease in the level of disruptive behaviors following the implementation of the LBG in each lunch period and a decreasing trend for two of the three lunch periods. Discussion focuses on the use of the LBG in preventing antisocial behavior and role expansion for school psychologists interested in promoting school-based prevention strategies. © 2008 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved. Keywords: Non-classroom setting; Disruptive behavior; Good Behavior Game

In recent years a number of researchers have noted an increase in the number of youth with serious and chronic antisocial behaviors (Walker, Ramsey, & Gresham, 2003/2004). The situation is particularly evident in urban areas where there is increased exposure to associated risk factors (McCurdy, Mannella, & Eldridge, 2003; Wagner, Kutash, ⁎ Corresponding author. E-mail address: [email protected] (B.L. McCurdy). ACTION EDITOR: Scott Ardoin. 0022-4405/$ - see front matter © 2008 Society for the Study of School Psychology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jsp.2008.09.003

40

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

Duchnowski, Epstein, & Sumi, 2005). The result is that in some school settings, particularly those with less structure and adult presence, the level of disruptive behavior may be critically unsafe. Nonclassroom settings, including hallways, the playground, and the cafeteria are often the most violence-prone areas in the school (Astor & Meyer, 2001; Astor, Meyer, & Pitner, 2001). It is reported that about 50% of problem behaviors in a given school occur in nonclassroom settings (Colvin, Sugai, Good, & Lee, 1997; Nelson & Colvin, 1996). For example, Craig, Pepler, and Atlas (2000) employed naturalistic observations of children on the playground and in the classroom and found, not surprisingly, that higher frequencies of aggression occurred on the playground than in the classroom. More recently, Fabiano, Pelham, Karmazin, Panahon, and Carlson (2008) documented an average daily frequency of about 0.5 rule violations per student, or one per every two students, in their study conducted in an elementary school cafeteria. Clearly, examples such as these provide compelling evidence for the need for intervention in non-classroom settings. Compounding the issue of elevated rates of problem behavior in non-classroom settings, there are inherent difficulties that contribute to the problem of managing behavior in such settings (Colvin et al., 1997). A prescribed curriculum and effective instruction serve to minimize problem behavior in the classroom setting (Lewis, Colvin, & Sugai, 2000; Sutherland, Wehby, & Yoder, 2002) whereas in the non-classroom setting the emphasis is primarily on the supervision of student behavior. Without an instructional focus, supervisors in non-classroom settings rely on students to self-manage their behavior. Moreover, in many non-classroom settings supervisors are more likely to be of classified, rather than certified, status. In urban schools, for example, the cafeteria and the playground are often times staffed by non-professionals hired directly from the community with little or no training in managing student behavior (Astor & Meyer, 2001; Astor et al., 2001). Recognizing the contribution of non-classroom settings to the escalation of problem behavior across the school, advocates of schoolwide positive behavior support (SW-PBS) promote the use of active supervision to maintain low levels of problem behavior (Colvin, Sugai, & Patching, 1993; Lewis et al., 2000; Lewis, Sugai, & Colvin, 1998). Active supervision refers to specific behaviors (i.e., scanning, moving, and interacting) utilized by supervising adults to prevent problem behavior and promote rule-following among students (Colvin et al., 1997). Applications of active supervision and precorrection have been incorporated within a school-wide model of positive behavior support with very promising results, including decreased frequencies of problem behavior across transition areas in the school as well as during recess (Colvin et al., 1997; Jeffrey & Horner, 2008; Lewis et al., 2000). However, despite the evidence in support of active supervision, the difficulty with improving active supervision duties by adults suggests that treatment fidelity may be compromised (Lewis et al., 2000). As is often the case in non-classroom settings such as the cafeteria, when adult presence is insufficient to monitor and reward student displays of rule-following, competing behaviors are likely to be reinforced by peers (Snyder, 2002; Sugai & Horner, 2002). Cushing, Horner, and Barrier (2003), for example, found that the probability of elementary school peers providing social positives to students engaged in moderately intense disruptive behaviors, including taunting and teasing, verbal and physical disruption, profanity and inappropriate affection, ranged from 0.54 to 0.97 (M = 0.77; SD = 0.11). In other words, on

