10.177/0145445503259828 SKILLS ASSESSMENT Bielecki, Swender / SOCIAL BEHAVIOR MODIFICATION / September 2004 The Assessment of Social Functioning in Individuals With Mental Retardation A Review JOANNE BIELECKI Louisiana State University STEPHEN L. SWENDER Pinecrest Developmental Center Social skills deficits and excesses are a defining aspect of mental retardation (MR). Research indicates that there is an established relationship between social skills and maladaptive behav- iors. A number of studies demonstrate that the social competence of individuals with MR and comorbid psychopathology can be enhanced with social skills training. However, to design an effective training package, an accurate assessment of adaptive and social functioning must first be conducted. Unique problems arise when assessing social skills in individuals with severe and profound MR (i.e., individuals often have limited verbal repertoires). Thus, a clinician must often rely on observable behavior and caregiver report rather than self-report. The three most common methods for assessing social skills are behavioral observations, role-playing, and checklists. These assessment strategies will be discussed, as well as suggestions for future research. Keywords: social skills; mental retardation; developmental disability; adaptive behavior scales; behavior checklists A universally accepted definition of social skills does not exist but major themes are reflected in the literature. Bellack (1983) defined social skills as observable and measurable interpersonal behaviors that promote independence, social acceptability, and quality of life. These skills are crucial to adjustment and normal functioning. They enable a person to obtain healthy social relationships, avoid interper- sonal conflict, and cope with stressful situations (Guralnick, 1986; Matson & Swiezy, 1994). Social skills are learned behaviors that are BEHAVIOR MODIFICATION, Vol. 28 No. 5, September 2004 694-708 DOI: 10.177/0145445503259828 © 2004 Sage Publications 694 Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 695 both situation-specific and context-dependent. Context is important because different social skills are important in different situations and for different tasks (La Greca, Stone, & Bell, 1982). Matson and Ollendick (1988) indicated that a socially skilled person adapts well to various situations because he or she is able to recognize social cues and appropriately respond to them. These individuals exhibit behav- iors that are reinforced and refrain from exhibiting behaviors that are punished (Libet & Lewinsohn, 1973). Individuals with developmental disabilities are generally deficient across a broad spectrum of social behaviors. As a result, mental retar- dation (MR) has become associated with social skill deficits (i.e., poor eye contact) and/or social excesses (i.e., holding onto others and not letting go); Duncan, Matson, Bamburg, Cherry, & Buckley, 1999). A positive correlation has been found between severity of MR and social impairment, with deficits greatest among those diagnosed with severe and profound MR (Wing & Gould, 1979). In persons with autistic dis- order, the social skill deficits appear to be even more pronounced than in people with other developmental disorders (Njardvik, Matson, & Cherry, 1999), and specific mental disorders such as psychosis are well known for their relationship to social behavior (Nihira, Price- Williams, & White, 1988). Matson, Smiroldo, and Bamburg (1998) investigated the relationship between psychopathology and social skills in 846 individuals with severe and profound MR. Their linear regression analysis indicated that increases in symptoms of psychopathology predicted increases in negative social behaviors. In addition, social impairments have been closely linked to behavior problems such as aggression and self-injury (Duncan et al., 1999). The results of the Duncan et al. (1999) study indicated that individuals displaying maladaptive behaviors exhibited a restricted range of social behaviors compared to those in the control group. Research demonstrates that there is an established relationship between social skills and maladaptive behaviors, but because the data is correlational, it is unclear if social skill deficits result in problem behaviors or if the presence of maladaptive behaviors results in social impairments. Social skills training procedures have been applied to enhance the social competence of persons with schizophrenia, perva- sive developmental disorders, social anxiety, depression, hearing and 