Publication Date


Advisor(s) - Committee Chair

Doris Redfield, Richard Miller, Daniel Roenker

Degree Program

Department of Psychology

Degree Type

Master of Arts


The purpose of this study was to investigate the interactions between direct care staff and the mentally retarded (MR) clients that they served in an institutional setting. It was hypothesized that (a) most of the staff-client interactions, regardless of quality (i.e., positive or negative) and/or level of intellectual functioning, would be instructional rather than conversational or no response interactions; (b) there would be an insufficient amount (i.e., duration and frequency) of conversational interaction, regardless of quality, across intellectual functioning levels; and (c) there would be a greater number of negative no response interactions (i.e., ignoring an appropriate client verbalization) than positive no response interactions (i.e., ignoring an inappropriate client verbalization). The impetus for these specific hypotheses was a concern that overall, in institutional settings, insufficient stimulation through direct care staff-client interaction occurred for proper implementation of individualized habilitation plans across IQ levels.

The subjects were 45 clients currently residing in an institutional setting in Kentucky. The criterion variable was IQ. The predictor variables were twelve possible combinations of type (i.e., conversational, instructional, and no response), quality (i.e., positive and negative), and amount (i.e., frequency and duration) of staff-client interaction. An experimenter-developed instrument was used to record the observational data of staff-client interactions. The experimenter-developed instrument yielded interobserver agreement, among six observers, ranging from .81 to 1.00; intraobserver stability, over a two-week period, ranged from .86 to 1.00 across the six observers. Subjects were observed four times for six minutes each on a varied schedule across four days.

Data were analyzed using a stepwise multiple regression procedure to determine the most significant model of staff-client interaction for Predicting 10. Pearson product-moment correlation coefficients were also calculated between each of the 12 predictor variables and the. criterion (IQ) since the number of subjects (n = 45) was small in comparison to the number of predictor variables.

Results of the data analyses supported the experimenter's hypotheses. The stepwise procedure indicated that the single significant predictor of IQ was duration of positive instruction. That is, a significant inverse relationship existed between level of intellectual functioning and duration of staff-client positive instructional interaction (F = 6.72, p <.01). A Pearson product-moment correlation coefficient (r = -.37) confirmed the results of the stepwise multiple regression procedure indicating a significant inverse relationship between IQ and duration of positive instruction (IPD). Pearson coefficients also indicated significant inverse relationships between IQ and frequency of positive instruction (r = -.35), frequency of negative instruction (r = -.33), and duration of negative instruction (r = -.32). Frequency of positive instruction (IPF), frequency of negative instruction (INF), duration of negative instruction (IND), and IPD shared significant amounts of variance indicating that instructional interactions, regardless of quality or amount, account for the only significant variance across IQ levels. Assuming linearity of the data, the hypothesis that negative no responses would exceed positive no responses across IQ levels was confirmed.


Psychology | Social and Behavioral Sciences

Included in

Psychology Commons