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

41

average, 77% of student displays of problem behavior in non-classroom settings result in some form of encouraging social response from peers. One possible solution may be to employ a group contingency. Group contingencies are often more practical than individual contingencies and, if arranged correctly, may capitalize on peer influence to produce a richer schedule of reinforcement for desired behavior in situations where adult-to-student contact is limited (Cooper, Heron, & Heward, 2007; Skinner, Skinner, & Sterling-Turner, 2002). One of the most well-researched group contingency interventions designed to counter the reinforcing effect of peer feedback is the Good Behavior Game (GBG; Barrish, Saunders, & Wolf, 1969). The GBG is an interdependent group contingency with demonstrated effectiveness in reducing disruptive behavior. In classroom versions of the GBG, students are grouped into teams and teams are assigned points when any team member violates a classroom rule. Teams with the fewest point totals at the end of the game period, or those with points totaling less than a set criterion, are considered game “winners” and are eligible for a reward (e.g., victory tag, free time). In the earliest investigation of the GBG, researchers decreased the out-of-seat and talking-out behaviors of a fourth grade class during math and reading. The class included several students with serious problem behaviors (Barrish et al., 1969). Since then, numerous follow-up studies have been conducted supporting the effectiveness of the intervention and extending the findings to include additional grade levels (Darveaux, 1984; Harris & Sherman, 1973; Lannie & McCurdy, 2007); additional target behaviors, including tattling, academic performance, aggression and shyness (Darveaux, 1984; Dolan et al., 1993; Harris & Sherman, 1973; Medland & Stachnik, 1972); and high risk populations, including students with emotional/behavioral disorders (Salend, Reynolds, & Coyle, 1989). More recent investigations have looked at the longitudinal impact of the GBG on populations of children with, and at risk for, disruptive behavior disorders. For example, researchers from the Prevention Research Center at Johns Hopkins (Ialongo, Poduska, Werthamer, & Kellam, 2001; Kellam, Ling, Merisca, Brown, & Ialongo, 1998) found a positive impact on the aggressive behaviors of first grade male students exposed to the GBG and lower levels of aggressive behavior among sixth grade students who were exposed to the GBG in first and second grades. Likewise, in a randomized control trial involving 671 students in first and fifth grades from schools in high juvenile crime areas, Eddy, Reid, and Fetrow (2000) incorporated a playground version of the GBG into a more comprehensive intervention that included classroom-based social and problem-solving skills training as well as group-based parent training. Results of the study showed the immediate effects of less aggressive behavior on the playground as well as improved teacher perceptions of classroom behavior and less aversive parent–child interactions. The longer-term impact on students in middle schools included significant delays in the time to first report of association with deviant peers, adoption of behaviors associated with deviant peers (i.e., alcohol and marijuana use), and the time to first arrest. Taken together, results of these epidemiological investigations have led Embry (2002) to promote the GBG as a “universal behavioral vaccine” and federal agencies, including the Substance Abuse and Mental Health Services Administration (SAMHSA), to dub the intervention as a “best practice” for the prevention of substance abuse and violent behavior. Despite the impressive number of studies and the consistent findings, to date there have been only two investigations to examine the impact of the GBG in non-classroom settings.

42

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

In addition to the study by Eddy et al. (2000), Fishbein and Wasik (1981) evaluated an adaptation of the GBG for use in a school library. The modification included awarding team points for following the rules rather than the traditional method of awarding points for not following the rules. Similar to previous investigations, disruptive and off-task behaviors decreased. Given the surge in antisocial behaviors across most communities and the resulting impact on schools, particularly non-classroom settings, questions remain as to the feasibility of implementing the GBG in those settings most prone to violent behavior — the cafeteria and playground (Astor & Meyer, 2001; Leff, Power, Costigan, & Manz, 2003). The current study investigated the impact of a modified GBG, herein referred to as the Lunchroom Behavior Game (LBG), in an urban, elementary school cafeteria. Specifically, the following questions were addressed: (a) what is the impact of the LBG on disruptive behavior, and (b) how acceptable is the intervention for cafeteria staff and students? Method Setting and participants The study was conducted in the lunchroom of a public elementary school housing grades K to 6. The school is part of a large urban school district located in the northeastern United States and was one of several schools targeted by the school district to receive grant-funded services to address drug, alcohol, violence and safety concerns. The school was not implementing SW-PBS at the time of the intervention. Alternately, the climate of discipline in the building was primarily reactive in nature, relying heavily on suspensions and involvement of the school police officer to address behavior issues. The school was comprised of approximately 615 students, with the racial demographics as follows: 79% African-American, 19% Hispanic, 0.8% Caucasian, 0.4% Native American, and 0.3% Asian. A majority of the students (86.9%) received free or reducedprice lunch. Results from the 2006 Terra Nova testing show that 38.4% and 50.1% of students in 1st through 6th grades performed in the bottom quartile for reading and math, respectively. The study was conducted in the lunchroom during the school lunch periods. The lunchroom was located in the front corner of the school. It was adjacent to the auditorium and accessible through adjoining doors. Typical entry to the lunchroom was gained through one of two doors. The lunchroom routine dictated that students enter one door for lunch and exit the other door following dismissal from lunch. The lunchroom contained ten rectangular tables, each comprised of four connected smaller table segments. The rectangular tables extended from the right and left sides of the lunchroom and were situated parallel to one another in two sections of five tables each. The two sections were separated by a 7-foot walkway for students to walk with their lunches to their assigned tables. Three lunch periods occurred daily, each serving approximately 200 students. Students were assigned to each lunch period by grade as follows: lunch period one served grades kindergarten and 3, lunch period two served grades 1 and 2, and lunch period three served grades 4, 5, and 6. Students in all lunch periods participated in the study. Adult participants and primary intervention agents in the study included 10 school staff members. All adult participants were African-American females. They ranged in age from

