ABSTRACT
Increasingly, researchers are turning to observational data to understand the development and functioning of family systems and to identify family interaction patterns that reliably differentiate problem from nonproblem families. The burgeoning interest in observational methods stems in large part from the emphasis of family systems theoretical orientations on the interdependency of individuals and relationships within the family and on the need to capture the nature of complex systemic interactions scientifically. Observational research is costly and labor-intensive, yet it offers investigators the opportunity to collect reliable and valid data unavailable through traditional assessment strategies. From a methodological perspective, whereas self-report measures are useful when one is interested in family members’ perceptions of family functioning (e.g., satisfaction with roles within the family), observational data are preferred for more complex aspects of family functioning. For example, observational methods can capture patterns of interaction, including responses that are difficult to describe such as behavior rates, affective expression sion, nonverbal behavior, and events that participants may be unwilling to report or else may distort as a function of the event’s social undesirability or of the effort required for adequate description (Hartmann & Wood, 1992). Direct observational methodology may be particularly useful for studying families with young children who do not have the language skills to respond to questionnaires or interviews. In addition, direct observations have high face validity and good generalizability and are less susceptible to confounding influences (Jacob, Tennenbaum, & Krahn, 1987). Data gathered through the direct observation of behaviors without an intermediary
questionnaire or other instrument reflect those behaviors directly. It therefore requires relatively little conceptual inference to connect the data to the phenomenon (Suen & Ary, 1989).
