Preliminary analysis via attention to fragments
Initial cataloging and review of the data can consume inordinate amounts of time. In those instances when this activity proves difficult to rein in, Heath, Hindmarsh, and Luff (2010) suggest setting the cataloging activity aside and momentarily shifting to preliminary analysis of a fragment of the data. Analysis of a fragment of interest that is ten seconds or less in length will likely yield great insight regarding the complexity of the phenomenon under investigation, and can help the researcher refocus attention on the essential factors of interest to the study, which will then refine the cataloging process.
Decisions regarding the nature of preliminary analysis to undertake will naturally reflect the researcher’s beliefs over what constitutes relevant and useful information to record.
Some options:
- Transcription of talk
- Transcription of action (through such frameworks as those recommended by Birdwhistell (1970) or more recently by Heath, Hindmarsh, and Luff (2010)).
- Analysis of the ways in which participants’ actions are arranged sequentially, which is to say, “They are designed with regard to, and occasioned by, prior actions and they form the foundation to subsequent actions” (Heath, Hindmarsh, & Luff, 2010, p. 68). This is particularly appropriate in studies with a sociological focus.
- Consideration of what occasioned particular choices in particular moments (“Why that now?” as suggested by Schegloff & Sacks, 1973, cited by Heath, Hindmarsh, & Luff, 2010, p. 83).
Through such investigation via transcription, the researcher can typically discover aspects of the phenomena of interest that would not otherwise be detected or investigated.
Guiding Questions
Video research is particularly easier to navigate when guiding questions or research questions are created. It is important to identify what phenomena are to be studied and what kinds of questions can be generated to be able to answer those questions (Barron & Engle, 2007). Video research then becomes more manageable because any of the information that is irrelevant and not pertinent to the study can be filtered. Moreover, video may also contain information that others may find useful in the future to answer other research questions. Though the data drawn from video may only be pertinent to a researcher’s study, others can use the same video to draw other data for other studies.
Progressive refinement of hypothesis
Knowing that there may be unexpected phenomena, the hypothesis can be further specified as more video is collected and analyzed (Engle, Conant, & Greeno, 2007). As the progressive hypothesis develops, it can also be framed and supported by other parts of the data generated from the videos. In addition, repeated social or group practices of viewing the data may strengthen interpretations. Having this developing hypothesis contributes to a great robustness and validity for the study (Barron & Engle, 2007).
References
Barron, B., & Engle, R. A. (2007). Analyzing data derived from video records. In S. J. Derry (Ed.), Guidelines for video research in education: Recommendations from an expert panel (pp. 24-33). Chicago, IL: Data Research and Development Center.
Birdwhistell, R. L. (1970). Kinesics and context: Essays on body motion communication. London: Allen Lane.
Engle, R., Conant, F. R., & Greeno, J. G. (2007). Progressive refinement of hypotheses in video-supported research. In R. Goldman, R. Pea, B. Barron, & S. J. Derry (Eds.), Video research in the learning sciences (pp. 239 –254). Mahwah, NJ: Lawrence Erlbaum Associates.
Heath, C., Hindmarsh, J., & Luff, P. (2010). Video in qualitative research. Thousand Oaks, CA: Sage Publications.
Schegloff, E. A. & Sacks, H. (1973). Opening up closings. Semiotica, 7, 289-327.