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

43

18 to 56 years. Six of the staff members were designated noon-time aides (NTAs) and each had primary responsibility for one classroom of 30 to 35 students per lunch period. Activities included monitoring student behavior, assisting with securing and opening student lunches, and cleaning tables after each lunch period. All NTAs had a high school diploma or its equivalent and from 3 to 25 years of experience working in the school lunchroom. Four additional staff members from the National School and Community Corps (NSCC) were assigned to the school to assist in monitoring classrooms during lunch. Information on the educational background of NSCC staff was not made available to the researchers. However, having a high school diploma is not a requirement for employment. NSCC staff members' experience working in the school and lunchroom ranged from 1 to 2 years. Note that there were originally five NSCC staff at the beginning of the initial baseline, but one staff member was transferred to another school. Her position was not replaced, and thus the number of NSCC staff remained at four for the duration of the study. Dependent variables and measurement Disruptive behavior served as the primary dependent variable. Disruptive behavior was defined along five categories of behavior: out of seat, play fighting, physical contact with force, throwing objects, and screaming. Out of seat was operationally defined as the student's buttocks not touching the seat after placing food on the table (without being excused), including straddling the bench and legs facing outward. Play fighting was operationally defined as moving body parts (e.g., arms, legs, head) in another student's direction to simulate fighting, without physical contact, including tugging at an object. Physical contact with force was operationally defined as physical contact between two or more students with the use of force, including, but not limited to, pushing, hitting, kicking, and punching. Throwing objects was operationally defined as propelling an object through the air, including, but not limited to, food, pencils, and balls. Screaming was operationally defined as a student's voice heard above all others in the lunchroom. All behaviors were recorded as a frequency count by two observers. Each observation ranged in length from 10 to 15 min, depending upon time availability. Observations were divided into 15-s intervals. The expiration of an interval served as a prompt to commence observing students at the next segment/table. Behavior occurrences were counted across each interval. The total frequency of defined behavioral occurrences was divided by the length of the observation in minutes to determine the rate per min of disruptive behavior. Observational data were collected for 20 sessions, with experimental sessions conducted 4 to 5 days per week for 4 weeks. Groups of students were observed for each interval. The rectangular tables at which the students were seated were naturally divided into four connected smaller table segments (segments 1–4) with approximately eight students seated at each segment. The observation commenced with a randomly selected rectangular table and the corresponding group seated at the first segment (i.e., segment 1). All four segments that made up a rectangular table were observed (four intervals of 15 s) prior to observations being conducted at another table. The observation then moved to the parallel table located

44

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

across the walkway. Table segments with less than three students were skipped given that this grouping occurred rarely and was not representative of the typical seating arrangements in the lunchroom. During the course of each observation, all groups of students were observed approximately twice. Materials The following materials were used during daily implementation of the LBG: clipboards, whistles, daily and weekly recording sheets, posters, calculators, and an electronic megaphone. Daily recording sheets were used to record points accrued by classrooms during the lunch period. The daily recording sheets were printed on 8.5 in. × 11 in. paper and contained a matrix with identifying classroom numbers in the left-hand column, a middle column to record points and a right-hand column to record the total points by classroom. Weekly recording sheets contained one box for recording the point totals for each classroom team from Monday through Friday. Dry-erase laminated posters for each lunch period, each measuring 42 in. × 53 in., were hung in visually and physically accessible areas in the lunchroom. Dry-erase markers were used to record daily team totals on the posters. Lunchroom staff used a hand-held electronic megaphone to announce team totals to all students prior to students exiting the lunchroom. Training for the NTA and NSCC staff Procedural training for the LBG occurred in two phases: (a) the development of clear expectations, or rules, for the lunchroom, and (b) training in the implementation of the LBG. Rule development During a 1-hour session, NTA and NSCC staff met with the research team to develop expectations for student behavior in the lunchroom. Staff members were asked to identify a list of undesirable behaviors that occurred with some frequency. Examples included leaving the lunch table without permission, running in the lunchroom, fighting, failing to clean the area after eating, and playing “squeeze the lemon” (i.e., students on both ends of a table bench slide inward to squeeze students sitting in the middle of the bench). Based on the list of non-sanctioned behaviors, staff members were then assisted in identifying a list of socially appropriate, alternative behaviors, or expectations, for students when in the lunchroom. Despite efforts to keep the number of expectations to five or fewer (Horner, Sugai, Todd, & Lewis Palmer, 2005), NTA and NSCC staff insisted on an expanded number of expectations to adequately account for the variety of problem behaviors frequently observed in the lunchroom. In an effort to promote acceptability of the intervention, expectations were combined and further refined into a final list of seven lunchroom rules that included (a) sit four (students) to a bench, (b) use an “indoor” voice, (c) ask permission to leave your seat, (d) keep your hands, body and objects to yourself, (e) follow staff directions the first time they are given, (f) keep your area clean, and (g) walk at all times. The research team then created posters listing the seven rules and posters were displayed in prominent locations around the lunchroom.

