Multilevel multiple measurement in an ecosystemic framework

This study is set within the context of an ecosystemic framework for the conceptualization and understanding of familial relationship patterns occuring between parents (usually mothers) and oppositional, noncompliant children. Emphasis is placed upon the development of a multiple measures assessment technology which might be effective in measuring and evaluating various levels of the family system. Specific levels measured here were (1) target child behavior and parent response in the home, (2) parental reports regarding the positive or negative valence of a particular 24 hour period, (3) coercive and noncoercively oriented verbal interactions in the home, and (4) parent reports during therapy sessions about coercive and noncoercive episodes during the preceding week. The study introduces a new observational technology in the form of an audio recording device which is placed in the home and which randomly chooses and records one ten minute sample of in-home verbal interaction each hour.

Four families referred for problems with oppositional children were treated with a three phase, expanded, parent training oriented approach to psychotherapy. Each family was assessed using the measures mentioned. Early-late comparisons were conducted for each family on each measure to see if that measure indicated significant change for the target period. A sign test was then used to compare the various pairs of measures within each family and test for agreement between measures as to whether there had been significant change. Finally a correlational comparison was run between the four measures using the total scores across families. The power of these statistical analyses was attenuated due to a lack of common data points between variables and the results were inconclusive with respect to possible agreement between variables. Questions then are asked about whether the obtained lack of agreement between variables is due to the lack of common data points or unique profiles within the families on the various measures. A hint of correlation between parents' reports in therapy and the in-home recorded interactions does suggest new possibilities for future study.