696 BEHAVIOR MODIFICATION / September 2004 visual impairments, and MR (Coe, Matson, & Fee, 1990; Matson et al., 1988; Matson, Fee, Coe, & Smith, 1991; Matson, Zeiss, Zeiss, & Bowman, 1980; Raymond & Matson, 1989). However, effective treat- ments are unlikely unless clinicians and researchers can reliably and validly assess social skills in the mentally retarded population (Bellack, 1983). Accurate assessment of adaptive and social skills is important because identification of common deficits may lead to training packages tailored to the special needs of persons with specific diagnoses and/or maladaptive behaviors. Such programs may dramat- ically affect the lives of persons with severe and profound MR and lead to more successful community integration. ASSESSMENT Unique problems arise when assessing social skills in individuals with severe and profound MR. These individuals often have very lim- ited verbal repertoires, which prevent the use of self-report assess- ment techniques. Persons with MR may have difficulties in reliable self-reporting even if the retardation is mild (Reiss, 1994). For exam- ple, researchers have demonstrated that persons with MR often look for outside approval (Zigler & Burack, 1989) and may readily agree with an interviewer’s leading questions. For these reasons, a clinician must often rely on observable behavior and caregiver report rather than self-report. The three most common methods for assessing social skills are behavioral observations, role-playing, and checklists (Bellack, 1979). BEHAVIORAL OBSERVATIONS Observation techniques can be conducted in either naturalistic or analogue conditions. The most direct form of assessment is the obser- vation of behavior within the actual environment (Shapiro & Browder, 1990). Naturalistic observation involves several raters who assess the individual on discrete social responses determined prior to the assessment (i.e., eye contact, use of gestures). Observing the indi- vidual in a naturalistic setting is often preferred because it allows cli- Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 697 nicians to sample behavior in the actual environment in which it occurs, surrounded by the people typically interacting with the indi- vidual and affecting relevant contingencies. Because this type of assessment occurs in the natural environment, results are usually more generalizable across setting and time (Gettinger & Kratochwill, 1987). There are some problems with this procedure. First, it is both costly and time-consuming, especially if the target behavior occurs infrequently. Second, the presence of an observer can cause reactivity with participants who may not exhibit their normal behavior. When observation of an individual in the natural setting is not possible, analogue settings may be used. ROLE-PLAYING Analogue observation often involves presentation of a role-play or simulated situation to which the individual responds. Typically, the experimenter provides a description of a scene and the individual acts out how he or she would respond in the situation. Role-playing allows the clinician to evaluate normally low-frequency behavior in a cost- effective manner. Several techniques have been developed to assess the interpersonal problem-solving skills of individuals with MR. They include the Behavioral Social Skills Assessment (BSSA) (Cas- tles & Glass, 1986), the Social Problem Solving Test (SPST) (Castles & Glass, 1986), and the Means-End Problem Solving Procedure (MEPS) (Platt & Spivack, 1975). The BSSA consists of 12 problem vignettes that were developed for use with role-play scenarios. The participant is read a description of the problem situation and shown an accompanying videotaped enactment. The individual is then asked to respond to the TV actors just as he or she would if the scenario were actually occurring. The measure has high interrater reliability (r = .93) but low test-retest reli- ability (r = .70). The MEPS and SPST also consist of problem vignettes but they are written in an open-middle format. Rather than working out the solution to a problem, the participant is given a prob- lem situation and its resolution and is then asked to provide a story that will connect the beginning with the end. Several probes are given to encourage the generation of different solutions. Responses are 698 BEHAVIOR MODIFICATION / September 2004 audiotaped and later transcribed for scoring. Researchers have dem- onstrated the scales have moderate to high reliability (MEPS: interrater r = .96, test-retest r = .69; SPST: interrater