John D. Cone

The relevance of reliability and validity for behavioral assessment

The differences between traditional and behavioral assessment are primarily conceptual, not methodological, and reliability and validity considerations therefore apply to both. Insufficient attention has been given these concepts by behavioral assessors. Classical reliability and validity issues are rephrased in terms of generalizability theory, and six universes of generalization are described. The relevance of each university for traditional and behavioral assessment is suggested, and a prototypical minimal generalizability study for the purveyors of new instruments is outlined.

The behavioral assessment grid (BAG): A conceptual framework and a taxonomy

To provide organization to the field of behavioral assessment, a threedimensional system for classifying behavioral measures is proposed that combines contents (cognitive, physiological, motor), methods (interviews, self-reports, ratings by others, self-observation, direct observation), and universes of generalizability (scorer, item, time, setting, method, dimension). Various theoretic, heuristic, and practical uses of the system are described.

Differences in abusive, at-risk for abuse, and control mothers' descriptions of normal child behavior

Differences in the descriptions of normal child behavior were assessed for mothers at-risk for abuse (N = 18) and control (N = 20) mothers, and compared to a criterion group of currently abusive mothers (N = 5). Three segments of each of nine videotaped scenes were rated by the mothers. At-risk mothers rated more segments as negative (M = 11.1) than did control mothers (M = 7.0), and were more like the abusive mothers who were the most negative of all (M = 16.4). When type of scene (e.g., noncompliant, mildly aversive, daily activities, child unattended) was analyzed, the at-risk mothers rated more of the mildly aversive (p less than .05) and child unattended (p less than .05) scenes as negative than did the controls. Abusing mothers rated significantly more segments as negative (p less than .05) than the other groups on all but noncompliant scenes. When specific risk factors were examined, insularity and directly observed negative interactions between mothers and their own child were significantly related to total segments rated as negative (p less than .02; p less than .05, respectively). The study was conceptualized and results discussed in terms of faulty stimulus discrimination; specifically, overgeneralizing from negative stimuli on the part of troubled mothers.

Intervention Planning using Adaptive Behavior Instruments

Adaptive behavior is defined as those interactions of an individual with the environment that are functionally effective in that environment. Using this definition it is shown that adaptive behavior instruments that are environmentally relevant can be used effectively in intervention planning. Five characteristics desirable in such instruments are listed. Selected instruments are reviewed, including the Adaptive Behavior Scale, Vineland Adaptive Behavior Scales, Scales of Independent Behavior, and The Pyramid Scales. A systems approach to assessing adaptive behavior is described, and it is argued that the multiple functions of adaptive behavior assessment require a set of carefully interrelated assessment procedures. No single instrument is sufficient. An example of such a system is presented and it is shown how using it can lead to the production of long-range, annual, monthly, and short-term goals, as well as to the identification of immediate instructional objectives.

Target Selection of Social Skills for Children The Template-Matching Procedure

The present studies represent a preliminary test of the template-matching procedure as a means of empirically identifying critical social behaviors for individual client children. Study 1 assessed the consistency with which female child informants identified behaviors they preferred in their friends (template behaviors). Study 2 validated these template behaviors on the same subjects by manipulating the behaviors in nonscripted confederates and assessing the impact of the manipulations on behavioral and sociometric measures. It was found that: (1) subjects were consistent when identifying most preferred template behaviors; (2) subjects reciprocated template behaviors in kind; (3) different patterns of acceleration of template behaviors in confederates were associated with collateral increases in untrained behavior in subjects and in subject responsivity to confederates; and (4) sociometric ratings obtained from some subjects increased, but these increases were not statistically significant. Implications of these findings for future development of the template-matching procedure are discussed.

Applied Behavior Analysis and the Solution of Environmental Problems

In 1970, Wohlwill described the emerging field of environmental psychology and suggested that its emergence was due partly to a general concern with the deteriorating quality of our physical environment. A good deal of research and writing has appeared since Wohlwill’s paper, and as the titles of the chapters in this and the previous volume of the Human Behavior and Environment series attest, the interests of persons in the discipline are quite varied. Surprisingly, only a few of the titles suggest concern with the quality of the physical environment itself.

Developing and maintaining social interaction in profoundly retarded young males.

Social interaction was increased in five profoundly retarded males using formal training, stimulus control, and rewarded generalization procedures. Seven behaviors were monitored for each boy to determine whether correlated changes occurred in maladaptive behaviors as social interaction varied over the several phases of a withdrawal design that included multiple-baseline features. All five boys increased their social interaction and reduced unoccupied and self-stimulatory behavior. These changes were maintained as continuous reinforcement was reduced to a single response-contingent reinforcer per 15- minute session. Four follow-up probes showed the stability of the changes.

Decelerating Environmentally Destructive Lawn-Walking.

Three general strategies (response difficulty, chaining, and prompting) were used to generate six interventions to decrease lawn walking in a park. Interventions were evaluated in separate ABA designs during which 1885 persons (61% male) were observed. While none of the three strategies appeared generally superior, some interventions were more effective than others. Formal prompts (signs) and procedures involving response difficulty (i.e., physical barriers to pathways) were most effective. An intervention designed by professional planners independent of this study was also evaluated and shown to increase lawn walking slightly, though not significantly, rather than to decrease it. The value of conducting a behavioral analysis prior to implementing permanent environmental rearrangements was emphasized.

Template Matching: An Objective Approach to Placing Clients in Appropriate Residential Services.

Informal strategies for placing clients in community residential arrangements are reviewed and contrasted with a more formal, objective strategy based on the template matching procedures of Bem (Bem & Funder, 1978; Bem & Lord, 1979). Group homes (N = 14) were described in terms of the minimum adaptive behavior required for successful placement therein. They were also described in terms of the maximum maladaptive behavior tolerated. The Pyramid Scales and Part II of the Adaptive Behavior Scale were used to develop the environmental templates. Considerable variation in behavior requirements was noted among group homes. A nonsignificant correlation between adaptive and maladaptive behavior requirements shows the need to consider both types of behavior when developing placement criteria. Implications of the template matching approach for decision making and for the design of various health and human services are discussed.

Social and solipsistic observer training: Effects on agreement with a criterion

The need to train accurate, not necessarily agreeing, observers is discussed. Intraobserver consistency as an intermediate criterion in such training is proposed and contrasted with the more familiar criterion of interobserver agreement. Videotaped observations of social interactions between handicapped and nonhandicapped preschoolers provided the medium for examining the criterion agreement of four observers trained against each type of standard. Observers generally failed to show high levels of criterion agreement whether trained to a within- or to a between-observer agreement standard. The results varied somewhat with the frequency of behaviors, however. Correlations between interobserver agreement and intraobserver consistency were variable but somewhat higher when interobserver agreement was the training criterion than when intraobserver consistency was the criterion. Correlations between interobserver agreement and criterion agreement ranged from — .16 to .89 during interobserver agreement training. Correlations between intraobserver consistency and criterion agreement ranged from — .23 to .99 during intraobserver consistency training.