TOC Electronic Journal: To print this article select pages 31-34.

Children’s Social Skills:

Assessment and
Classification Practices

STEPHEN N. ELLIOTT and FRANK M. GRESHAM

After defining social skills, the authors review recent advances in the The behavioral approach to defining social skills does not ensure
assessment and classification of children’s social skills. Social skills have that these social behaviors are, in fact, socially important or
been demonstrated to be critical prerequisites to academic and inter- socially significant. Merely increasing the frequency of certain
personal success. Therefore, knowledge of recent advances in the eval- behaviors that researchers define a priori as social skills may not
uation of social skills is essential to the functioning of counselors in affect goals or outcomes valued by society at large.
schools. The final and most heuristic definition of social skills is social
validity. According to this definition, social skills are those be-

T
he development of children’s social skills is critical both haviors exhibited in specific situations that help in predicting a
to children and to the educational enterprise, as can be child’s attitude on important social outcomes (Gresham & Elliott,
seen by the increasing educational and psychological re- in press). Important social outcomes for children and young
search literature on the development, assessment, and treat- people of school age include (a) acceptance by the peer group,
ment of children’s social behaviors. Some of the interest among (b) significant others’ judgments of social skills (e.g., teachers,
educators and psychologists in children’s social skills is probably parents), (c) academic competence, (d) adequate self-concept or
based on research findings demonstrating the relationship be- self-esteem, and (e) adequate psychological adjustment (i.e.,
tween academic achievement and social skills (e.g., Feldhusen, absence of psychopathology). This definition essentially is a
Thurston, & Benning, 1970; Kim, Anderson, & Bashaw, 1968) hybrid of the peer acceptance and behavioral definitions and is
and the growing belief that schools are a fertile environment for more comprehensive than either definition alone. Specific be-
developing socially competent individuals. haviors are identified and related to important criteria for in-
dexing adequate social functioning.
DEFINITION OF SOCIAL SKILLS
In general, social skills have been described as the interaction
between an individual and the environment and the tools used INAPPROPRIATE SOCIAL BEHAVIORS:
to initiate and maintain vital interpersonal relations (Phillips, TARGETS FOR ASSESSMENT
1978). Recent interest in social skills has focused primarily on A review of the literature indicated that social skills correlate
building positive social behaviors (i.e., social skills) into a child’s with overall adjustment and later functioning in society. Re-
repertoire. Several definitions of children’s social skills have search has revealed that children with deficient social skills have
been discussed in recent years. These definitions can be grouped a high incidence of dropping out of school (Ullman, 1957), bad
into three categories: (a) the peer acceptance definition, (b) the conduct discharges from military service (Roff, Sells, & Golden,
behavioral definition, and (c) the social validity definition. 1972), school maladjustment (Gronlund & Anderson, 1963), ju-
In the peer acceptance definition of social skills, a child is labeled venile delinquency (Roff et al., 1972), and adult mental health
as socially skilled if he or she is accepted by peers. Various difficulties, as evidenced by psychiatric referrals up to 13 years
sociometric indices of peer acceptance status have been used to later (Cowen, Pedersen, Babigian, Izzo, & Trost, 1973). Social
operationalize social acceptance (see McConnell & Odom, 1986, skills deficits have also been related to numerous problems in
for a review). The major disadvantage of this definition is that adjusting to the normal classroom environment (Stumme, Gres-
the specific behaviors that lead to a child being accepted or ham, & Scott, 1983). Socially unpopular children have been shown
rejected by peers cannot be identified. Thus, the peer acceptance to be lacking in various social skills, such as communicating
definition is of little use in designing remedial strategies for needs, cooperating, responding positively to peers, and making
social skills deficits. friends (Gottman, Gonso, & Rasmussen, 1975; LaGreca & San-
In the behavioral definition of social skills, socially skilled be- togrossi, 1980; Oden & Asher, 1977).
haviors are characterized as those behaviors exhibited in specific In summary, children who exhibit social skill deficits expe-
situations in which there is maximum probability of punishment rience both short- and long-term negative consequences, and
contingent on one’s social behavior. An advantage of this def- these negative consequences seem to be precursors of more
inition is the identification of specific social behaviors, their con- severe problems in adolescence and adulthood. Because of the
trolling variables (i.e., antecedents and consequences), and the paucity of formal measures of children’s (ages 5-18) prosocial
situations in which these behaviors are performed. Thus, the behaviors, psychologists and educators face some significant
behavioral definition of social skills has direct relevance for in- assessment and classification challenges. Thus, the purpose of
tervention strategies designed to remediate social skill deficits. this article is to examine research on the assessment and clas-