r = .93, test-retest r = .61) (Mathias & Nettlebeck, 1992; Castles & Glass, 1986) but questions have been raised regarding the validity of these behavioral role-play tests. Castles and Glass (1986) estimated the discriminant validity of the BSSA, the SPST, and the Interpersonal Self-Efficacy Scale (ISES) by calculating the intercorrelations among the three measures and corre- lating each of the three measures with IQ. Correlations were modest (r ranging from .06 to .47). Matson et al. (1986) also demonstrated that role-play measures correlate poorly with other social skills assess- ment techniques. These tests do not correlate with appropriate or inap- propriate social interactions, peer or staff measures of social behavior, or scores on self-report measures. An additional weakness of role-play assessments such as the BSSA, MEPS, and SPST is they are verbally demanding measures. Role-play situations are difficult to do with persons with severe dis- abilities because such simulations require complex communication. An individual with severe MR would not have the cognitive capacity to generate a number of solutions much less be able to articulate the steps involved in solving the problem. For this reason, the most com- monly used source of information regarding the social skills of men- tally retarded people is an informant who is familiar with that individ- ual’s typical behavior. CHECKLISTS Behavior checklists and rating scales are the simplest and most fre- quently used tools to summarize skills that have been observed in per- sons with severe and profound MR (Browder & West, 1992). They are relatively quick and inexpensive, and they allow a clinician to obtain a wide range of information from a variety of informants with little dif- ficulty. As such they are often the only practical way to systematically assess social skills in applied settings. However, informant-based evaluations have been criticized because they lack psychometric soundness (e.g., problems with reliability between raters and across Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 699 time), they lack precise terminology, and informant biases can affect the results (Marchetti & Campbell, 1990). A number of rating scales have been developed for assessing social skills. However, only a few have been designed to assess deficits and excesses of social skills, at least in part, in individuals with MR. They include adaptive behavior scales such as the American Association of Mental Deficiency's (AAMD) Adaptive Behavior Scale and Vineland Adaptive Behavior Scales (VABS), and social skills measures such as the Social Perfor- mance Survey Schedule (SPSS), the Matson Evaluation of Social Skills for Individuals with Severe Retardation (MESSIER), and the Assessment of Social Competence (ASC). Adaptive behavior scales. Measures of adaptive behavior generally place considerable emphasis on social skills and therefore contain domains with items designed to measure behaviors linked to this con- struct. However, these scales only contain a limited number of items pertaining to social skills and they reflect only social abilities rather than social deficits or inappropriate behavioral excesses. Overall, measures of adaptive behavior are most useful in providing a global picture of an individual’s social skills functioning (Marchetti & Campbell, 1990), but less useful in identifying specific social skills deficits to target for remediation (Bellack, 1979), or for measurement of treatment efficacy (Matson & Hammer, 1996; Perry & Factor, 1989). Despite these limitations, two of the most widely used adaptive behavior scales, the VABS and the AAMD Adaptive Behavior Scale, will be considered. The VABS consists of three editions: the Interview Expanded Form (Sparrow, Galla, & Cecchetti, 1984a), the Interview Survey Form (Sparrow, Galla, & Cecchetti, 1984b), and the Classroom Edition (Sparrow, Galla, & Cecchetti, 1985). The Expanded and Survey forms are normed for use with persons with MR. The tests must be adminis- tered by someone with a graduate degree and specific training in test administration and interpretation. The respondent is an adult who is familiar with the person being evaluated. The interviewer uses a semistructured interview method in which general questions are asked within each domain. More specific questions are asked as needed. The Expanded and Survey forms both contain Communica- 700 BEHAVIOR MODIFICATION / September 2004 tion, Daily Living Skills, Socialization, Motor Skills, and Mal- adaptive Behavior domains and an Adaptive