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

45

Training in LBG procedures The researchers trained both the NTA and NSCC staff in the procedures of the LBG. Training occurred during one 90-min session. Once trained, members of the research team coached the NTA and NSCC staff in the implementation of the LBG procedures during the first lunch period across two consecutive days. Following an overview of the LBG, training addressed the following topics: (a) moving and scanning to monitor student behavior, (c) identifying undesirable behaviors and recording point losses, (c) totaling the point scores for each class, (d) recording point scores and providing feedback to all students, and (e) selecting winning teams. After reviewing the topics in detail, the NTA and NSCC staff practiced the LBG procedures during a role-play activity in the lunchroom while the trainers provided feedback. Preparing students for the Lunchroom Behavior Game Prior to implementing the LBG in the lunchroom, the authors introduced the LBG procedures to all teachers of students in grades K to 6. Teachers were asked to review the LBG procedures with their students and to specifically teach students the lunchroom rules using lesson plans developed by the authors. Teachers were prompted to explain the game and teach the rules to their students 1 week prior to implementation of the LBG for their respective Cohort. Lesson plans incorporated a direct instructional format including the use of role play to present exemplars and non-exemplars of following the rules (Engelmann & Carnine, 1991; Gunter & Reed, 1997). Teachers accompanied their students to the lunchroom to teach the 15-min lesson plans and to review the procedures for the LBG. Experimental procedures Baseline Observations of disruptive behavior were conducted across the three lunch periods. Eight to 10 classrooms attended each lunch period. Whole classrooms were assigned to one rectangular table (which incorporated four table segments). During baseline, the NTA and NSCC staff monitored student behavior utilizing their typical procedures of verbal reprimands, brief time-out in the lunchroom but away from the assigned lunch table, permanent removal to an empty lunch table for the entire lunch period, and removal from the lunchroom by school police. Data were only collected on rates of disruptive behavior. No data were collected on the use of disciplinary procedures by the NTA and NSCC staff. Lunchroom Behavior Game The LBG was implemented daily during each lunch period. Each classroom functioned as a team. The LBG commenced when more than 75% of students had received their lunch and were seated at their assigned tables. For each observed occurrence of disruptive behavior, the NTA or NSCC staff member identified the behavior and made a tick mark (i.e., recorded a point) on the recording sheet corresponding to the student's team (i.e., classroom). Specifically, the NTA/NSCC staff member performed the following: (a) blew the whistle to obtain student/team attention, (b) identified the rule infraction and appropriate,

46

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

alternative behavior (e.g., “you are out of your seat without permission. You are to raise your hand and ask permission to leave your seat”), and (c) made a tick mark corresponding to the student's team on the recording sheet attached to the clipboard. The 10 NTA/NSCC staff members continued to identify and record occurrences of disruptive behavior throughout the lunch period. With 5 min remaining in the lunch period, a designated NTA collected the recording sheets from all adults in the lunchroom. She tallied the total number of tick marks or points across all adult recordings for each classroom. Once tallied, she announced the day's point tally for each classroom via a megaphone to all students in the lunchroom while a second designated NTA recorded the point tally for each classroom on the poster. At the end of the week, the daily point tally and weekly point tally were announced. On Monday morning, the Dean of Students identified the weekly winning classrooms for each lunch period during the morning announcements. Weekly winners were those classrooms that did not exceed the criterion for the week. The Dean of Students chose the criterion weekly so that classrooms with the lowest point totals (i.e., the majority) were designated as “winners” and those with clearly discrepant and higher totals did not win. To prevent students from becoming discouraged early in the week, the weekly criterion remained unknown to both the students and adults in the lunchroom until announced on Monday morning. Rewards for winning classrooms included edible items, small tangibles, and certificates to earn movie time and special parties in the classroom. Experimental design A multiple-baseline design across lunch periods was utilized to measure the impact of the LBG on student disruptive behavior. The intervention was implemented with each lunch period when the baseline data were stable or ascending and a clear decrease in behavior occurred in concert with implementation of the intervention in the previous lunch period. Interobserver agreement The second and third authors served as secondary and primary observers for the study, respectively. The observers attained a level of 80% interobserver agreement (IOA) prior to commencement of the study. IOA between the two observers was assessed during 30% of experimental sessions across all conditions. IOA was calculated for each discrete behavior by interval utilizing point-by-point agreement, and dividing the number of agreements by the number of agreements plus disagreements and multiplying the result by 100 to obtain a percentage (Kazdin, 1982). IOA was then aggregated across behaviors to yield a mean IOA for each session. Mean IOA across lunch periods was 88% (SD = 3.15; range, 85%–93%) across all sessions sampled. Treatment fidelity Protocols listing the components of the intervention were devised to assess treatment fidelity. Fidelity measurement included procedural steps (e.g., recording sheets attached to clipboards, carry whistle, hold clipboards) and adult behavior for carrying out the intervention