96 JOURNAL OF COUNSELING AND DEVELOPMENT / OCTOBER 1987 / VOL. 66

 Children’s Social Skills
sification of social skill problems and to outline pragmatic strat- in sociometric techniques keyed to nonbehavioral criteria, chil-
egies that will lead to treatment of children’s problems. dren’s attitudes or preferences for engaging in certain activities
with peers are assessed rather than their perceptions of peers’
METHODS OF ASSESSING behavior. In contrast, behavioral criteria are based on peers’
SOCIAL SKILLS perceptions of the specific behaviors of their peers. Peer nom-
Numerous assessment methods have been used to evaluate the inations and ratings focusing on behavioral criteria may be more
social skills of children and adolescents. Recent reviewers (Asher accurately termed behavioral sociometrics.
& Hymel, 1981; Foster & Ritchey, 1979; Gresham, 1985) have Self-report measures of children’s behavior are not as fre-
identified the following assessment procedures: (a) ratings by quently used as are teacher ratings or sociometric techniques,
others, (b) sociometric techniques, (c) self-report measures, (d) but they represent a potentially important source of assessment
behavioral role play measures, (e) behavioral interviews, and (f) information. The only self-report social skills rating scale of which
naturalistic observations. we are aware is How I Act Towards Others (HIATO) (Gresham &
Several of these assessment procedures generate information Elliott, 1986). This scale is a reworded version of the TROSS for
that helps in identifying and classifying social skills problems. children with at least a third-grade reading level. In preliminary
The assessment procedures used most frequently in generating research on the HIATO with both handicapped and nonhan-
information are ratings by others, sociometric techniques, self- dicapped children, the same factor structure and high internal
report measures, and behavioral role play tasks (Gresham, 1985; consistency demonstrated by the TROSS have been upheld.
Gresham & Elliott, 1984). Most of these procedures are very Some assessment techniques are better suited than are others
practical because of the ease of administration and the relatively for obtaining information that is useful in designing social skills
short time consumed. treatment programs. The extent to which an assessment tech-
Ratings provided by teachers, parents, and school counselors nique results in a functional analysis (i.e., an understanding of
have been a prime source of information in evaluating behavioral controlling antecedent events or contingencies) of behavior de-
problems of children and adolescents (Edelbrock, 1983). Re- termines its usefulness as an intervention-treatment assessment
cently, ratings provided by adults, particularly teachers, have method. Two techniques that permit a functional analysis, be-
been used to assess social skills. The most comprehensive teach- havioral interviews and naturalistic observations, are discussed
er rating scale is the Social Behavior Assessment (SBA), devel- below.
oped by Stephens (1978). The SBA contains 136 social skills on Behavioral interviews have not been investigated systemati-
which teachers rate children’s behavior on a 3-point scale (i.e., cally as a social skills assessment technique, although they are
acceptable level, less than acceptable level, or never). There are no perhaps the most frequently used form of assessment in the
reliability data, validity data, or norms in the handbook that initial stages of intervention (Gresham, 1983). Behavioral inter-
accompanies the test, and teachers find it very time consuming views are very useful in (a) defining social behaviors in observ-
to complete. In a recent review, Drabman (1985) concluded that able terms; (b) identifying the antecedent, sequential, and con-
“the SBA currently promises more than it offers” (p. 1410). sequent conditions surrounding target behaviors; and (c)