Behavior Composite. Of these domains, the Socialization Domain and, to a much lesser extent, the Communication and Maladaptive Behavior domains contain items that fall under the rubric of social skills, depending on the defi- nition being used. The Socialization Domain consists of 134 (Expanded), 66 (Survey), and 53 (Classroom) items that are further classified into the subdomains of Interpersonal Relationships, Play and Leisure Time, and Coping Skills. Items range from very basic (e.g., looks at face of caregiver) to more complex (e.g., apologizes for unintentional mistakes) behaviors. The Communication Domain con- tains the subdomains Receptive, Expressive, and Written. Although this domain consists of items that primarily reflect behaviors related to basic communication skills, there are some items that can also be seen to overlap with social skills (e.g., spontaneously relates experi- ences in simple terms, listens attentively to instructions, smiles in response to a caregiver). Similarly, the Maladaptive Behavior Domain contains items that are also related to social skills (e.g., shows lack of consideration, swears in inappropriate situations, has poor eye con- tact). Each item on the VABS is rated 0 (no, never), 1 (sometimes or partially), 2 (yes, usually), or DK (don’t know), although some items may be rated N (no opportunity). Standard Score, National Percentile Rank, Adaptive Level, and Age Equivalent Score are calculated. The VABS has supplementary norms for Ambulatory and Nonambulatory mentally retarded people in residential facilities and mentally retarded adults in nonresidential facilities (Sparrow et al., 1984a, 1984b). Supplementary norms for persons with autism have also been published (Carter et al., 1998). The VABS is considered to have good test-retest reliability, interrater reliability, subscale intercorrelations, and construct and criterion validity. The test manu- als can be consulted for more details regarding reliability and validity. In addition, standard errors of prediction and confidence intervals have been published that can be used by clinicians when interpreting changes in obtained scores across repeated administration (Atkinson, 1990). The American Association of Mental Retardation (AAMR) Adap- tive Behavior Scale–Residential and Community, Second Edition Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 701 (ABS-RC:2) (Nihira, Leland, & Lambert, 1993), is an adaptive behav- ior scale designed and normed for use with developmentally disabled persons through 79 years of age. The scale contains 356 items and is completed in an interview format with a respondent who knows the individual being examined. The test is divided into two parts: (a) Indi- vidual Responsibility and Daily Living and (b) Social Behaviors. Part 1 contains the subdomains of Independent Functioning, Physical Development, Economic Activity, Language Development, Numbers and Time, Domestic Activity, Prevocational/Vocational Activity, Self-Direction, Responsibility, and Socialization. Part 2 consists of Social Adjustment, and Personal and Social Responsibility. Part 2’s subdomains are Social Behavior, Conformity, Trustworthiness, Ste- reotyped and Hyperactive Behavior, Self-Abusive Behavior, Social Engagement, and Disturbing Interpersonal Behavior. Items are scored yes/no or statement that best applies is selected. Domain and Subdomain age, percentile, and standard scores are calculated. The ABC:RC-2 does not have a total score. The unidimensionality of the five factors proposed in the AAMR Adaptive Behavior Scales manual has been questioned and a two-factor model proposed (Stinnett, Fuqua, & Coombs, 1999). Social skills measures. The SPSS was developed by Lowe and Cautela in 1978 (Lowe & Cautela, 1978). It is a 100-item measure used to assess positive and negative social behaviors. Items are scored on a 5-point Likert-type scale (0 = not at all, 1 = a little, 2 = a fair amount, 3 = much, 4 = very much). Typical items include the follow- ing: has eye contact, shows enthusiasm for others’good fortune, inter- rupts others, threatens others verbally or physically, and knows when to leave people alone. The scale was developed for adults of normal intelligence and could be used as a self-report or significant-other rat- ing scale. Numerous studies have demonstrated the scale has adequate psychometric properties (internal consistency, r = .88; test-retest reli- ability, r = .87). The SPSS has also been shown to correlate with other’s ratings of a patient’s social skills (Lowe & D’Ilio, 1985). Matson, Helsel, Bellack, and