Unstructured Observation

In unstructured observation, the researcher enters the field with some general ideas of what might be salient, but not of what specifically will be observed. Therefore, observation is holistic, unstructured, and unfocused, with the investigator attempting to document as much as possible about the setting and its participants in order to discover themes of interest. Unstructured observation is not constrained by checklists and coding schemes; rather, the researcher [Page 908]reports in narrative style about observations that are relevant to the research questions. Unstructured observation is most frequently associated with an interpretivist, constructivist paradigm that emphasizes the importance of context, sees knowledge as being co-constructed by both participants and investigators, and asserts it is impossible to completely separate the observer from the observed.

Unstructured observation is characterized by emergent research design, recognizing that what is observed may change as experience is gained in the setting. Data are collected about the physical setting, its history (often through documents), the context, the participants (including their physical characteristics such as age, gender, race and stature, their activities and their interactions with others), and the rhythm of activities. Investigators look and listen, and data collection is very thorough and detailed, especially at the beginning of a project when what is significant is still unknown. Unstructured observation recognizes that important themes can emerge from the mundane. Data are usually recorded as fieldnotes that include jottings, maps, and diagrams. Unstructured observation usually leads to more structured observation as more is learned about the context. This increased structure is better regarded as focused rather than as structured or systematic observation, which uses checklists.

Unstructured observation is particularly useful in the early stages of an observational study. It is more holistic than other forms of observation, capturing more information about the setting of interest. Unlike methods such as interviews, unstructured observation allows investigation of context and process in an ongoing rather than episodic manner. It is effective for looking at interaction among individuals and between groups. When used with interviews, unstructured observation allows for comparison between participant accounts and actual behavior. Unstructured observation is a flexible method that allows for the collection of comprehensive and rich data.

A major weakness of unstructured observation is that it is susceptible to observer bias as investigators choose what to observe and how to process and analyze that information. In keeping with the concept of researcher as instrument, the quality of data collected and the findings are heavily dependent on the skill and experience of the investigator. Reactivity is also a potential problem, although it is usually overcome through habituation. The very nature of unstructured observation means that it is impossible to predict in advance all that will be of interest. Therefore, for example, obtaining informed consent from all participants (if necessary) requires ongoing effort as individuals move in and out of the research context.