Date of Award

Spring 1981

Document Type

Dissertation - Restricted

Degree Name

Doctor of Philosophy (PhD)


Educational Policy and Leadership

First Advisor

Ivanoff, John

Second Advisor

Bardwekk, Rebecca

Third Advisor

Topetzes, Nick


Crockett's (1965) imbuement of personal construct theory (Kelly, 1955) with Werner's (1957) orthogenetic principle of development has stimulated theoretical and empirical investigation of the manner in which children's interpersonal perceptions might vary as a function of developmental level. Research applying this cognitive-developmental model through content analyses of children's descriptions of others has shown a systematic increase with age in the number of constructs used (degree of differentiation) as well as shifts in the type of construct used from egocentric/concrete to nonegocentric/abstract (level of organization). The goal of the present investigation was to provide initial information regarding the influence of emotional disturbance upon degree of differentiation (number) and level of organization (type) of interpersonal constructs at several developmental levels. It was hypothesized that the interpersonal construct systems of emotionally disturbed (ED) youngsters would evidence greater simplicity with respect to these structural variables than would the construct systems of non-emotionally disturbed (non-ED) age-mates. It was further hypothesized that differences between ED and Non-ED groups would increase with corresponding increases in age. The present investigation conceptualized the effects of emotional disturbance upon the development of interpersonal constructs in terms of a developmental delay. Forty-five ED and forty-five Non-ED male subjects of average-range intelligence and representing three developmental levels (Mean CAs = 9.3, 11.3, 13.4) completed a four-peer version of the Role Category Questionnaire. Content analyses of the resulting descriptions yielded measures of differentiation (number of constructs) as well as measures of organization (predominant type) of interpersonal constructs. Multiple regression analysis was employed to determine the unique contribution provided by group membership to the reduction of residual variance in each of the measures of the dependent variables. Analysis of data for number of constructs supported the hypothesis regarding differences associated with group membership. ED subjects used significantly fewer constructs in their descriptions of others than did Non-ED subjects. No general age-related trends were observed and, subsequently, the lack of a significant age x group interaction failed to support the anticipated increased difference between groups with increases in age. A significant interaction of fluency x group emerged which required establishing regions of significance with respect to values of fluency within which the differential effects of group membership could be interpreted as significant. A substantial proportion of the transcripts comprising the data set of the study exceeded the critical values for fluency and, thus, were interpretable with respect to significant group differences. Analysis of data for level of organization of constructs failed to support the hypothesis regarding group differences. Although a general increase in organization associated with increased age was observed, the lack of a significant age x group interaction failed to support the hypothesized increased difference between groups associated with increase in age. Suggestions for further research were discussed including: a method for addressing the issue of fluency; the need for a measure of organization of constructs which has greater sensitivity to structural variability than that provided by type of construct; further systematic study of the apparent reciprocal interaction of differentiation and organization within a limited age range; and extension of the present study to include complexity of interpersonal constructs, emotional disturbance, and communicative effectiveness.



Restricted Access Item

Having trouble?