Senatore (1983) revised the SPSS for use with individuals with mild and moderate MR. The resulting scale is a multidimensional measure that is completed by a family member 702 BEHAVIOR MODIFICATION / September 2004 or direct-care staff member who knows the individual well and has worked with him or her for at least 6 months. The scale consists of 57 of the original 100 SPSS items. These items were retained based on Pearson product-moment correlations of .30 or greater with the total score. A principal components factor analysis was performed on the SPSS scores of 207 adults with MR (Matson et al., 1983). Four factors emerged from the data: Appropriate Social Skills, Communication Skills, Inappropriate Assertion, and Sociopathic Behavior. The pri- mary limitation of the revised scale is the lack of research examining the reliability and validity of the measure. The SPSS also lacks norms for the MR population, and its applicability has only been studied with regard to individuals in the mild and moderate ranges of MR. The MESSIER (Matson, 1994) was specifically designed to assess social skills in persons with severe and profound MR. It consists of 85 items generated from a review of existing social skills measures for children and adults, items from the social and communication domains of adaptive behavior scales, and items nominated by experts. The items are grouped into six clinically derived subscales: (a) posi- tive verbal, (b) positive nonverbal, (c) positive general, (d) negative verbal, (e) negative nonverbal, and (f) general negative. Each item is rated on frequency using a 4-point Likert-type scale: 0 (never), 1 (rarely), 2 (sometimes), and 3 (often). The MESSIER is administered by a trained examiner in a semistructured interview format. The respondent should be a caregiver who has worked for the individual for at least 6 months. Items endorsed are then transcribed onto a scoring profile under their respective subscales. The psychometric properties of the MESSIER have been studied. Internal consistency as measured by coefficient alpha was high for the MESSIER (r = .94). The test-retest reliability, as measured by a Pearson product-moment correlation, was also quite high (r = .86), indicating that peoples’ ratings remained fairly consistent over short periods of time. Good correlations were also found between raters for the total MESSIER score (r = .73) and for all positive (r = .79) and negative MESSIER items (r = .71). These numbers support the utility of the MESSIER for making consistent social skills ratings (Matson, LeBlanc, & Weinheimer, 1999). Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 703 The convergent validity of the MESSIER was evaluated by com- paring it to (a) the Socialization domain of the VABS and (b) sociometric ratings. Significant positive correlations were found between corresponding MESSIER subscales and VABS subdomains on social behaviors. Sociometric ratings also correlated highly with total MESSIER scores for 80% of raters, suggesting that the ratings on the MESSIER seem to accurately reflect a person’s social status. Although the MESSIER and VABS (Socialization Domain) seem to measure similar constructs, the MESSIER has several advantages over the VABS in measuring social skills in persons with severe or profound MR. First, the MESSIER has a larger number of social- skills-specific questions than the VABS. Second, the MESSIER can be used to illustrate an individual’s social skills and deficits, as well as maladaptive behavior excesses, whereas the VABS is limited to pro- viding information about social abilities only. This information can be useful when developing treatment and habilitation plans. The ASC (Meyer et al., 1985) was designed to assess social skills at all levels of intellectual functioning. It contains 252 discrete behaviors that are organized into 11 dimensions of social competence. Within each of the 11 dimensions, the items are grouped into eight levels rep- resenting a hierarchy of increasing social ability. Items include behav- iors that may be referred to as behavior problems. The ASC is to be completed by someone who knows the individual well. Each item is given one of three scores: “no evidence of the behavior,” “someone else’s report of the behavior only,” and “direct observation of the behavior.” A person passes a level if at least one item is passed within that level. A total score on the ASC consists of the sum of the scores for each of the 11 functions. Overall scores range from 11 to 87. The initial psychometric