Observation as a Research Method

Observation as a Research Method

In addition to interviewing, I also sought to gather data about the Justice Citizens through participant observation. Observation has a long history within qualitative research, originally being associated with ethnography and occurring within settings that already exist. Observation in educational research has been drawn from anthropology, and is either unobtrusive observation or, more likely, participant observation. In the 1950s, ethnographers started exploring educational sites through observation. Lichtman writes, ‘Researchers immersed themselves in classrooms and conducted extensive observations and took field notes’. Interest in this approach to gathering data increased, as well as the number of different sites that were available for research. By the dawn of the 21st century, online classrooms and forums, as well as more traditional settings, were being investigated through participant observation.

The main reason that observation was embraced so quickly by qualitative researchers is that it allowed for a deeper and more nuanced understanding of human behaviour. When coupled with interviews, observation allowed researchers to understand the subjects of their research at a much deeper level. Lichtman writes, ‘Observing humans in natural settings assists our understandings of the complexity of human behaviour and interrelationships among groups’ (p 165). One of the most commonly cited qualitative researchers, Geertz, argued that such observations provided the opportunity for ‘thick description’  (Geertz, 1973). In support of this argument, Cohen, Manion and Morrison (p 396) write, ‘The use of immediate awareness, or direct cognition, as a principal mode of research thus has the potential to yield more valid or authentic data than would otherwise be the case with mediated or inferential methods.’ Cohen, Manion and Morrison also identify other advantages. Firstly, by observation, the researcher gets to see what people do, rather than what they say they do. This means that there is an added level of truth in observations that might be obscured in data gathered through other means. In addition, observations are also less intrusive and time-consuming than an interview.

However, there are issues related to observation of which researchers need to be aware (Lichtman 2010). The first is that any observation necessitates an interaction between the facts and the interpretation of those facts. Cohen, Manion and Morrison (INSERT REFERENCE) argue that observers produce data on a continuum between the facts of the observation and the interpretation that is applied to those facts

There is also the question of the choice of the subject: the researcher must choose who is to be studied, and in what situations they are to be studied. For example, a researcher in an educational setting must choose whether to observe students or teachers – and whether to observe groups in formal or informal groups. Linked to these decisions is the requirement to gain access. Schools and other educational institutions often have stringent rules about access to students, and what can be discussed with them. Other, less formal rules, might dictate access to teachers or other staff, who might feel uncomfortable speaking with a researcher in their place of work.

There are also decisions to be made about what is to be studied. An observer cannot possibly document everything that he or she sees, even in a narrowly defined space, and nor should he or she try to do so – much of what will be observed might not be relevant to the researcher’s interests. Although the advent of cheap video technology has allowed researchers to film settings for observation, the use of such technology brings its own challenges – some subjects might feel that being filmed – even with their consent – is more obtrusive than having a researcher in the room in person, and thus their behaviour might be significantly changed. Morrison (1993) details that a researcher should take note of four categories: the physical setting, the human setting, the interactional setting and the program setting.

Another decision that a researcher needs to make about observation is to decide on both the frequency and duration of the observation. A central tenet of qualitative researcher validity is prolonged exposure; that is, if a researcher is to draw valid conclusions from his or her observations, then it is necessary for them to spend lengthy periods of time with the subjects they are researching. In the case of early anthropologists, this often meant years. Now, however, the length of a research project might be limited to days or weeks. In the case of the Justice Citizens project, the length of the project extended over 6 months. Although this was quite a long time, I was, at first, concerned that I would not have enough exposure to the students as I was only seeing them for an hour each week. However, my concerns were allayed because I spent a lot more time with the students than I had expected: in addition to scheduled class times, we spent time before and after school, during lunchtime and recess working on the films. This meant that I was in regular, frequent contact with the students, in varying degrees of formality, which, I think, added to the trustworthiness of the research.

The final issue for the researcher to consider is what his or her role might be. This becomes particularly important when the researcher must decided to be either a participant observer or simply an uninvolved observer. Participant observers take part in the actions of the research subjects as much as possible; they seek to learn about the lived experiences that they are studying by undertaking those experiences in the same way as the participants. Of course, such an experience is impossible to fully reproduce, and as such participant observations are often artificial to some degree. Gold (1988) argues that all observers exist on a continuum between that of a complete participant and complete observer. Some category within this continuum are ‘participant as observer’ and ‘observer as participant’.  The traditional role of an observer is to be non-interventionist, but this was not appropriate in the context of Justice Citizens.

With Justice Citizens, this decision about the role was made even more complex by my role as a teacher in the class. Although I did not act as a participant observer in the sense that I was a student in the class, I was nevertheless a participant in the class as a teacher. The nature of the project – that it was YPAR and not a more formal classroom setting – did blur some of the lines between my role as a teacher and the role of the students. In some ways, I felt that I was acting more as a facilitator of learning experiences than as a teacher; the newness of this style and the experiences that I organised for the students allowed me to participate in the project as well.

Lichtman (2010) provides some excellent advice for the novice researcher about how to conduct an observation. She cautions that preparation is vital. A good observation is not simply a case of ‘going in and having a look’ (INSERT REFERENCE). Rather, there should be a focus for the observation. Lichtman suggests choosing 3-5 areas to focus upon. For example, a researcher examining a teacher’s practice might look for verbal commands, non-verbal signals, and the clarity of instructions given.