Several researchers (Gresham & Reschly, 1985; Stephens, 1980; designing observational systems to measure target behavior.
Stumme, Gresham, & Scott, 1982, 1983), however, have found Analyzing children’s behavior in natural settings (e.g., class-
it to be a stable and moderately reliable rating scale. rooms, playgrounds) is the mast ecologically valid method of
Gresham and Elliott (Clark, Gresham, & Elliott, 1985; Gres- assessing children’s social skills. Naturalistic observations per-
ham, Elliott, & Black, 1985, 1986) recently developed the Teacher mit a functional analysis of behavior in the setting and at the
Rating of Social Skills (TROSS) scale. The TROSS is a 50-item, time the behavior occurs. The decision of which behaviors to
4-factor (academic performance, social initiation, cooperation, be observed can be based on teacher ratings of frequency and
and peer reinforcement) rating scale on which teachers rate so- importance, and children can be observed through behavioral
cial behavior on a 3-point frequency dimension (often true, some- interviews. Reviews of commonly used social behaviors in ob-
times true, or never true). A unique feature of the TROSS is the servational research can be found in other sources (Asher &
inclusion of an important dimension on which teachers specify Hymel, 1981; Foster & Ritchey, 1979; Gresham, 1981b; Hops &
whether each behavior is critical for classroom success, impor- Greenwood, 1981).
tant for classroom success, or unimportant for classroom suc-
cess. Having teachers evaluate behaviors using the importance CLASSIFYING SOCIAL SKILL DIFFICULTIES
criterion ensures the social validity of the behaviors being as- Most authors agree that social incompetencies observed in chil-
sessed and has implications for selecting target behaviors for dren can result from difficulties in response acquisition or re-
intervention. For example, behaviors receiving a low frequency sponse performance (Bandura, 1977). Response acquisition def-
rating and a high importance rating should have top priority for icits occur when an individual has not learned skills that are
social skills interventions. necessary for a socially competent response; performance def-
Preliminary evidence on the TROSS indicates that it has a icits arise when a child fails to perform successfully behaviors
stable factor structure, adequate internal consistency (.97), and of which he or she is capable (Kratochwill & French, 1984).
efficiency (75% accuracy) in discriminating mainstreamed, mild- Gresham and Elliott (1984, in press) extended this two-way clas-
ly handicapped children from nonhandicapped children (Black, sification scheme to include four general areas of social skill
1985; Clark et al., 1985; Gresham et al., 1985, 1986). The TROSS problems. As indicated in Table 1, this model of social skill
includes information comparable to that found in the SBA but difficulties helps in determining whether a child knows how to
requires less time for teachers to complete and provides an im- perform the target skills and whether emotional arousal re-
portance dimension for enhancing the identification of target sponses (i.e., anxiety or impulsivity) are present.
behaviors. Social skill deficits. This social skill problem is characteristic of
Sociometric techniques are useful in determining which chil- children who have not acquired the necessary social skills to
dren are poorly accepted, rejected, or unpopular. Two basic interact appropriately with others or those who have failed to
types of sociometric procedures commonly used are peer nomi- learn a critical step in the performance of the skill. For example,
nations and peer ratings. Both can be keyed to nonbehavioral and a child may not know how to give a compliment to a peer. A
behavioral criteria. Nonbehavioral criteria are based on activities child’s knowledge or past performance of the appropriate be-
or attributes of children rather than specific behaviors. As such, havior is the basis for determining the existence of a social skill

