|
Abstract
|
This study represents an attempt to discover whether there are distinctive patterns of day-to-day interactions that distinguish abusive and neglectful families from families with no known history of abusing or neglecting their children. Observational data collected in the homes of 17 abuse, 17 neglect, and 19 control families indicated that the abusive and neglectful parents demonstrated lower rates of interaction, overall, and were more likely to emphasize the negative in their relationships with their children. The implications of these data for further research and for intervention efforts are discussed.
VioLit summary:
OBJECTIVE:
This aim of this study by Burgess and Conger was to investigate whether there are distinctive patterns of day to day interactions that distinguish abusive and neglectful families from families with no known history of abusing or neglecting their children. Implications for research and intervention were discussed.
METHODOLOGY:
A quasi-experimental design was employed. Three family types were selected for the study: (1) those where nonaccidental physical injury had happened to one or more children by a caretaker (abuse); (2) those where one or more children experienced a lack of parental attention so severe that it reached the attention of legal authorities, i.e., inadequate nourishment (neglect); (3) families with no official record of abuse or neglect but were similar to the first two types in terms of demographics (control). The families classified as abusive (17) and neglectful (17) were recruited through the Commonwealth of Pennsylvania's Department of Public Welfare. The control families (19) were contacted by the Pennsylvania Field Research Laboratory. The mean income for all the families was less than $9,000 per year. Parents averaged between 10 and 11 years of formal schooling. The overall mean was 3 children per family, and the median age for children was 6.5 years. Participating families were paid $10 for each of four visits. Two observers were scheduled to see each family at home for 6 hours during a given week. Four hours of the 6 were spent observing family interaction while the additional 2 hours were used to administer questionnaires. Each of the four observation sessions were separated into 3 distinct tasks: construction, skill, and discussion. The tasks were selected to facilitate interaction in situations common to most families. Construction tasks i.e., tinker toys, set the occasion for cooperative interaction while skill tasks, i.e., throwing bean bags through a hoop, allowed for competitive interaction. Finally, the discussion tasks including issues like "What would your family like to do for a vacation?" facilitated verbal exchange among family members. Within each task a particular family member was the focal subject twice for a minimum of 1 1/2 minutes. The principle dependent variable was the observed pattern of interaction among family members. The observational code used was recorded through the Behavioral Observation Scoring System. (BOSS) which used a 10 digit keyboard, a stopwatch and a modified tape recorder. When a particular behavior occurred, the observer would depress a key and an energy impulse would be transmitted into a magnetic tape. Special programs deciphered the impulses from the magnetic tape. The behavior code consisted of 5 column entries. Column (1) Type of interact: For any interval for which a particular family member was the focal subject he or she could direct a verbal (=1) or physical (=3) response to another family member. Or, the focal subject would be the recipient of another's verbalization (verbal receive=2) or physical response (=4). Column (2) Emotional affect of interact: Modified the general interact entry by coding the emotional affect for that behavior (1=neutral, 2=positive, 3=negative). Columns (3) and (4) identified the person interacting with the focal subject. Column (5) indicated the occurrence of a command (prescriptive command=1, proscriptive command=2, comply=3, refuse=4, neither=5). Once the family had been observed, the cassette tapes were played through a special electronic interface which noted the events as they occurred as well as the passage of time between data entries, allowing for comparisons to be made within and across families. In an attempt to minimize observer bias, the observers were not informed until all observations were completed, whether the family they were observing were abuse, neglect, or control. The observers operated in pairs which were randomly shifted in composition every 2 or 3 weeks to prevent the development of unique definitions of the behavior codes. In order to check reliability, both observers coded family interaction over all sessions. One observer's tapes were randomly selected for complete analysis. Then a probe tape was randomly selected from the second observer's coding to compare with the data from the first. Observer reliability was estimated by computing correlation coefficients, slopes, and intercepts using the rates of a particular behavior scored for each family member by the two observers. Rates of behaviors were calculated and F-test and the Duncan multiple range test were used to analyze the data.
FINDINGS/DISCUSSION:
Overall, F tests showed significant differences between parents for total behavior rates and for behaviors directed toward children. For total behavior rates, the qualitative dimensions of parental activities appeared to differ most among the groups. Duncan multiple range test was employed to make abuse control and neglect control comparisons. Overall abusive parents exhibited lower rates of verbal and physical behavior than the control parents. Given these depressed levels of activity, they were relatively more aversive than controls. Abusive parents were also less compliant and less positive with their children. Neglect parents were the most negative and least positive of the 3 groups. They also made more requests of their children and yet were least compliant with requests from them. Parental behavior for each group were analyzed separating the means for mothers and fathers. Levels of statistical significance for control and problem group comparisons were completed using the Duncan multiple range test. Mothers in the abuse sample were most different from control counterparts showing lower rates of verbal behavior and positive contacts with their children. Abuse fathers differed from control fathers only in their overall rate of compliant behavior, with no significant differences between abuse and control children in their responsiveness to their fathers. In addition, abuse fathers spoke to their children more frequently than their spouses, a finding unique to abuse parents. Neglect mothers stood out for their extremely negative behavior and low rates of positive interaction. However, there were no statistically significant differences between neglect and control children in their responding to their mothers. Neglect fathers were characterized by low rates of positive and compliant responding. Children in neglect families spoke to their fathers less often than controls, interacted less positively with them, and initiated fewer physical contacts. Overall, there was a general trend toward higher rates of positive interactions in control families compared to problem families.
AUTHORS' RECOMMENDATIONS:
The authors argued that in order to understand the processes that maintain patterns of abusive interaction, we must inspect and analyze the contingencies which exist between the patterns themselves. Molecular analyses focusing on the response sequences and conditional probabilities must be carried out. Ultimately, experimental analyses should be conducted to determine the extent to which certain behavior patterns are functionally related to specific others. This was necessary to determine what is presently maintaining such dysfunctional patterns of interaction and also to gain further insight into how such patterns develop.
EVALUATION:
The research design employed in this study is unique and manages to capture some of the strengths of quantitative and qualitative study. The efforts made to try to ensure standardization and eliminate bias give more credence to the findings. The sample selection from state-identified abusive and neglectful families, the relatively small sample size (17 each for abusive and neglectful; 19 for control), and the possibility of bias from refusals all call for cautious interpretation of the findings. Also, such variables as race, SES, and geographic region, potentially important, were not or could not be addressed. The basic finding of this study was that abusive parents were least interactive and that neglectful parents were negative and were the least positive; this reflects the need to better understand family dynamics not only to identify how abuse may start but how intervention may be done. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado)
KW - Pennsylvania
KW - Domestic Violence Causes
KW - Domestic Violence Offender
KW - Child Abuse Causes
KW - Child Abuse Offender
KW - Child Physical Abuse Causes
KW - Child Physical Abuse Offender
KW - Child Neglect Causes
KW - Child Neglect Offender
KW - Adult Offender
KW - Adult Parent
KW - Adult Violence
KW - Parent Offender
KW - Offender Characteristics
KW - Family Relations
KW - Family Environment
KW - Parent Child Relations
KW - Parent Relations
Language: en |