Date of Award


Degree Name

Doctor of Philosophy



First Advisor

Dr. Malcolm Robertson

Second Advisor

Dr. Chris Koronakos

Third Advisor

Dr. R. Wayne Fuqua

Fourth Advisor

Dr. Sid Dykstra


The theoretical concepts of rule-governed behavior and five controlling variables identified from the literature were applied to the treatment of first-time offenders in a half-way house. The subjects had similar family, educational, and criminal histories. This population differed from those used in previous studies of rule-governed behavior (i.e., college students). Given the fact that reinforcement histories do affect rule compliance, it was expected that some differences would emerge in the way these subjects responded to rule changes.

Rules in the Center were operationally defined by specification of behavior, occasion, and consequence. Subsequent rate of rule violation was compared to the rate obtained during pre-baseline conditions. Rules were then assigned to three equal clusters based on rule type and probability of violation. All rule clusters were presented in three formats (oral, written, and demonstration). A multiple baseline design allowed for measurement of presentation effects. A second experimental condition was introduced to measure sensitivity to rule consequation. Consequences for rule violations were removed and reinstated at unannounced intervals varying in length from 2 hours to 48 hours. Violations during "no consequence" and "consequence" intervals were compared to determine sensitivity of resident behavior to the existing contingencies.

Operationally defining existing rules for the staff reduced violation rates among residents when compared to pre-baseline rates. Subsequent presentations of revised rules to the residents produced an initial increase in violation rates; however; overall rule violations decreased with repeated presentation. Rules presented orally resulted in the greatest disruption (i.e., increased rule violations) across all clusters. Resident sensitivity to change in contingencies was evident during the consequence - no consequence phase. This condition produced a general increase in violation rate but produced an increase of a greater magnitude during intervals when consequences were applied. Some level of rule-control was apparent for both staff and inmate behavior.

Access Setting

Dissertation-Campus Only