JOURNAL OF COUNSELING AND DEVELOPMENT / OCTOBER 1987 / VOL. 66

Elliott and Gresham
TABLE 1
Conceptual Classification System for Children’s Social Skills

Deficit Domain
Emotional Arousal Response Acquisition Performance
Absent Social Skills Deficit Social Performance Deficit
Present Self-Control Skill Deficit Self-Control Performance Deficit
Note. Emotional arousal responses, which can be anxiety, fear, anger, or impulsivity, interfere with the acquisition or performance of
appropriate social behaviors.

deficit. Direct instruction, modeling, behavioral rehearsal, and appropriate social behavior are often used to ameliorate this
coaching are frequently used to remediate social skill deficits social skill problem (Bolstad & Johnson, 1972; Kendall & Bra-
(Gresham, 1982). swell, 1985; Rosenbaum & Drabman, 1979).
Social performance deficits. Children with social performance
deficits have appropriate social skills in their behavior reper- CONCLUSION
toires but fail to perform them at acceptable levels. For example, When social skills are deficient, children often miss rewarding
a child may have learned all the requisite steps in giving a interpersonal opportunities and are likely to experience negative
compliment to a peer but does not satisfactorily emit the re- consequences such as rejection. Therefore, the major goal of this
sponse in all social settings. Social performance deficits can be article has been to examine the various empirically proven as-
related to a lack of opportunity or to a lack of motivation to sessment methods that educators and psychologists can use to
perform the behavior. Determining whether a child can exhibit evaluate and classify social skill problems. We concluded that
the behavior in question is the key to identifying a social per- a multimethod approach to assessment in which various raters
formance deficit. Typically, social performance deficits have been (e.g., teachers, parents) use different methods (e.g., direct ob-
modified by manipulating antecedents and consequences. In- servation, rating scales, interviews) is desirable for selecting
terventions have included peer initiations (Strain, Shores, & children with skill difficulties, identifying specific skill deficits,
Timm, 1977), contingent social reinforcement (Allen, Hart, Har- and evaluating treatment outcomes. There are numerous treat-
ris, & Wolf, 1964), and group contingencies (Gamble & Strain, ment options available for children with social skills deficits.
1979). Several authors have conducted comprehensive reviews of school-
Self-control social skill deficits. This social skill problem is char- based interventions for social skill problems (e.g., see Cartledge
acteristic of a child for whom an emotional arousal response has & Milburn, 1980; Elliott, Gresham, & Heffer, in press; Goldstein,
prevented skill acquisition. Anxiety is an emotional arousal re- Sprafkin, Gershaw, & Klein, 1980; Michelson, Sugai, Wood, &
sponse that has been shown to prevent acquisition of appro- Kazdin, 1983). It seems clear that if the development of socially
priate coping behaviors, particularly with respect to fears and skilled individuals is a desired goal of the educational endeavor,
phobias (Bandura, 1977). Therefore, a child may not learn to then the significant adults in young children’s lives must know
interact effectively with others because social anxiety inhibits how to assess and classify social skill deficits so that effective
social approach behavior. Impulsivity (i.e., a tendency toward treatment strategies can be implemented.
short response latencies) is another emotional arousal response REFERENCES
that can hinder social skills acquisition (Kendall & Braswell, Allen, K.E., Hart, B.M., Harris, F.R., & Wolf, M.M. (1964). Effects of
1985). A child who exhibits impulsive behavior may fail to learn social reinforcement on isolate behavior of a nursery school child.
appropriate social interaction skills because his or her behavior Child Development, 35, 7-9.
Asher, S.R., & Hymel, S. (1981). Children’s social competence in peer
often results in rejection by peers (O’Leary & Dubey, 1979). In relations: Sociometric and behavioral assessment. In J.D. Wine & M.D.
turn, peer rejections lead to decreased exposure to appropriate Smye (Ed.), Social competence (pp. 125-157). New York: Guilford Press.
social models or to loss of appropriate social skills. Two impor- Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-
tant criteria signal the existence of a self-control social skill def- Hall.
icit: (a) the presence of an emotional arousal response and (b) Black, F. (1985). Social skills assessment for mainstreamed handicapped stu-
dents: The discriminative efficiency of the Teacher Ratings of Social Skills.
lack of knowledge or performance of the skill in question. In- Unpublished doctoral dissertation, Louisiana State University, Baton
terventions designed to remediate this problem usually involve Rouge.
emotional arousal reduction techniques (e.g., desensitization or Bolstad, O.D., & Johnson, S.M. (1977). The relationship between teach-
flooding) and self-control strategies (e.g., self-talk, self-moni- er’s assessment of students and students’ actual behavior in the class-
toring, and self-reinforcement) (Kendall & Braswell, 1985; Mei- room. Child Development, 48, 570-578.
chenbaum, 1977). Cartledge, G., & Milburn, J. (Eds.). (1980). Teaching social skills to children:
Innovative approaches. New York: Pergamon.
Self-control social performance deficits. Children with these def- Clark, L., Gresham, F.M., & Elliott, S.N. (1985). Development and val-
icits have a given social skill in their behavior repertoires, but idation of a social skills assessment measure: The TROSS-C. Journal
skill performance is hindered by an emotional arousal response of Psychoeducational Assessment, 4, 347-356.
and by problems in antecedent or consequent control. For ex- Cowen, E.L., Pedersen, A., Babigian, H., Izzo, L.D., & Trost, M.A.
ample, a child may know how to give a compliment to a peer (1973). Long-term follow-up of early detected vulnerable children.
Journal of Consulting and Clinical Psychology, 41, 438-446.
but does so infrequently or inconsistently because of social anx-
Drabman, R.S. (1985). Review of Social Behavior Assessment. In J.V.
iety and because he or she has limited opportunities to practice Mitchell, Jr. (Ed.), The ninth mental measurements yearbook. Lincoln, NE:
the behavior or has not been contingently reinforced for doing Buros Institute.
so. Identification of a self-control social performance deficit rests Edelbrock, C. (1983). Problems and issues in using rating scales to assess
on two criteria: (a) the presence of an emotional arousal response child personality and psychopathology. School Psychology Review, 12,
and (b) inconsistent performance of the social skill in question. 293-299.
Elliott, S.N., Gresham, F.M., & Heffer, R.W. (in press). Social skills
Self-control strategies for teaching inhibition of inappropriate interventions: Research findings and training techniques. In C.A. Maher
behavior, stimulus control training for teaching discrimination & J.E. Zins (Eds.), Psychoeducational interventions in schools: Methods and
skills, and contingent reinforcement for increasing displays of procedures for enhancing student competence. New York: Pergamon Press.