properties of the ASC have been studied. The measure proved to be internally consistent when administered to a sample of children and young adults diagnosed with MR (children, r = .93; young adults, r = .95). The ASC has also demonstrated good test-retest reliability (r = .90) and adequate interrater reliability (r = .70) within a school-age population of students with severe and pro- found MR. Convergent validity was assessed by examining correla- tions between the ASC and AAMR diagnoses. Scores on the total ASC correlated moderately with overall levels of MR (r = .51). These 704 BEHAVIOR MODIFICATION / September 2004 results indicate the ASC shows promise as a valid and reliable mea- sure of social competence, but a weakness is its apparent ceiling effect for young adults. IMPLICATIONS AND FUTURE DIRECTIONS Persons with MR experience varying degrees of deficits in social skills, which can lead to isolation, stigmatization, lack of coping skills, increased maladaptive behaviors, and increased rates of psychopathology (Matson et al., 2000; Matson & Hammer, 1996). Social skills deficits and excesses are a defining aspect of MR and this fact is reflected in the amount of emphasis placed on social skills within measures of adaptive functioning (Sparrow et al., 1984a). Spe- cific measures of interactive social skills can be combined with more global measures of functioning as part of an overall, individualized social skills training program. Despite gains in assessment and treat- ment technologies related to social skills in persons with developmen- tal disabilities, more work needs to be done by researchers and clini- cians in this area to help reverse the emphasis found in most institutions and community settings on the containment of behavior rather than on proactive skills building. Social skills training has been shown to be effective, and should be a priority for persons with devel- opmental disabilities (Matson & Hammer, 1996). With the advent of measures such as the MESSIER and the SPSS, researchers and clini- cians have at their disposal reliable and valid methods for selecting target behaviors for intervention and for tracking treatment effective- ness. As such, more studies that assess the effectiveness of social skills training in persons with MR are needed to identify and develop the most effective training methods. Future tasks for researchers in this area should include the estab- lishment of scaling norms for mentally retarded persons in the com- munity. Much of the research completed thus far has been conducted within large, long-term residential settings. Reliability and validity studies for the existing instruments are needed to insure the feasibility of their use within a community setting. Research with community- based samples may also help illuminate specific social skills compo- Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 705 nents that predict successful community placement. Such research could aid in the development of socially valid interventions. Further- more, in following Goldfried and D’Zurilla’s (1969) model of content validity for behavioral assessment techniques, additional situations and responses within community settings could be identified, studied, and possibly added to future editions of these social skills measures. Given the current political environment that emphasizes community placement, the need for such research is past due. Future research should also focus on developing norms for children with MR. Although some scales, such as the ASC (Meyer et al., 1985), have been normed for children with MR, such norms do not exist for MESSIER, SPSS, and other measures. Measures of adaptive behavior generally contain norms for children with physical handicaps such as deafness, but do not have supplementary norms for MR such as those that exist for adults. Attention also needs to be given to the empirical validation of a battery method for assessing social skills in mentally retarded persons. Much work has been done within the field toward developing different behavioral assessment techniques (behavioral observations, checklists, behavioral interviews, etc.) and examining their reliability, validity, and practicality, but there has not been suffi- cient attention in the literature on standardizing a battery approach to the assessment of social skills in this population. Such a battery could become an important component of an overall assessment package. Researchers could then focus on how social skills assessment inter- faces with other types of assessment such as intellectual assessment, assessment of problem behaviors, and assessment of psycho- pathology. More knowledge could then be gained regarding the rela- tionship between specific social skills variables and variables from these other areas of assessment. Finally, research should focus on the development of more specialized social skills assessment instru- ments, particularly for those areas that are found to be crucial in improving the quality of life for mentally retarded persons. REFERENCES Atkinson, L. (1990). Standard errors of prediction for the Vineland Adaptive Behavior Scales. Journal of School Psychology, 28, 355-359. 