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

47

for behavioral infractions (e.g., when a behavioral infraction occurs, inform the student of the rule infraction and the appropriate alternative behavior, and make a tick mark) for a total of 8 steps. The measurement of fidelity involved the observation of all staff for procedural steps (e.g., recording sheets attached to clipboards, carry whistle), as well as individual staff for occurrences of discrete behaviors (e.g., responding to behavioral infractions). The most commonly missed steps included (a) failing to consistently follow the steps for addressing behavioral infractions and (b) failing to record team point losses on the poster during announcement. During 45% of the sessions, the experimenter recorded whether each component of the intervention was completed. Across all sessions sampled, the average percent of steps followed was 89% (SD = 1.0; range, 75%–100%). Treatment acceptability To assess students' acceptability of the LBG, the Children's Intervention Rating Profile (CIRP; Turco & Elliott, 1986) was administered. The CIRP is a seven-item, one-factor scale assessing the acceptability of the intervention with an average coefficient alpha of 0.86 (Turco & Elliott, 1986). The CIRP was adapted by (a) including the name of the intervention within each item (e.g., the lunchroom game is fair), (b) shortening the scale to six items (i.e., the item regarding difficulty of intervention implementation was omitted since students did not implement), and (c) truncating the Likert-type rating scale from 6 points to 3 points for students in kindergarten through 2nd grade. Three to four students from each classroom were randomly selected to complete the CIRP. Approximately 20% of the student population was sampled. Higher scores on the CIRP indicated more favorable approval by students. The Intervention Rating Profile (IRP; Martens, Witt, Elliott, & Darveaux, 1985) was administered to assess the lunchroom staff's intervention acceptability. The IRP is a 15item, one-factor scale. A Cronbach's alpha of 0.91 was found in the original study of its psychometric properties (Martens et al., 1985). To reflect the intervention implemented, the IRP was adapted by including the name of the intervention in each of the items. The IRP contains 15 items and is assessed along a 6-point, Likert-type scale. Responses range from 1 (“Strongly disagree”) to 6 (“Strongly agree”). Three additional items were added to the IRP to evaluate (a) the ease of daily use of the LBG, (b) the ease of use in managing student behavior, and (c) the likelihood of continued use of the LBG. Results Disruptive behavior Fig. 1 depicts the disruptive behaviors per min across lunch periods. The data depicted represent a total count of disruptive behavior per observation (Lewis et al., 1998). A total count was depicted in favor of discrete behaviors due to the low frequency of occurrence for some behaviors (e.g., screaming). The total count was then divided by the length of the observation (in min) to calculate the rate of disruptive behavior. Rate of, rather than total, disruptive behavior was calculated to permit comparisons within and across lunch periods, as the length of the periods could vary from 10 to 15 min, depending upon when students arrived in the lunchroom following teacher escort from their classroom. Upon implementation of the LBG

48

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

Fig. 1. Rate per minute of student disruptive behavior across three lunch periods.

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

49

during each lunch period, there was an immediate decrease in the rate of disruptive behavior. The rate of disruptive behavior remained below baseline levels during all lunch periods. In the first lunch, the mean (and standard deviation) per min rate of disruptive behavior decreased from 4.73 (0.65) during baseline to 2.02 (0.78) during intervention, for an overall reduction of 2.71 disruptive behaviors per min. Similarly, in the second lunch period, disruptive behavior decreased from a mean rate of 6.66 (0.97) during baseline to 2.75 (0.87) during intervention, for an overall reduction of 3.91 disruptive behaviors per min. Finally, in the third lunch period disruptive behavior decreased from a mean rate of 7.1 (1.67) during baseline to 2.46 (0.46) during intervention, marking the largest reduction of disruptive behaviors (4.64 per min). In addition to a change in level, the implementation of the LBG produced a downward trend in the data across the first two lunch periods followed by a slightly increasing trend in the third lunch period. Within the data series there are two data points that should be noted. The first is the final point of baseline for the second lunch period (i.e., session 8). This point represents the lowest rate of disruptive behavior during baseline for this period. Although it descended in the desired direction prior to implementation of the LBG, there was an immediate change in level upon implementation of the LBG, with all data points for the intervention condition remaining below the datum from session 8. The second notable datum occurred in the third lunch period where a single overlapping data point (i.e., session 17) occurred. This datum overlapped with one session of particularly low rate disruptive behavior during baseline (i.e., session 4), which was not reflective of the overall baseline rate of behavior. Overall, the LBG resulted in clear changes in the level and trend of disruptive behavior in the first and second lunch period and a change in the level of disruptive behavior in the third lunch period. Treatment acceptability Students in grades kindergarten through 2nd grade as well as students in three special education classrooms which spanned grades K to 6 completed the CIRP along a 3-point, Likert-type scale, ranging from 1 (“Not at all”) to 3 (“A lot”). The mean rating was 2.25 out of 3 (SD = 0.44; range, 1.6 to 2.76), suggesting that the intervention was moderately acceptable to younger students. Students in grades three to six completed the measure along a 6-point, Likert-type scale, ranging from 1 (“I do not agree”) to 6 (“I agree”). The mean rating for this older group of students was 4.34 out of 6 (SD = 1.2; range, 2.47 to 5.29), suggesting that older students found the intervention to be acceptable (higher scores indicate greater acceptability). All ten participating NTA/NSCC staff members completed the IRP. The mean item rating across the 15 items was 5.03 (SD = 0.49). The mean item rating for the three additional items was 5.33 (SD = 0.15). The results from the staff member ratings suggest that the LBG was highly acceptable. Discussion The purpose of the present study was to investigate the effects of the LBG, a modified version of the Good Behavior Game, on the disruptive behavior of students in an urban,

