Handbook of psychology volume 7 educational psychology

The Content of School Interventions for Students with Behavior Disorders

523

It is instructive that they found only 34 evidence-based pro-

grams that met rigorous standards of efficacy. Their report

provides the following valuable information for professional

consumers:

1. It identifies critical issues and themes in prevention re-

search with school-age children and families. 

2. It profiles universal, selected, and indicated programs that

reduce symptoms of both externalizing and internalizing

symptoms.

3. It summarizes state-of-the-art programs in the prevention

of mental disorders in school-age children.

4. It identifies the key elements that contribute to program

success.

5. It provides suggestions to improve the quality of program

development and evaluation.

Aside from its overall quality, there are a number of spe-

cific strengths to the report of Greenberg et al. (1999). Sepa-

rating the review into universal, selected, and indicated

interventions and then determining whether the interventions

reviewed address primarily externalizing or internalizing

behavior problems and disorders substantially increases the

relevance of the review for school applications and provides

needed clarity for consumers. In our view, the great majority

of adjustment problems that students manifest in the school

setting are either directed outwardly toward the external

social environment (i.e., aggression, defiance, bullying, coer-

cion) or internally (i.e., social withdrawal, anxiety, depres-

sion, phobias) representing problems with others versus

problems with self. Externalizing disorders usually require a

reduce-and-replace intervention strategy, whereas internaliz-

ing disorders typically call for a focus on skill development

and performance enhancement. 

Greenberg et al. (1999) found that the most effective

interventions were those that (a) had multiple components,

(b) involved multiple social agents (parents, teachers, peers),

(c) were implemented across several settings (classroom,

playground, home), and (d) were in place for a sufficient pe-

riod of time to register socially valid outcomes—usually a

minimum of 1 year. This review is being continuously up-

dated and expanded as new evidence-based interventions

come on line. It is highly recommended as a blueprint or

roadmap for BD professionals to use in upgrading school-

based practices for the behaviorally at-risk K–12 population.

A special issue of the APA journal Prevention and Treatment

(March, 2001) was devoted to the report and its findings.

The U.S. Public Health Service classification system for

differing types of prevention is well suited for the delivery of

these three types of interventions profiled in the Greenberg

et al. (1999) review (i.e., universal, selected, indicated). As a

rule, universal interventions are used to achieve primary pre-

vention goals and outcomes (i.e., to prevent harm); selected

interventions are used for secondary prevention efforts (i.e., to

reverse harm); and indicated interventions are used for ter-

tiary prevention applications (i.e., to reduce harm). Walker

et al. (1995, 1996) have adapted this classification schema

for the delivery of proven interventions within school settings

that address the needs of all students. Figure 20.3 illustrates

this classification schema. School settings are ideally suited to

implement this delivery structure because they are naturally

organized to implement schoolwide interventions (e.g., a

school discipline plan, a school safety plan, a school improve-

ment plan), small-group interventions (e.g., resource and self-

contained classrooms), and individually tailored interventions

(e.g., counseling). In the last 5 years, this three-level interven-

tion delivery system has been widely adopted by researchers

and school personnel alike across the country.

School personnel are especially amenable to universal in-

tervention approaches because they treat all students equi-

tably and in the same manner. Thus, the fairness issue that

resonates so strongly with most teachers is indirectly ad-

dressed as every student is exposed to the intervention in an

identical fashion. Those students for whom the universal in-

tervention is insufficient then receive secondary and possibly

tertiary prevention interventions. One of the great advantages

of a universal intervention is that it creates a context in which

more intensive small-group and individually tailored inter-

ventions can achieve greater effectiveness, which are then

applied only after the failure of a universal intervention ap-

proach for certain students. Another is that it addresses

the problems of mildly involved, at-risk students in a cost-

effective manner. The scaled-up adoption of this integrated

delivery system, when combined with proven intervention

models that have been adapted to and tested within the school

setting, has the potential to improve substantially the effec-

tiveness of schooling and to create much more positive school

climates.

THE CONTENT OF SCHOOL INTERVENTIONS

FOR STUDENTS WITH BEHAVIOR DISORDERS

When children begin their school careers, they are required to

make two critically important social-behavioral adjustments

teacher-related and peer-related (Walker et al., 1995). That is,

they must negotiate a satisfactory adjustment to the academic

and behavioral expectations of teachers and conform to the de-

mands of instructional settings. Of equal importance, they

must negotiate a satisfactory adjustment to the peer group, find

a niche within it, and develop social support networks consist-

ing of friends, affiliates and acquaintances. Walker, Irvin,

524

School-Related Behavior Disorders

Noell, and Singer (1992) have developed an interpersonal

model of social-behavioral competence for school settings.

This model identifies the adaptive and maladaptive behavioral

correlates of successful student adjustment in the domains of

teacher-related and peer-related functioning. The model also

describes the long-term outcomes that are commonly asso-

ciated with the adaptive (e.g., school success, friendship-

making, peer and teacher acceptance) versus maladaptive

(e.g., school failure and dropout, assignment to restrictive set-

tings, delinquency) pathways contained within it.

The adaptive and maladaptive behavioral correlates in-

cluded in the teacher- and peer-related adjustment dimensions

of this model are based on empirical evidence generated by

the present authors and their colleagues as well as research ev-

idence presented in the professional literature on social com-

petence. The long-term outcomes listed for each path under

these two forms of adjustment are based on longitudinal and

cross-sectional studies reported in the literature over the past

two decades (see Loeber & Farrington, 1998; Patterson et al.,

1992; Strain, Guralnick, & Walker, 1986).

Failure in either of these critically important areas impairs a

student’s overall school adjustment and success; failure in both

puts a student’s overall quality of life at risk and is a harbinger

of future problems of potentially severe magnitude. Students

with behavior disorders are invariably below normative levels

and expectations on the adaptive behavioral correlates of

teacher- and peer-related adjustment and usually outside the

normative range on the maladaptive behavioral correlates. In

the great majority of cases, the intervention of choice for stu-

dents with behavior disorders involves developing their social

skills and overall social competence while teaching them alter-

natives to the maladaptive forms of behavior that tend to dom-

inate their behavioral repertoires. In our view, the potential of

social skills instruction (SSI) for students in general, and par-

ticularly for students with behavior disorders, has yet to be

realized in spite of a substantial investment in SSI efforts over

the past two decades by school personnel (Bullis, Walker, &

Sprague, 2001; Elksnin & Elksnin, 1995).

Recent reviews of the efficacy of SSI with the K–12 school

population with behavior disorders have not been encourag-

ing (see Gresham, 1997, 1998a; Kavale, Mathur, Forness,

Rutherford, & Quinn, 1997). These authors have conducted

and reviewed meta-analyses of social skills interventions and

concluded that the average effect sizes in the studies they

reviewed are minimal to moderate at best, generally ranging

between .30 and .45. Given the level of effort invested in these

studies, these results do not appear to be terribly cost-effective.

A number of reasons have been hypothesized for these disap-

pointing outcomes, including (a) the failure to match deficits

in social skills with the intervention, (b) the absence of

theoretical models to guide SSI, (c) implementing the SSI

procedure in artificial instructional settings and expecting

generalization of newly taught skills to natural settings, and

(d) implementing the SSI for insufficient amounts of time for

it to impact the student’s behavioral repertoire. We believe that

an equally powerful, but infrequently mentioned, reason con-

cerns the failure to address the competing, maladaptive behav-

ior problems of students with behavior disorders who are the

targets of SSI. SSI alone is rarely sufficient to teach prosocial

skills and simultaneously to address a well-developed mal-

adaptive behavioral repertoire. Direct intervention techniques

designed to reduce and eliminate maladaptive forms of behav-

ior are required for this purpose. Some best-practice principles

and guidelines for conducting SSI with students with behavior

disorders are described next.

Social Skills Instruction for Students with

Behavior Disorders

The school is an ideal setting for teaching social skills be-

cause of its accessibility to children and their peers, teachers,

and parents. Fundamentally, social skills intervention takes

place in school and home settings, both informally and for-

mally, using either universal or selected intervention proce-

dures. Informal social skills interventions are based on the

notion of incidental learning, which takes advantage of

naturally occurring behavioral incidents or events to teach

appropriate social behavior. Most of the SSI in home, non-

classroom school contexts, and community settings can be

characterized as informal or incidental. Literally thousands

of behavioral incidents occur in these naturalistic home,

school, and community settings, creating rich opportunities

for making each of these behavioral incidents a potentially

successful learning experience. Formal SSI, on the other

hand, can take place seamlessly within a classroom setting in

which (a) the social skills curriculum is exposed to the entire

class or it is taught to selected students within small-group

formats and (b) social skills are taught as subject matter in the

same way as are social science, history, biology, and other

academic subjects. However, unless formal and informal

methods of teaching social skills are combined with each

other, there is likely to be a disconnect between conceptual

mastery of social skills and their demonstration and applica-

tion within natural settings.

Objectives of Social Skills Instruction

(From Kathleen L. Lane, Frank M. Gresham, & Tam E.

O’Shaughnessy,  Interventions for Children with or At Risk

for Emotional and Behavioral Disorders. Published by Allyn

and Bacon, Boston, MA. Copyright © 2002 by Pearson

Education. Reprinted by permission of the publisher.)

The Content of School Interventions for Students with Behavior Disorders

525

deficits or competing problem behaviors that the student ex-

hibits. A common misconception is that one seeks to facilitate

generalization and maintenance after implementing proce-

dures for the acquisition and performance of social skills. The

evidence is clear that the best and preferred practice is to in-

corporate generalization strategies from the beginning of any

SSI program (Gresham, 1998b). 

Promoting Skills Acquisition

Procedures designed to promote skill acquisition are applica-

ble when students do not have a particular social skill in their

repertoire, when they do not know a particular step in the per-

formance of a behavioral sequence, or when their execution

of the skill is awkward or ineffective (i.e., a fluency deficit).

It should be noted that a relatively small percentage of stu-

dents would need social skills intervention based on acquisi-

tion deficits; far more students have performance deficits

(Gresham, 1998a). 

Three procedures represent pathways to remediating

deficits in social skill acquisition: modeling, coaching, and

behavioral rehearsal. Social problem solving is another path-

way, but it is not discussed here because of space limitations

and because it incorporates a combination of modeling,

coaching, and behavioral rehearsal. More specific informa-

tion on social problem solving interventions can be found in

Elias and Clabby (1992). 

Modeling is the process of learning a behavior by observ-

ing another person performing it. Modeling instruction pre-

sents the entire sequence of behaviors involved in a particular

social skill and teaches the student how to integrate specific

behaviors into a composite behavior pattern. Modeling is one

of the most effective and efficient ways of teaching social be-

havior (Elliott & Gresham, 1992; Schneider, 1992). 

Coaching is the use of verbal instruction to teach social

skills. Unlike modeling, which emphasizes visual displays

of social skills, coaching utilizes a student’s receptive language

skills. Coaching is accomplished in three fundamental

steps: (a) presenting social concepts or rules, (b) providing

opportunities for practice or rehearsal, and (c) providing

specific informational feedback on the quality of behavioral

performances.

Behavioral rehearsal refers to practicing a newly learned

behavior in a structured, protective situation of role-playing.

In this way, students can enhance their proficiency in using

social skills without experiencing adverse consequences.

Behavioral rehearsal can be covert, verbal, or overt. Covert

rehearsal involves students’ imagining certain social interac-

tions (e.g., being teased by another student or group of stu-

dents). Verbal rehearsal involves students’ verbalizing the

specific behaviors that they would exhibit in a social situation.

SSI has four primary objectives: (a) promoting skill acqui-

sition, (b) enhancing skill performance, (c) reducing or elimi-

nating competing problem behaviors, and (d) facilitating

generalization and maintenance of social skills. Most students

with behavior disorders will likely have some combination of

acquisition and performance deficits, some of which may be

accompanied by competing problem behaviors. Any given

student may require some combination of acquisition, perfor-

mance, and behavior-reduction strategies. All students will re-

quire procedures to facilitate generalization and maintenance

of previously learned social skills (see Gresham, 2002).

Table 20.1 lists specific social skills and behavior-reduc-

tion strategies for each of the four goals of SSI. Appropriate

intervention strategies should be matched with the particular

TABLE 20.1

Objectives and Strategies of Social Skills Instruction

I. PROMOTING SKILLS ACQUISITION

A. Modeling.

B. Coaching.

C. Behavioral rehearsal.

II. ENHANCING SKILLS PERFORMANCE

A. Manipulation of antecedents.

1. Peer initiation strategies.

2. Proactive classroom management strategies.

3. Peer tutoring.

4. Incidental teaching.

B. Manipulation of consequences.

1. Contingency contracting.

2. Group-oriented contingency systems.

3. School-home notes.

4. Verbal praise.

5. Activity reinforcers.

6. Token and point systems.

III. REMOVING COMPETING PROBLEM BEHAVIORS

A. Differential reinforcement.

1. Differential reinforcement of other behavior (DRO).

2. Differential reinforcement of low rates of behavior (DRL).

3. Differential reinforcement of incompatible behaviors (DRI).

B. Overcorrection.

1. Restitution.

2. Positive practice.

C. Time-out.

1. Nonexclusionary (contingent observation).

2. Exclusionary.

D. Systematic desensitization (for anxiety-based competing behaviors).

E. Flooding and exposure (for anxiety-based competing behaviors).

IV. FACILITATING GENERALIZATION

A. Topographical generalization.

1. Training diversely.

2. Exploiting functional contingencies.

3. Incorporating functional mediators.

B. Functional Generalization.

1. Identify strong competing stimuli in specific situations.

2. Identify strong competing problem behaviors in specific

situations.

3. Identify functionally equivalent socially skilled behaviors.

4. Increase reliability and efficiency of social skilled behaviors

(build fluency).

5. Decrease reliability and efficiency of competing problem

behaviors.

Source: Gresham (1998b).

526

School-Related Behavior Disorders

Overt rehearsal is the actual role-playing of a specific social

interaction.

Enhancing Skills Performance

Most social skills interventions involve procedures that in-

crease the frequency of particular prosocial behaviors in spe-

cific social situations because most social skills difficulties

involve performance deficits rather than acquisition deficits.

This suggests that social skills interventions for most stu-

dents should take place in naturalistic environments (e.g.,

classrooms, playgrounds) rather than in small, pullout-group

situations. Failure to perform certain social skills in specific

situations results from two fundamental factors: (a) inappro-

priately arranged antecedents and (b) inappropriately

arranged consequences. A number of specific procedures can

be classified under the broad rubric of antecedent and conse-

quent strategies. 

Interventions based on antecedent control assume that the

environment does not set the occasion for the performance of

prosocial behavior. That is, cues, prompts, or other events ei-

ther are not present or are not salient in order for the child to

discriminate these stimuli in relation to the performance of

prosocial behavior. A cuing and prompting procedure uses

verbal and nonverbal cues or prompts to facilitate prosocial

behavior. Simple prompts or cues for some children may be all

that is needed to signal them to engage in socially appropriate

behavior (e.g., “Say thank you,” “Ask Katrina to join your

group”). Cuing and prompting represent one of the easiest and

most efficient social skills intervention strategies (Elliott &

Gresham, 1992; Walker et al., 1995).

Interventions based on consequent control can be classi-

fied into three broad categories: (a) reinforcement-based

strategies, (b) behavioral contracts, and (c) school-home

notes. Reinforcement-based strategies assume that the student

knows how to perform a social skill but is not doing so be-

cause of little or no reinforcement for the behavior. The objec-

tive in using these strategies is to increase the frequency of

reinforcement for prosocial behavior. Reinforcement strate-

gies include attention, social praise, tokens and points, and

activity reinforcers as well as group-oriented contingency

systems. Extensive discussions of behavioral contracts,

school-home notes, and group-oriented contingency systems

can be found in more comprehensive treatments of these sub-

jects (Kelley, 1990; Kohler & Strain, 1990; Stuart, 1971)

Removing or Eliminating Competing Problem Behaviors

The focus of SSI is clearly on developing and refining proso-

cial behaviors. However, the failure of some students to either

acquire or perform certain social skills may be due to the pres-

ence of competing problem behaviors. This is particularly true

of students having behavior disorders whose externalizing or

internalizing symptoms compete with or block the acquisition

and performance of prosocial behaviors. For example, aggres-

sive behavior may be performed instead of a prosocial behav-

ior because it may be more efficient and reliable in achieving

one’s social goals and producing reinforcement. A number of

techniques that are effective in reducing competing problem

behaviors are presented in Table 20.1.

Facilitating Generalization and Maintenance

Basically, there are only two processes that are essential to all

behavioral interventions: discrimination and generalization

(Stokes, 1992). Discrimination occurs within the context of

stimulus control. A major problem confronting social skills in-

terventions is that it is much easier to prompt the occurrence of

some behaviors in one place, for a limited period of time, than

it is to get those same behaviors to occur in a variety of other

places for an extended period of time. That is, it is infinitely

easier to teach discriminations than it is to teach generaliza-

tion and maintenance.

Generalization of behavior change is related directly to the

principle of resistance to intervention. If social skill deficits

occur at low frequencies, competing problem-behavior ex-

cesses will likely occur at high frequencies, and both of these

deficits and excesses tend to be chronic with students with be-

havior disorders (i.e., they have lasted a relatively long period

of time), and they will tend to show less generalization across

different nontraining conditions as well as less durability over

time as SSI is withdrawn (Gresham, 1991). In effect, these

students quickly discriminate training from nontraining

conditions, particularly when training conditions are notice-

ably different from nontraining conditions.

Students with behavior disorders often show excellent ini-

tial behavior change in response to well-designed, powerful

school interventions of a secondary or tertiary prevention

nature, particularly in relation to their competing problem-

behavior excesses, but they tend not to show generalization or

maintenance of these behavior changes. One reason for this

may be that exclusive attention often is focused on decreasing

the momentum of undesirable behavior to the exclusion of

facilitating the momentum of desirable behaviors such as crit-

ically important social skills. The primary reason for this

frequently observed lack of generalization and maintenance is

that essential components of behavior change are not actively

programmed to occur as part of SSI.

Various generalization programming strategies are pre-

sented in Table 20.1 under the headings of topographical and