706 BEHAVIOR MODIFICATION / September 2004 Bellack, A. S. (1979). A critical appraisal of strategies for assessing social skill. Behavioral Assessment, 1, 157-176. Bellack, A. S. (1983). Recurrent problems in the behavioral assessment of social skill. Behav- ioral Research and Therapy, 21, 29-41. Browder, D.M., & West, B.J. (1991). Assessment of social skills and interfering behavior. In D.M. Browder (Ed.), Assessment of individuals with severe disabilities: An applied behavior nd approach to life skills assessment (2 ed., pp. 245-271). MD: Paul H. Brookes Publishing Co. Carter, A. S., Volkmar, F. R., Sparrow, S. S., Wang, J., Lord, C., Dawson, G., et al. (1998). The Vineland Adaptive Behavior Scales: Supplementary norms for individuals with autism. Journal of Autism and Developmental Disorders, 28(4), 287-302. Castles, E. E., & Glass, C. R. (1986). Empirical generation of measures of social competence for mentally retarded adults. Behavioral Assessment, 8, 319-330. Coe, D., Matson, J. L., & Fee, V. (1990). Training nonverbal and verbal play skills to mentally retarded and autistic children. Journal of Autism & Developmental Disorders, 20, 177-187. Duncan, D., Matson, J. L., Bamburg, J. W., Cherry, K. E., & Buckley, T. (1999). The relationship of self injurious behavior and aggression to social skills in persons with severe and profound learning disability. Research in Developmental Disabilities, 20, 441-448. Gettinger, M., & Kratochwill, T. R. (1987). Behavioral assessment. In C.L. Frame & J. L. Matson (Eds.), Handbook of assessment in childhood psychopathology: Applied issues in differential diagnosis and treatment evaluation(pp. 131-161). New York: Plenum Press. Goldfried, M. R., & D’Zurilla, T. J. (1969). A behavior-analytic method for assessing compe- tence. In C. D. Spielberger (Ed.), Current topics in clinical and community psychology (Vol.1, pp. 151-196). New York: Academic Press. Kazdin, A. E., & Matson, J. L. (1981). Social validation in mental retardation. Applied Research in Mental Retardation, 2, 39-53. La Greca, Annette M., Stone, Wendy L., Bell, & Charles, R. (1982). Assessing the problematic interpersonal skills of mentally retarded individuals in a vocational setting. Applied Research in Mental Retardation, 3, 37-53. Libet, J. M., & Lewinsohn, P. M. (1973). Concept of social skill with special reference to the behavior of depressed persons. Journal of Consulting & Clinical Psychology, 40, 304-312. Lowe, M. R., & Cautela, R. (1978). A self report measure of social skill. Behavior Therapy, 9, 535-544. Lowe, M. R., & D’Ilio, V. (1985). Factor analysis of the Social Performance Survey Schedule. Journal of Psychopathology & Behavioral Assessment, 7,13-22. Marchetti, A. G., & Campbell, V. A. (1990). Social skills. In J. L. Matson (Ed.), Handbook of behavior modification with the mentally retarded (2nd ed., pp. 333-355). New York: Plenum. Mathias, J. L., & Nettelbeck, T. (1992). Reliability of seven measures of social intelligence in a sample of adolescents with mental retardation. Research in Developmental Disabilities, 13,131-143. Matson, J. L. (1986). Assessing social behaviors in the visually handicapped: The Matson Evalu- ation of Social Skills with Youngsters (MESSY). Journal of Clinical Child Psychology, 15,1986. 