Observations for scientific purposes (see also Bake-man & Gnisci, chap. 10, this volume) may significantly vary with respect to their obtrusiveness and hence (non)reactivity. When the social sciences started developing its arsenal of methods, alternatives of—at least at that time—highly obtrusive procedures had been discussed. In the beginning of the debate, "participatory observation" was—and still is—called a type of research with fuzzy definitions of the relations between investigators and targets (Couto, 1987). The investigators are not unobtrusive at all because they act in the field under investigation. Their specific impact on the processes to be observed, however, is considered low because their positions as researchers are masked by their activity in the field. For targets under investigation, the researchers' position is less extraordinary—and hence probably less salient to the subjects—than the position of a scientific observer in a laboratory. Presently, the concept is of some importance in culture-specific contexts or some areas of clinical and family research.
"Simple observations" are considered as potentially nonreactive. Bloom and Fischer (1982) discussed four types:
■ The observation of physical and bodily signs (jewelry, changes in hairstyle, clothing, and makeup) that are potential indicators of attitudes or behaviors. This is also addressed by studies on symbolic self-completion (Wicklund & Gollwitzer, 1981).
■ The analysis of expressive movements (smiles, frowns, gestures) as indicators of attitudes and feelings.
■ Physical location analysis that can indicate certain attitudes (e.g., seating patterns as an index of interracial relations).
■ The analysis of language behavior (which is tape-recorded for various research goals).
The process of observation may not be unobtrusive in its literal sense because either people or equipment or both are present. However, as long as "the observer has no control over the behavior or sign in question, and plays an unobserved, passive and unobtrusive role in the research situation" (Webb et al., 1966, p. 112), reactivity should be reduced considerably.
Was this article helpful?