Non participant observation and self reports
Observation and self-report are appropriate data collection methods for health care studies, given that much of the health care providers’ professional lives are spent observing and interviewing patients. Researchers who use observation aim to describe practice and behaviours as they unfold, collecting ‘first hand’ data from the research setting itself. Self-report measures use the perceptions of the participants themselves to study needs assessments and evaluations.
Observation can be structured or unstructured, participant or non participant, overt or covert. A non – participant structured approach provides a better control of reliability and validity of measurements than the unstructured approach, however, unlike the latter, it has limitations in the richness and the large variety of data which can be collected (Waltz, Strickland, Lenz, 2010).
Wadensten (2005) uses overt, structured, non participant observation to investigate morning conversations in a nursing home. The researcher observed the subjects in their own environment, with their verbal and written consent without engaging in the activity himself. He is an unobtrusive bystander who does not intentionally influence the phenomenon under study (non- participant). Despite its appeal, non participant observation can present problems. Many individuals find observation too invasive and might not consent. Furthermore, there remains a possibility that the observer or the automated devices used can be a disruptive influence, which can cause Hawthorne effect (impact of the researcher’s presence on respondents’ behaviour) (Cormack, 1996). Blending in unobtrusively and also if the researcher is not a total stranger to the setting, can decrease potential validity problems due to people playing to the gallery of the researcher (Stephens, Taggart, Jones and Andrew, 1998). In the case of Wadensten’s (2005) study the researcher had been a regular visitor for some time and was familiar with the staff and residents. For this reason he opted not to use a recorder.
Unstructured non-participant observation is used in qualitative research, while structured observation is used in both quantitative and qualitative depending on the aim of the study. Unstructured observations are recorded (as field notes, audio or video), transcribed verbatim then analysed. This method involves the collection of large amounts of information since there are few restrictions. In the assigned study, the researcher records qualitative data using structured observations. A tool, in this case, a checklist or observation schema, prepared by the researcher himself is used to record the behaviour under study, by ticking off the item as it is demonstrated. This systematic observation minimizes, possibly eliminates, the variations that might arise from different observers and facilitate categorization (Bell, 2005). It makes data collection more straight forward, limiting subjectivity and observer bias, while enhancing validity and reliability (Gerrish and Lacey, 2010). However, it must be noted that over simplistic checklists can be vague and miss important detail. Conversely, highly structured observation schedules can neglect spontaneous activities. Checklists need to be validated in terms of inter-observer reliability and intra-observer reliability so as to ensure that the data being collected are consistent and replicable when used by different researchers and on different occasions (Gerrish and Lacey, 2010).
Structured observation methods are often used to study focused aspects of a setting as part of descriptive or cross-sectional studies; in fact, the given study is part of a larger one addressing interaction and communication in nursing homes. This type of data collection is not flexible like participant observation and, while making explicit what people do, does not delve into the reasons why. Wadensten’s (2005) study provides us with the variety of topics of morning conversation, but does not attempt to enlighten us as to possible reasons for this behaviour. Thus, non participant observation can be used to inform on the science of health care, but not to study the art of a profession (Cormack, 1996).
Furthermore, observational methods of data collection are time-consuming, hard work and expensive and only a small number of people can be observed at one time. Barbro Wadensten (2005) took six weeks to carry out the study and only observed one or two subjects at a time.
Quantitative self-reports, obtaining the respondents’ perceptions of themselves and of their world, is a very popular data collection method used in most areas of the social sciences (Schwarz, 1999). Written questionnaires, face-to-face interviews, telephone interviews, and internet surveys are methods of collecting self-report data. Li and Lambert (2008), considered the Chinese cultural perspective and thus, opted to use the most anonymous, unthreatening and confidential method available to them, self-report questionnaires, to study coping skills, stressors, job satisfaction and demographic characteristics in of ICU (Intensive Care Unit) nurses.
Questionnaires, telephone and internet surveys are less expensive than other data collection techniques. They generate a large amount of data in a relatively short period of time and allow collection of data despite the geographical distance separating the researchers and the participants (Cormack, 1996) - 102 respondents from four hospitals in two cities participated in Li and Lambert’s study (2008). Although data collected using questionnaires may lack depth and context, it is a rich means for correlation (Li and Lambert make correlations between the findings in their four questionnaires) providing a clear picture of the processes under study.
Structured face-to-face interviews have the advantage of higher response rates (Edwards et al. 2002), but are more costly in terms of time and energy. Low response rates is the key difficulty in questionnaires, especially in those which ask for personal and idiosyncratic information. This can affect the significance of the results.
The presentation, wording, and sequence of questions in self-report questionnaires and interviews can be critical in gaining the necessary information. Strong evidence of the validity and the reliability of the assessment scales used is essential (Gerrish and Lacey, 2010), in fact in the assigned study the researchers use three specially-designed scales (Brief Cope (Carver,1997), Nursing stress scale (Gray-Toft and Anderson, 1981) and Job satisfaction survey (Spector, 1997)), so that bias can be measured and controlled more efficiently. The authors, however, do acknowledge the fact that these instruments were developed for a western culture and this might be a limitation in an Asian context. This outlines the sensitivity of questionnaires themselves and the necessity to take all aspects into consideration to minimize the factors which contribute to invalidity.
Another potential validity problem can arise due to the respondents not being truthful despite the fact that participation is voluntary. Individuals can distort their answers by reporting in a socially accepted way or in a way they would like to present themselves. This can be corrected by lie scales or other such controls, but, some might genuinely have misperceptions about themselves or are unable to provide the level of detail, or understand the meaning of specific questions (Cormack, 1996). On the same prejudicing lines, relatives or friends of the participants might have assisted or completed the questionnaire for them. In the study assigned the researchers made themselves available by telephone and email to minimize the possibility that questions were misinterpreted. This step was taken in view of the fact that the questionnaires were translated from English to Chinese (Li and Lambert, 2008).
Both non-participant observation and self report methods of data collection aim at simplifying complex processes while bringing us closer to the real world of intricate human phenomena. It is up to the researcher to exploit the strengths, while remaining vigilant to the limitations.
If you are the original writer of this essay and no longer wish to have the essay published on the UK Essays website then please click on the link below to request removal: