Mixed methods

These are just some notes on a workshop on mixed methods in which we were looking at An Innovative Mixed Methods Approach to Studying the Online Health Information Seeking Experiences of Adults With Chronic Health Conditions by Joanne Mayoh, Carol S Bond, and Les Todres. I doubt that quantitative research will be that appropriate within what I’m trying to do, but I have been looking at things like pronoun use over time in some of my transcripts of maths talks and finding it fascinating, so getting some kind of a view of the considerations seemed wise.

To reform traditional notions of complementarity into a paradigmatically considerate approach, Sale et al. (2002) suggest using a mixed methods approach that acknowledges and respects paradigmatic differences, while still allowing for the combination of different methods within a single study. They note the importance of accepting that qualitative and quantitative research will inevitably look at different phenomena within the same research area, and therefore suggest that that the explicit phenomenon examined by each method must be explicitly labeled to state and emphasis the paradigmatic differences in qualitative and quantitative research. The utilization of this approach within the current study allowed paradigmatically opposing methodologies to be combined within a single project in order to provide a multidimensional understanding of a complex phenomenon,
while still honoring epistemological and ontological differences.

Formats involving mixed methods:

  • Exploratory: qual followed by quan
  • Explanatory: quan followed by qual
  • Concurrent triangulation – both provide a view on the same topic
  • Embedded mixed methods

A good framework for phenomenological research:

Giorgi’s Descriptive Phenomenological Method
Giorgi and colleagues (Giorgi, 1985, 2009; Giorgi & Giorgi, 2004) have developed a systematic procedure for analyzing descriptive phenomenology that has become popular with researchers within a wide range of disciplines, such as nurses, sociologists, and anthropologists. This method was adopted as it provides clear and logical stages that must be conducted to achieve a full descriptive–phenomenological analysis. These are the following:
1. Obtaining rich life-world descriptions
2. Reading the transcriptions thoughtfully to get a narrative sense of the text as a whole piece
3. Dividing the descriptions up into units, signifying changes in meaning—“meaning units”
4. Expressing the sense of each meaning in a general manner
5. Developing a “structure” that integrates the common meanings throughout all lifeworld descriptions
6. Opening out the structure and elaborating on the common themes from within by using original participant data to develop the richness of the analysis; it also provides an opportunity to discuss the unique ways that each theme was experienced by participants

A couple of notes:

The idea is that qualitative research is hypothesis-generating, while quantitative is hypothesis-testing.

Descriptive statistics can be given as simple percentages, while correlations need p-values.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s