98 JOURNAL OF COUNSELING AND DEVELOPMENT / OCTOBER 1987 / VOL. 66

 Children’s Social Skills
Feldhusen, J.F., Thurston, J.R., & Benning, J.J. (1970). A longitudinal Kendall, P.C., & Braswell, L. (1985). Cognitive-behavioral therapy for im-
analysis of classroom behavior and school achievement. Journal of pulsive children. New York: Guilford.
Experimental Education, 38, 4-10. Kim, Y., Anderson, H.E., & Bashaw, W.L. (1968). Social maturity
Foster, S.L., & Ritchey, W.L. (1979). Issues in the assessment of social achievement and basic ability. Educational and Psychological Measure-
competence in children. Journal of Applied Behavior Analysis, 12, 625- ment, 28, 535-543.
638. Kratochwill, T.R., & French, D.C. (1984). Social skills training for with-
Gamble, R., & Strain, P.S. (1979). The effects of dependent and inter- drawn children. School Psychology Review, 13, 331-338.
dependent group contingencies on socially appropriate responses in LaGreca, A.M., & Santogrossi, D.A. (1980). Social skills training with
classes for emotionally handicapped children. Psychology in the Schools, elementary school students: A behavioral group approach. Journal of
16, 253-260. Consulting and Clinical Psychology, 48, 220-227.
Goldstein, A.P., Sprafkin, R.P., Gershaw, N.J., & Klein, P. (1980). Skill- McConnell, S.R., & Odom, S.L. (1986). Sociometrics: Peer-referenced
streaming the adolescent. Champaign, IL: Research Press. measures and the assessment of social competence. In P.S. Stram,
Gottman, J.M., Gonso, J., & Rasmussen, B. (1975). Social interaction, M.J. Guralnick, & H.M. Walker (Eds.), Children’s social behavior: De-
social competence, and friendship in children. Child Development, 46, velopment, assessment, and modification (pp. 181-209). Orlando, FL: Ac-
708-718. ademic Press.
Gresham, F.M. (1981a). Assessment of children’s social skills. Journal of Meichenbaum, D. (1977). Cognitive behavior modification. New York: Pen-
School Psychology, 27, 120-133. um.
Gresham, F.M. (1981b). Social skills training with handicapped children: Michelson, L., Sugai, D.P., Wood, R.P., & Kazdin A.E. (1983). Social
A review. Review of Educational Research, 51, 139-176. skills assessment and training with children: An empirically based approach.
Gresham, F.M. (1982). Social skills: Principles, practices, and procedures. New York: Plenum.
Des Moines: Iowa Department of Public Instruction. Oden, S.L., & Asher, S.R. (1977). Coaching children in social skills for
Gresham, F.M. (1983). Behavioral interviews in school psychology: Is- friendship making. Child Development, 48, 496-506.
sues in psychometric adequacy and training. School Psychology Review, O’Leary, S.G., &Dubey, D.R. (1979). Applications of self-control pro-
12, 17-25. cedures for children: A review. Journal of Applied Behavior Analysis, 12,
Gresham, F.M. (1985). Conceptual issues in the assessment of social 449-465.
competence in children. In P. Strain, M. Gurolnick, & H. Walker Phillips, E.L. (1978). The social skills basis of psychopathology. New York:
(Eds.), Children’s social behavior: Development, assessment, and modifica- Grune & Stratton.
tion (pp. 143-179). New York: Academic Press. Roff, M., Sells, S.B., & Golden, M.M. (1972). Social adjustment and per-
Gresham, F.M., & Elliott, S.N. (1984). Assessment and classification of sonality development in children. Minneapolis: Minneapolis Press.
children’s social skills: A review of methods and issues. School Psy- Rosenbaum, M.S., & Drabman, R.S. (1979). Self-control training in the
chology Review, 13, 292-301. classroom: A review and critique. Journal of Applied Behavior Analysis,
Gresham, F.M., & Elliott, S.N. (1986). How I Act Towards Others. Un- 12, 467-485.
published test, Louisiana State University, Baton Rouge. Stephens, T.M. (1978). Social skills in the classroom. Columbus, OH: Ce-
dars Press.
Gresham, F.M., & Elliott, S.N. (in press). The relationship between
Stephens, T.M. (1980). Technical manual for the Social Behavior Assessment.
adaptive behavior and social skills: Issues of definition and assess-
Columbus, OH: Cedars Press.
ment. Journal of Special Education.
Strain, P.S., Shores, R.E., & Timm, M.A. (1977). Effects of peer social
Gresham, F.M., Elliott, S.N., & Black, F.L. (1985). Factor structure rep- initiations on the behavior of withdrawn preschool children. Journal
lication and bias investigation of the Teacher Rating of Social Skills. Manu- of Applied Behavior Analysis, 10, 289-298.
script submitted for publication. Stumme, V.A., Gresham, F.M., & Scott, N.A. (1982). Validity of Social
Gresham, F.M., Elliott, S.N., & Black, F.L. (1986). Social skills of children Behavior Assessment in discriminating emotionally disabled from
from different diagnostic subgroups. Manuscript submitted for publica- nonhandicapped students. Journal of Behavioral Assessment, 4, 327-342.
tion. Stumme, V.A., Gresham, F.M., & Scott, N.A. (1983). Dimensions of
Gresham, F.M., & Reschly, D.J. (1985). Social skills and peer acceptance children’s classroom social behavior: A factor analytic investigation.
differences between learning disabled and nonhandicapped students. Journal of Behavioral Assessment, 5, 161-177.
Learning Disability Quarterly, 10, 65-81. Ullman, C.A. (1957). Teachers, peers, and tests as predictors of adjust-
Gronlund, H., & Anderson, L. (1963). Personality characteristics of so- ment. Journal of Educational Psychology, 48, 257-267.
cially accepted, socially neglected, and socially rejected junior high
school pupils. In J. Sederman (Ed.), Educating for mental health. New Stephen N. Elliott is an assistant professor, Department of Educational Psy-
York: Crowell. chology, University of Wisconsin—Madison. Frank M. Gresham is a profes-
Hops, H., & Greenwood, C.R. (1981). Social skills deficits. In E.J. Mash sor, Department of Psychology, Louisiana State University, Baton Rouge. Cor-
& L.G. Terdal (Eds.), Behavioral assessment of childhood disorders (pp. respondence regarding this article should be sent to Stephen N. Elliott, Department
347-394). New York: Guilford. of Psychology, Louisiana State University, Baton Rouge, LA 70803-5501.

JOURNAL OF COUNSELING AND DEVELOPMENT / OCTOBER 1987 / VOL. 66 99