78-87. Matson, J. L. (1994). Autism in children and adults: Etiology, assessment, and intervention. Pacific Grove, CA: Brooks/Cole. Matson, J.L. (1995). The Matson Evaluation of Social Skills for Individuals with Severe Retarda- tion. Baton Rouge, LA: Scientific Publishers, Inc. Matson, J. L., Fee, V. E., & Coe, D. A. (1991). A social skills program for developmentally delayed preschoolers. Journal of Clinical Child Psychology, 20,428-433. Bielecki, Swender / SOCIAL SKILLS ASSESSMENT 707 Matson, J. L, & Hammer, D. (1996). Assessment of social functioning. In J. W. Jacobson & J. A. Mulick (Eds.), Manual of diagnosis and professional practice in mental retardation (pp. 157- 163). Washington, DC: American Psychological Association. Matson, J. L., Helsel, W. J., Bellack, A. S., & Senatore, V. (1983). Development of a rating scale to assess social skill deficits in mentally retarded adults. Applied Research in Mental Retar- dation, 4, 399-407. Matson, J. L., LeBlanc, L. A., & Weinheimer, B. (1999). Reliability of the Matson Evaluation of Social Skills in Individuals With Severe Retardation. Behavior Modification, 23, 647-661. Matson, J. L. & Ollendick, T. H. (1988). Enhancing children’s social skills: Assessment and training. Elmsford, NY: Pergamon. Matson, J. L., Smiroldo, B. B., & Bamburg, J. W. (1998). The relationship of social skills to psychopathology for individuals with severe or profound mental retardation. Journal of Intellectual and Developmental Disability, 23, 137-145. Matson, J. L. & Swiezy, N. (1994). Social skills training with autistic children. In J. L. Matson (Ed.), Autism in children and adults: Etiology, assessment, and intervention(pp. 241-260). Belmont, CA: Brooks/Cole Publishing. Matson, J. L., Zeiss, A. M., Zeiss, R. A., & Bowman, W. (1980). A comparison of social skills training and contingent attention to improve behavioral deficits of chronic psychiatric patients. British Journal of Social and Clinical Psychology, 19, 57-64. Meyers, A. W., Schleser, R., & Cooke, C. J. (1979). Cognitive contributions to the development of gymnastics skills. Cognitive Therapy & Research, 3, 75-85. Nihira, K., Leland, H., & Lambert, N. (1993). AAMD Adaptive Behavior Scale–Residential and Community (2nd ed.). Austin, TX: Pro-Ed. Njardvik, U., Matson, J. L., & Cherry, K. E. (1999). A comparison of social skills in adults with autistic disorder, pervasive developmental disorder not otherwise specified, and mental retar- dation. Journal of Autism and Developmental Disorders, 29, 287-295. Nihira, K., Price-Williams, D.R., Douglas & White, J. F. (1988). Social competence and maladaptive behavior of people with dual diagnosis. Journal of the Multihandicapped Per- son, 1, 185-199. Perry, A., & Factor, D. C. (1989). Psychometric validity and clinical usefulness of the Vineland Adaptive Behavior Scale and the AAMD Adaptive Behavior Scale for an autistic sample. Journal of Autism and Developmental Disorders, 19(1), 41-55. Platt, J. J., & Spivack, G. (1975). Unidimensionality of the Means-Ends Problem-Solving (MEPS) procedure. Journal of Clinical Psychology, 31,15-16. Raymond, K. L., & Matson, J. L. (1989). Social skills in the hearing impaired. Journal of Clinical Child Psychology, 18, 247-258. Reiss, S. (1994). Issues in defining mental retardation. American Journal on Mental Retardation, 99, 1-7. Shapiro, E. S., & Browder, D. M. (1990). Behavioral assessment. In J. L. Matson (Ed.), Hand- book of behavior modification with the mentally retarded (2nd ed., pp. 93 - 122). New York: Plenum Press. Sparrow, S. S., Galla, D. A., & Cicchetti, D. V. (1984a). Vineland Adaptive Behavior Scales: Sur- vey form manual. Circle Pines, MN: American Guidance Service. Sparrow, S. S., Galla, D. A., & Cicchetti, D. V. (1984b). Vineland Adaptive Behavior Scales: Expanded form manual. Circle Pines, MN: American Guidance Service. Sparrow, S. S., Galla, D. A., & Cicchetti, D. V. (1985). Vineland Adaptive Behavior Scales: Classroon edition. Circle Pines, MN: American Guidance Service. Stinnett, T. A., Fuqua, D. R., & Coombs, W. T. (1999). Construct validity of the AAMR Adaptive Behavior Scale–2. School Psychology Review, 28(1), 31-43. 708 BEHAVIOR MODIFICATION / September 2004 Wing, L., & Gould, J. (1979). Severe impairments of social interaction and associated abnormal- ities in children: Epidemiology and classification. Journal of Autism and Developmental Disorders, 9, 11-29. Zigler, E., & Burack, J. A. (1989). Personality development and the dually diagnosed person. Research in Developmental Disabilities, 10, 225-240. JoAnne Bielecki (M. A.) is currently pursuing her doctoral degree in psychology at Loui- siana State University under the supervision of Dr. Johnny L. Matson. She will be com- pleting her predoctoral internship in June of 2003 at UCLA’s Neuropsychiatric Institute and Hospital. JoAnne received her bachelor’s degree from the University of Connecticut in Storrs. Dr. Stephen L. Swender is the Associate Director of Psychology at Pinecrest Develop- mental Center, Central Louisiana State Hospital APA-Approved Internship Consortium. His interests include the assessment and treatment of behavior problems and psychiatric conditions in the developmentally disabled.