50

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

elementary school lunchroom. Results of the study indicate that the LBG was effective in reducing disruptive behavior in the lunchroom during each lunch period and across all grade levels. Implementation of the LBG in each consecutive lunch period resulted in an immediate and sustained decrease in the level of disruptive behavior with only one instance of overlapping data (i.e., see sessions 4 and 17 during third lunch). The overlap was due to an unusually low baseline data point that occurred on the fourth day of the study. On this day, the investigators observed one staff member spontaneously reviewing the rules with individual students similar to a teaching format. After reviewing the purpose of the project and importance of delaying intervention, the staff member ceased “teaching” and the level of disruptive behavior immediately returned to previous levels and continued in an upward trend. In addition to the effects on student behavior, the LBG was rated as highly acceptable by lunchroom staff and acceptable by a sample of the school's student population. The results of the acceptability measures are just as important as the reductions in disruptive behavior. Although the behavioral outcomes obtained with the present study testify to the strength of the intervention, the acceptability results provide a glimpse of the feasibility of such interventions, possibly the vital ingredient, for promoting generalization in non-classroom settings (Elliott, Witt, Kratochwill, & Stoiber, 2002). Impacting antisocial behavior Although the current investigation successfully demonstrated the use of a group contingency on a large scale in one school lunchroom, it has broader implications for school systems mired with growing discipline problems. Many schools concerned with the escalation of antisocial behavior have adopted a proactive approach to discipline known as SW-PBS (Horner et al., 2005; McCurdy et al., 2003). Based on a three-tiered model of prevention adapted from the U.S. public health service, SW-PBS offers a systemic approach to discipline that emphasizes clearly defined expectations for student behavior, the active teaching of rules, publicly and positively acknowledging students for following the rules, a hierarchy of corrective responses to address problem behavior, and team-based formative evaluation. As typically recommended within a SW-PBS model, active supervision is promoted to enhance behavior management in non-classroom settings (Jeffrey & Horner, 2008). The difficulty associated with promoting the maintenance of active supervision skills, however, combined with the potentially high rate of peer reinforcement for other, disruptive behavior in settings where adult-to-student ratios are particularly lean may serve to limit the impact of an active supervision intervention (Sugai & Horner, 2002). For example, it is a widely acknowledged fact that urban school environments experience a higher number of discipline problems and that aggressive and deviant behavior in these settings is often reinforced by peers (Baker, Kamphaus, Horne, & Winsor, 2006; Dishion, Spracklen, Andrews, & Patterson, 1996; Snyder, 2002). Attempts to impact this behavior through the differential reinforcement of other, prosocial behaviors (i.e., active supervision) is likely to be too weak to effectively reduce disruptive and aggressive behavior, particularly in schools where a SW-PBS model is not in place. Alternately, Henry et al. (2000) found that disruptive and aggressive behaviors are less likely to increase when supervisors as well as

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

51

peers actively discourage such behavior. The LBG with an emphasis on identifying nonsanctioned behaviors and discouraging their occurrence through point assignment promotes peer involvement in discouraging such behavior. A possible alternative explanation for the obtained results is that peer involvement may have functioned to punish problem behavior (Bellafiore & Salend, 1983). Although the sole presentation of aversive consequences can inadvertently increase the rate of inappropriate behavior (i.e., deleterious intervention effects), this phenomenon was not observed in the present study. The addition of a system to promote positive behavior, like the weekly rewards in the current study, may have been critical in minimizing the negative side effects of punishment (Salend, Jantzen, & Giek, 1992). Limitations of the study Positive results aside, there were several limitations of the current investigation that should be noted. Despite acceptable measures of interobserver agreement, it is likely that the overall mean agreement would have been enhanced by improving the operational definitions of the target behaviors. For example, IOA may have been better had the definition of “physical contact with force” been “non-sanctioned physical contact, including pushing, hitting with open hand or fist, and kicking.” In addition, the use of partialinterval recording, as opposed to frequency counts, would better account for behaviors that have no clear termination point (e.g., screaming). A second limitation of the study is that there was no planned attempt to assess the maintenance of student behavior in the absence of the intervention. Except for the few investigations designed to examine the impact of the GBG using a withdrawal design (e.g., ABAB; Barrish et al., 1969; Medland & Stachnik, 1972), no investigations have examined the maintenance effects on student behavior after a lengthy period of implementation. It may be that, given the ease of implementation, the game-like quality of the intervention and the powerful effect on behavior, teachers and others implementing the GBG see no reason to terminate its use. However, since implementation within a non-classroom setting and for a large number of students often requires more planning, future investigations should include an evaluation of fading procedures as well as a follow-up phase to assess the longterm impact on student behavior. Finally, anecdotal evidence suggests that teachers introduced additional classroombased contingencies intending to enhance the effects of the intervention. For example, one third grade teacher implemented a no homework contingency if the class earned zero points for the lunch period. Similarly, this same teacher engaged in a between-classroom competition with another third grade teacher to see which of their two classrooms could earn the fewest weekly points. Although the introduction of additional contingencies by teachers indicates their approval of and support for the intervention, this response introduced an unexpected limitation to the results. However, given that the evidence for additional contingencies affected only two classrooms at most, it is unlikely that the effects of the LBG were primarily due to the additional contingencies. Nevertheless, future investigations should find a way to assess and control for supplemental teacher interventions, particularly classroom-based contingencies that may serve to confound the results of the investigation.

52

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

Implications for school psychologists The LBG offers school psychologists interested in prevention and early intervention yet another opportunity to expand their roles and, ultimately, the scope of their service delivery. Literally hundreds of students can be positively affected by the implementation of the LBG in non-classroom settings like the lunchroom or even the playground. These areas of the school environment are reputed locations where bullying and other acts of violence are routinely committed and where students, particularly those already exposed to numerous risks for the adoption of antisocial behavior, learn to survive (Leff, Power, Manz, Costigan, & Nabors, 2001; Leff, Power, Costigan, & Manz, 2003). School psychologists, with their depth of training in and knowledge of behavioral principles, are in the ideal position to assist schools through training and coaching professionals as well as para-professionals in the delivery of the LBG and formatively evaluating its impact on the overall climate of the school. References Astor, R. A., & Meyer, H. A. (2001). The conceptualization of violence-prone school subcontexts: Is the sum of the parts greater than the whole? Urban Education, 36, 374−399. Astor, R. A., Meyer, H. A., & Pitner, R. O. (2001). Elementary and middle school students' perceptions of violence-prone school subcontexts. The Elementary School Journal, 101, 511−528. Baker, J. A., Kamphaus, R. W., Horne, A. M., & Winsor, A. P. (2006). Evidence for population-based perspectives on children's behavioral adjustment and needs for service delivery in schools. School Psychology Review, 35, 31−46. Barrish, H. H., Saunders, M., & Wolf, M. M. (1969). Good Behavior Game: Effects of individual contingencies for group consequences on disruptive behavior in a classroom. Journal of Applied Behavior Analysis, 2, 119−124. Bellafiore, L. A., & Salend, S. J. (1983). Modifying inappropriate behaviors through a peer confrontation system. Behavioral Disorders, 8, 274−279. Colvin, G., Sugai, G., Good, R. H., & Lee, Y. (1997). Using active supervision and precorrection to improve transition behaviors in an elementary school. School Psychology Quarterly, 12, 344−363. Colvin, G., Sugai, G., & Patching, B. (1993). Precorrection: An instructional approach for managing predictable problem behaviors. Intervention in School and Clinic, 28, 143−150. Cooper, J. O., Heron, T. E., & Heward, W. L. (2007). Applied behavior analysis, (2nd ed.) Pearson: Upper Saddle River, NJ. Craig, W. M., Pepler, D., & Atlas, R. (2000). Observations of bullying in the playground and in the classroom. School Psychology International, 21, 22−36. Cushing, L. S., Horner, R. H., & Barrier, H. (2003). Validation and congruent validity of a direct observation tool to assess student social climate. Journal of Positive Behavior Interventions, 5, 225−237. Darveaux, D. X. (1984). The Good Behavior Game plus merit: Controlling disruptive behavior and improving student motivation. School Psychology Review, 13, 510−514. Dishion, T. J., Spracklen, K. M., Andrews, D. M., & Patterson, G. R. (1996). Deviancy training in male adolescent friendships. Behavior Therapy, 27, 373−390. Dolan, L. J., Kellam, S. G., Brown, C. H., Werthamer-Larsson, L., Rebok, G. W., Mayer, L. S., et al. (1993). The short-term impact of two classroom-based preventive interventions on aggressive and shy behaviors and poor achievement. Journal of Applied Developmental Psychology, 14, 317−345. Eddy, J. M., Reid, J. B., & Fetrow, R. A. (2000). An elementary school-based prevention program targeting modifiable antecedents of youth delinquency and violence: Linking the interests of families and teachers (LIFT). Journal of Emotional and Behavioral Disorders, 8, 165−176. Elliott, S. N., Witt, J. C., Kratochwill, T. R., & Stoiber, K. C. (2002). Selecting and evaluating classroom interventions. In M. R. Shinn, H. M. Walker, & G. Stoner (Eds.), Interventions for academic and behavior

