Date of Graduation


Document Type


Degree Type



Eberly College of Arts and Sciences



Committee Chair

Cheryl B McNeil


This study sought to identify differences in caregiver-child interactions between caregivers with and without prior reports of child physical abuse. Data for the abuse group were made available for secondary analyses by the National Archive for Child Abuse and Neglect via Cornell University. Data for the comparison group were collected for a previous study at West Virginia University. For both studies, caregiver-child dyads participated in a 5---minute, videotaped observation in a situation in which the parent had to exert moderate control over the child. Data that had been coded using the Dyadic Parent---Child Interaction Coding System (Eyberg, et al., 1994; Eyberg, et al., 2005) were compared in 70 caregiver-child dyads. Groups were similar with regard to child behavior, parent age, child gender, parental education level, caregiver marital status, and relation of parent to child. Analyses were conducted with both the sample of 70 dyads, in which children in the abuse group were older, and with a subsample of 41 dyads, in which all children were 4- to 5-years-old and significantly more caregivers in the abuse group were male. Race was considered in both sets of analyses. Results with both samples revealed that, over and above demographic differences, parents in the abuse group talked significantly more and, after additionally controlling for total talk, gave significantly less praise. These results inform parental fitness examinations, but observations should be examined within a broader assessment. Limitations of this study include comparison of two pre-existing samples with some procedural and sample differences. Future research should consider demographic differences when examining caregiver-child differences in abusive samples and attempt to include fathers.