What the Research Says (or Doesn’t Say)
Regarding In- and Out-of-School Suspension Practices and Alternatives
NWREL’s Office of Planning and Service Coordination responds to requests that come from the field regarding current educational concerns and issues. Our staff provide readily available research articles or make referrals to technical service providers and other sources.
Recently, a middle school principal from Maryland requested research on the effectiveness of in- and out-of-school suspension practices to alleviate disciplinary issues.
Traditionally out-of-school suspension has been a disciplinary measure for serious infractions, such as physical violence, alcohol and drug possession, and possession of weapons. However out-of-school suspension is often also used for offenses such as poor attendance, truancy, and discipline issues. For example, the state of Tennessee reported administering almost 30,000 suspensions for attendance issues in 2001-2002 (Palmisano & Potts, 2004). In fact, data from the Indicators of School Crime and Safety indicate that although fewer schools are reporting the use of suspension for serious infractions, more schools are using suspension for “insubordination” in 2003 (22 percent) than four years earlier (18 percent) (Devoe et al., 2004; Dinkes, Cataldi, Kena, & Baumet, 2006).
What does the research say about the effectiveness of out-of-school suspension on such disciplinary issues?
In our scan of the literature, we did not find any experimental studies that analyzed the effectiveness of out-of-school suspension. Skiba (2000) analyzed school discipline practices and noted that “there appears to be little evidence, direct or indirect, supporting the effectiveness of suspension or expulsion for improving behavior or contributing to overall school safety” (p. 13). The general discussion about out-of-school suspension practices is that is “pushes” students out of school for major and relatively minor infractions. Also, data show that many students receiving suspensions are repeat offenders and that suspensions can actually reinforce negative behaviors rather than punish them (Tobin, as cited in Skiba, 2000). A recent report from the National Dropout Prevention Center reiterates this and indicates that suspensions are considered a predictor for students dropping out of school altogether (Hammond, 2007).
What about in-school suspension as an alternative?
If out-of-school suspension does not appear to be working to change student behavior, what about the popular practice of in-school suspensions (ISS) as an alternative or precursor to out-of-school prevention programs? ISS programs typically involve removing offending students from their regular classroom and placing them in a separate classroom for a certain time period. ISS programs vary—a common approach is simply to have students monitored while they do schoolwork; however, other programs provide tutoring, instruction, counseling, behavior skills coaching, violence prevention, conflict resolution, and character education. Indeed, one problem with evaluating the effectiveness of ISS programs is that they vary so widely in purpose, organization, curriculum, and duration.
Unfortunately, as with out-of-school suspension, much of the literature on ISS consists of descriptions of programs in various schools, with anecdotes about its impact, or opinions based on surveys of staff and students. As Blomberg explains in his literature review, “The story of in-school suspension is one of both promise and pessimism” (2004, p. 5). There are very few objective studies examining ISS effects on behavior. Jackson (2006), who did an extensive review of the research for his dissertation, remarks that although ISS is widely used and popular, “it is both astonishing and troubling … that so few studies exist to support such wide-spread adoption” (p. 144).
In our review of the literature, we found that most objective studies were doctoral dissertations done in the last few years. Jackson wanted to know if the duration of ISS periods made any difference in classroom behavior, since the issue of students’ losing instructional time is important. He compared the effects of short, two-hour assignments to in-school suspension with two-day assignments on the classroom behavior of seven elementary students—four were randomly assigned to the two-hour period, and the other three assigned to the two-day. Results of the study revealed “virtually no difference” between the shorter or longer in-school suspension periods on affecting overall classroom behavior. A small decrease in classroom misbehavior and increase in classroom engagement following the suspension periods was observed, but this was not sustained. Jackson speculated that one reason is that the stimuli for the disruptive behavior within the classroom are likely not being addressed and the infrequent periods of in-school suspensions are not likely to have an effect.
Devlin (2006) sought to determine if a middle school’s ISS program using a non-punitive program with conflict resolution training was more effective than a program without conflict resolution training in reducing out-of-school suspensions and referrals. She found that in-school suspension with conflict resolution had no effect on reducing out-of-school suspension or referral rates. In fact, suspension rates went up during the pre- and post-test period for the treatment group. Devlin theorizes that conflict resolution training may be more effective with longer duration and embedded in the school’s culture rather than just “taught” during the in-school suspension period. She also thought that the students who had so many risk factors would need more sustained and layering of interventions to make a difference.
One large-scale study by Leapley (1997) does provide some promise for the benefits of ISS that uses a therapeutic program. He compared middle and high schools using an ISS program that was led by a behavior specialist who worked with students on academics, problem-solving skills, and conflict resolution, with control schools that had similar suspension rates. Not only did the ISS schools have lower suspension rates, but they also had a lower rate of violent incidents than did schools without the ISS program.
What should be considered when designing in-school suspension programs?
Although the research does not give much guidance on how to design effective ISS programs, or whether they should be used at all, it seems to be the preferred option over out-of-school suspension if schools have a goal of keeping students in school. If punitive programs are not working, investigating how to create ISS programs that are focused on helping students change their behavior and providing support for them as they do it, would be beneficial. Additionally, because students are referred to ISS programs for a variety of reasons, it is important to create a clear and consistent policy that outlines for both teachers and students why ISS exists and for why a student is placed there. As research does not indicate that longer periods of time in ISS make an impact, the time spent away from the classroom should be minimized if possible.
There are many other descriptions of ISS programs and opinions of how these programs should look. Our bibliography lists some of the most useful ones.
What are the most promising alternatives to suspension?
In conducting our literature review, we found a great article that sums up some promising alternatives to suspension. Author Dr. Reece L. Peterson is professor of special education at University of Nebraska, Lincoln. He specializes in the education of students with emotional or behavioral disorders, teaches undergraduate and graduate courses related to student behavior, and recently co-directed with Dr. Russell Skiba a federally funded project that developed a “Safe and Responsive Schools Framework” to improve student behavior and prevent school violence through three levels: Creating a Positive Climate, Early Identification and Intervention, and Effective Responses.
Ten Alternatives to Suspension (Peterson, 2005)
- Problem solving with students to identify alternative behavior choices, and following up with developing a contract that reminds students to engage in a problem-solving process and includes reinforcers for success and consequences for not following the contract
- Restitution. In-kind restitution, that is directly related to the problem caused by the student’s behavior (e.g., cleaning graffiti) or generally related to improving the school environment (e.g., garden maintenance)
- Mini-courses or skill modules. These should be on topics related to the inappropriate behavior and should facilitate behavior change
- Parent involvement/supervision. Better and more frequent communication about issues can be formalized into a disciplinary consequence
- Counseling to focus on problem solving related to behavioral issues
- Community service
- Behavior monitoring that permits rewards for successful performance of specified objectives
- Coordinated behavior plans
- Alternative programming such as permitting the student to make short- or long-term changes in their schedule, taking alternative classes, or participating in independent study or work-experience programs. Students should be able to receive appropriate credit and progress toward graduation
- Appropriate in-school suspension. To include academic tutoring, skill building related to the behavior problem, and a clearly defined procedure for students returning to class contingent on their progress or behavior
Dr. Peterson recommends that for these strategies to be most effective, certain “prerequisites” be in place that focus on creating a caring and positive school climate, building student-adult relationships, increasing family involvement, schoolwide character education, bullying prevention, and school discipline programs, consistent school values, early identification and intervention, and recognizing and rewarding positive behavior. Says Peterson, “If they are to be effective, these ’foundations’ must be implemented in a way that they become a normal part of that school’s culture.”
Used with permission from the author. To read his complete article, see http://ici.umn.edu/products/impact/182/default.html
Are schoolwide positive behavior supports a valid alternative?
“When student misbehaviors are treated as the problem, rather than a symptom of some deeper problem, the disciplinary measures fail more often than succeed (Chung & Paul, 1996, p. 7).
Many schools are realizing that the “get tough” approach is not working to address behavior issues, and are looking at schoolwide approaches to positive behavior support systems. They are also looking at how the school structures, culture, academics, and other factors contribute to disciplinary problems. Certainly there are many examples of promising programs. We chose to highlight one particular schoolwide support system that was developed in the Northwest.
Schoolwide positive behavior supports (SWBPS) is a promising approach to create a school environment that addresses problem behavior in a positive and preventive way, and provides a continuum of behavior support from schoolwide to classroom to individual (Sugai & Horner, 2006). Positive behavior supports are “rules, routines, and physical arrangements that are developed and taught by staff to prevent initial occurrences of the problem behavior” (from www.pbis.org/primaryprevention.htm). For example, if the school wishes to prevent running in the halls, they would set up a system where all students understood the rule and are taught not to run in the halls, have a routine where teachers are monitoring students in the hallway during key times, and have a teacher with students in large groups.
Key components of SWBPS typically include a planning team made up of school staff, administrators, parents, and others to guide the process; clearly defined definitions of schoolwide rules and expectations for behavior; appropriate behavioral expectations that are directly taught; established systems for recognizing appropriate behaviors and discouraging inappropriate behaviors; and systems for continual monitoring and evaluation (Lassen, Steele, & Sailor, 2006).
The intention for primary, or schoolwide, support is to teach students appropriate behavior, reinforce appropriate behavior to prevent it from happening, and create a culture where positive behavior is the norm. Of course, this does not work for everyone, for many reasons, and therefore more intensive systems of support are necessary.
Research on the effectiveness of SWPBS is growing with implementation evaluations showing positive results. A rural middle school saw a reduction in “discipline detentions” from the first to fourth year of implementation. A few studies of urban schools have “generally found reductions in the frequency of overall problem behavior” as measured primarily by office discipline referrals. One study found that the number of suspensions per student was significantly reduced each year during the three-year period (Lassen, Steele, & Sailor, 2006). Longitudinal evaluations also indicate that schools are sustaining SWPBS over a long period of time, are able to establish a training and coaching structure within a two-year initiative, and are not adding additional costs (Horner, Sugai, & Vincent, 2005).
For schools seeking assistance and more information about implementing SWPBS, the U.S. Department of Education, Office of Special Education Programs, established a national technical assistance center co-directed by the University of Oregon and University of Connecticut, to develop and enhance knowledge about school-based Positive Behavior Supports practices, systems and, outcomes. You can visit the Center’s Web site at www.pbis.org
Conclusion
“A school climate supportive of positive behavior, efforts to build positive interactions, appropriate instruction, and ongoing close supervision may prevent behavior problems from growing to crisis proportions and requiring disciplinary consequences” (Peterson, p. 11).
Although clear directions for solving disciplinary problems may not exist in the research, it is apparent that providing a continuum of supports that focuses on prevention, education, and positive feedback can be more effective in the long term than using primarily punitive, exclusionary measures that do not focus on why the student is having a behavioral problem.
When determining what kinds of practices to implement, whether it be a schoolwide behavior support program, a character-education focused ISS program, or a violence prevention program, Sugai and Horner (2006) recommend that innovative practices be pilot tested and evaluated frequently, and to base it on a “documented need.” They suggest applying the use of four evaluation questions: 1) What is the likelihood that the desired outcome will be achieved by this practice? 2) What are the costs and benefits of adopting and sustaining this practice? 3) Is the practice relevant to our situation and does it contextually fit with our culture? 4) What supports are necessary to ensure that the practice will be sustained over time?
It is clear that more research is needed to determine what factors specifically contribute to reducing disciplinary problems and how in-school suspension and alternatives can be most effective in doing this. Varying methods should be compared so that schools can make the right decisions when investing resources into a particular or schoolwide approach.
References and Additional Resources
Blomberg, N. (2004, Spring). Effective discipline for misbehavior: In school vs. out of school suspension. Concept, an Interdisciplinary Journal of Graduate Studies. Retrieved September 21, 2007 from www.publications.villanova.edu/Concept/2004/Effective_Discipline.htm
Chung, G., & Paul, R. (1996). School-wide discipline policies: In-school suspension in one middle school. (ERIC Document Retrieval No. ED420105)
Devlin, N.F. (2006). Effects of in-school suspension with conflict resolution training on suspensions, referrals, and normative beliefs about aggression. Doctoral dissertation, Walden University, 2006. University Dissertation Abstracts International, 67(1).
DeVoe, J.F., Peter, K., Kaufman, P., Miller, A., Noonan, M., Snyder, T.D., & Baum, K. (2004). Indicators of school crime and safety: 2004 (NCES 2005–002/NCJ 205290). Washington, DC: U.S. Government Printing Office. Retrieved September 17, 2007 from http://nces.ed.gov/pubs2005/2005002.pdf
Dinkes, R., Cataldi, E.F., Kena, G., & Baum, K. (2006). Indicators of school crime and safety: 2006 (NCES 2007–003/NCJ 214262). Washington, DC: U.S. Government Printing Office. Retrieved September 17, 2007, from http://nces.ed.gov/programs/crimeindicators/
Hammond, C. (with Linton, D., Smink, J., & Drew, S.). (2007). Dropout risk factors and exemplary programs: A technical report. Clemson, SC: National Dropout Prevention Center at Clemson University and Communities in Schools, Inc. Retrieved September 17, 2007, from www.dropoutprevention.org/resource/major_reports/communities_in_schools.htm
Horner, R., Sugai, G. & Vincent, C. (2005). Impact: Feature Issue on Fostering Success in School and Beyond for Students with Emotional/Behavioral Disorders, 18(20), 4–5.
Jackson, A.K. (2006). The effects of reduced duration in-school suspension on the classroom behavior of elementary school students. Doctoral dissertation, Florida State University, 2007. University Dissertation Abstracts International, 68(2).
Lassen, S.R., Steele, M.M., & Sailor, W. (2006). The relationship of school-wide Positive Behavior Support to academic achievement in an urban middle school. Psychology in the Schools, 43(6), 701–712.
Leapley, L.D. (1997). In-school suspension/intervention room: A means to reduce violence in schools. Doctoral dissertation, Eastern Michigan University, 1997. University Dissertation Abstracts International, 54(1), 127.
Palmisano, A., & Potts, K. (2004). Teaching to empty desks: The effects of truancy in Tennessee schools. Nashville, TN: The State of Tennessee. Retrieved September 21, 2007, from www.comptroller.state.tn.us/orea/reports/truancy.pdf
Peterson, R.L. (2005). Ten alternatives to suspension. Impact: Feature Issue on Fostering Success in School and Beyond for Students with Emotional/Behavioral Disorders, 18(20), 10–11. Retrieved September 17, 2007, from http://ici.umn.edu/products/impact/182/default.html
Skiba, R.J. (2000). Zero tolerance, zero evidence: An analysis of school disciplinary practice (Policy Research Report #SRS2). Bloomington, IN: Indiana Education Policy Center. Retrieved September 21, 2007, from www.unl.edu/srs/pdfs/ztze.pdf
Sugai, G., & Horner, R.R. (2006). A promising approach for expanding and sustaining school-wide positive behavior support. School Psychology Review, 35(2), 245–259.
Note: NWREL Planning and Service Coordination staff members have extensive expertise in policy analysis, literature review, and online information retrieval. For assistance in locating current educational research, policy, and statistics, contact info@nwrel.org or 503-275-0454. 
A Room for Character Building, not Punishment
Staff at Longfellow Elementary School (Great Falls, Montana) recognize that students often have difficulties in their lives that can affect their ability to learn at school. Rather than banishing them to a suspension room when they are having trouble managing their behavior, they are given a quiet, constructive place to “regroup” when the hustle and bustle of their classroom is too much for them. In this special room, a teacher certified in character education helps students work on specific character traits such as cooperation, kindness, peacefulness, respect, and responsibility. The children are filtered back to their regular classroom as soon as possible. Students know that there are academic expectations, but they also know that they are not there to be punished. “Given the living conditions of most of our students, I don’t know how we could punish them anymore,” says Principal Cal Gilbert. The mission statement of the school, posted on their Web site is “Learning, Whatever it Takes.” Learning the skills that children need to be successful in and out of school is certainly part of that mission. Since the character education classroom was implemented, disciplinary actions have decreased, says Gilbert.