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

53

problems II: Preventive and remedial approaches (pp. 243−294). Bethesda: MD: National Association of School Psychologists. Embry, D. D. (2002). The Good Behavior Game: A best practice candidate as a universal behavioral vaccine. Clinical Child and Family Psychology Review, 5, 273−297. Engelmann, S., & Carnine, D. (1991). Theory of instruction: Principles and applications. Eugene, OR: ADI Press. Fabiano, G. A., Pelham, W. E., Karmazin, K., Kreher, J., Panahon, C. J., & Carlson, C. (2008). A group contingency program to improve the behavior of elementary school students in a cafeteria. Behavior Modification, 32, 121−132. Fishbein, J. E., & Wasik, B. H. (1981). Effect of the Good Behavior Game on disruptive library behavior. Journal of Applied Behavior Analysis, 14, 89−93. Gunter, P. L., & Reed, T. M. (1997). Academic instruction of children with emotional and behavioral disorders using scripted lessons. Preventing School Failure, 42, 33−38. Harris, V. W., & Sherman, J. A. (1973). Use and analysis of the “Good Behavior Game” to reduce disruptive classroom behavior. Journal of Applied Behavior Analysis, 6, 405−417. 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. Horner, R. H., Sugai, G., Todd, A. W., & Lewis-Palmer, T. (2005). Schoolwide positive behavior support. In L. M. Bambara & L. Kern (Eds.), Individualized supports for students with problem behaviors: Designing positive behavior plans (pp. 359−390). New York: Guilford. Ialongo, N., Poduska, J., Werthamer, L., & Kellam, S. (2001). The distal impact of two first-grade preventive interventions on conduct problems and disorder in early adolescence. Journal of Emotional and Behavioral Disorders, 9, 146−160. Jeffrey, J.L. & Horner, R.H. (2008). Evaluation of a cafeteria social skills program on student behavior and peer-to-peer interactions in a non-classroom setting within an urban elementary school. Kazdin, A. E. (1982). Single-case research designs: Methods for clinical and applied settings. New York: Oxford University Press. Kellam, S. G., Ling, X., Merisca, R., Brown, C. H., & Ialongo, N. (1998). The effect of the level of aggression in the first grade classroom on the course and malleability of aggressive behavior into middle school. Development and Psychopathology, 10, 165−185. Lannie, A. L., & McCurdy, B. L. (2007). Preventing disruptive behavior in the urban classroom: Effects of the Good Behavior Game on student and teacher behavior. Education and Treatment of Children, 30, 85−98. Leff, S. S., Power, T. J., Manz, P. H., Costigan, T. E., & Nabors, L. A. (2001). School-based aggression prevention programs for young children: Current status and implications for violence prevention. School Psychology Review, 30, 344−362. Leff, S. S., Power, T. J., Costigan, T. E., & Manz, P. H. (2003). Assessing the climate of the playground and lunchroom: Implications for bullying prevention programming. School Psychology Review, 32, 418−430. Lewis, T. J., Sugai, G., & Colvin, G. (1998). Reducing problem behavior through a school-wide system of effective behavioral support: Investigation of a school-wide social skills training program and contextual interventions. School Psychology Review, 27, 446−459. Lewis, T. J., Colvin, G., & Sugai, G. (2000). The effects of precorrection and active supervision on the recess behavior of elementary students. Education and Treatment of Children, 23, 109−121. Martens, B. K., Witt, J. C., Elliott, S. N., & Darveaux, D. X. (1985). Teacher judgments concerning the acceptability of school-based interventions. Professional Psychology — Research and Practice, 16, 191−198. McCurdy, B. L., Mannella, M. C., & Eldridge, N. (2003). Positive behavioral support in urban schools: Can we prevent the escalation of antisocial behavior? Journal of Positive Behavior Interventions, 5, 158−170. Medland, M. B., & Stachnik, T. J. (1972). Good-Behavior Game: A replication and systematic analysis. Journal of Applied Behavior Analysis, 5, 45−51. Nelson, R. J., & Colvin, G. (1996). Designing supportive school environments.. Special Services in the Schools, 11, 169−186. Salend, S. J., Jantzen, N. R., & Giek, K. (1992). Using a peer confrontation system in a group setting. Behavioral Disorders, 17, 211−218. Salend, S. J., Reynolds, C. J., & Coyle, E. M. (1989). Individualizing the Good Behavior Game across type and frequency of behavior with emotionally disturbed adolescents. Behavior Modification, 13, 108−126.

54

B.L. McCurdy et al. / Journal of School Psychology 47 (2009) 39–54

Skinner, C. H., Skinner, A. L., & Sterling-Turner, H. E. (2002). Best practices in contingency management: Application of individual and group contingencies in educational settings. In A. Thomas & J. Grimes (Eds.), Best practices in school psychology IV (pp. 817−830). Bethesda, MD: National Association of School Psychologists. Snyder, J. (2002). Reinforcement and coercion mechanisms in the development of antisocial behavior: Peer relationships. In J. Reid, G. Patterson, & L. Snyder (Eds.), Antisocial behavior in children and adolescents: A developmental analysis and model for intervention (pp. 101−122). Washington, DC: American Psychological Association. Sugai, G., & Horner, R. (2002). The evolution of discipline practices: School-wide positive behavior supports. Child and Family Therapy, 24(1–2), 23−50. Sutherland, K. S., Wehby, J. H., & Yoder, P. J. (2002). Examination of the relationship between teacher praise and opportunities for students with EBD to respond to academic requests. Journal of Emotional and Behavioral Disorders, 10, 5−13. Turco, T. L., & Elliott, S. N. (1986). Students' acceptability ratings of interventions for classroom misbehaviors: A study of well-behaving and misbehaving youth. Journal of Psychoeducational Assessment, 4, 281−289. Wagner, M., Kutash, K., Duchnowski, A. J., Epstein, M. H., & Sumi, W. C. (2005). The children and youth we serve: A national picture of the characteristics of students with emotional disturbances receiving special education. Journal of Emotional and Behavioral Disorders, 13, 79−96. Walker, H. M., Ramsey, E., & Gresham, F. M. (2003/2004). Heading off disruption: How early intervention can reduce defiant behavior and win back teaching time. American Educator, Winter, 6−21.