Final essays

Grounded theory is an approach to data analysis that uses theoretical sampling.

Voicing is a characteristic of narrative design.

Local knowledge and situated lives are characteristics of ethnographic design.

By leaving out triangulation, member checks, a peer debriefer, memos, and fieldnotes, your study will be weak in terms of external validity.

Grounded theory is oriented towards action and process whereas interpretivism would have the researcher spend more time with the data, returning to it over and over as assertions develop. Grounded theory is geared to build a theory from interpretations of the data by the researcher's background of literature, personal experience, and interactions with the data. The product is an inductively derived theory from the phenomena it represents. The theory must meet four criteria to be considered valid: fit, understanding, generality, and control. Grounded theory is intended to work on a "best fit" basis rather than exact fits with similar cases. Readers look for credibility of the data, adequacy of the process, and the empirical grounding of the research findings; empirical grounding includes generation of systemically related concepts, range of variation and specificity built into the theory, and considerations of process. There seems to be a general theme of flexibility when working with grounded theory. With interpretivism, instead of drawing conclusions from the data, the data is is used to reinforce assertions. The product of interpretivism is a written report consisting of empirical assertions, analytic narrative vignettes, quotes from field notes, interviews, data reports, interpretive commentary framing, theoretical discussion, and a report of the natural history of inquiry in the study. The grand idea behind the report is to allow the reader to experience the setting, survey the range of evidence, and they allow the reader to consider the personal basis of the author's perspective throughout the study. With regards to validity, Erickson calls for deliberate searches for disconfirming evidence, avoiding the problem of premature typification.

Trustworthiness: Conform to standards of acceptable and competent practice with ethical conduct.

  1. Triangulation - multiple data sources ensures a complete study
  2. "Being there" - a long time with participants gives a better perspective instead of a snapshot
  3. Participant validation - member checks; get more information from the participants after the initial findings
  4. Using critical friend - peer debriefer
  5. Community of practice - using colleagues to bounce ideas off of

Four factors affect the credibility of a study for cross-group comparisons: selection effects, setting effects, history effects, and construct effects (LeCompte and Goetz). Morse recommends investigator responsiveness, methodological coherence, theoretical sampling and sampling adequacy, and active analytic stance, and saturation as strategies for ensuring rigor. Establishing validity includes determining the extent to which conclusions represent reality and assessing whether constructs represent or measure categories of human experience.