1/1/2023 0 Comments Hyperresearch codingRegardless of the study's research method, several commonalities exist among methods used in qualitative data analysis. Making conceptual or theoretical coherence (linking the findings into an overarching "how" and "why" of the phenomenon under study) Building a logical chain of evidence (validating each of the relationships identified)ġ3. Finding intervening variables (discerning other variables that may link findings together)ġ2. Noting relationships between variables (depicting the relationships between the findings)ġ1. Factoring (generating words to express common findings)ġ0. Subsuming particulars into the general (using a higher level of abstraction)ĩ. Partitioning variables (breaking down the themes into smaller units)Ĩ. Making contrasts or comparisons (comparing sets of things)ħ. Counting (noting that something is happening a number of times)Ħ. Making metaphors (using a literary device in which different things are compared to make sense of the experience)ĥ. Clustering (grouping together things that seem to share characteristics)Ĥ. Seeing plausibility (realizing that the finding or conclusion sounds true or makes sense)ģ. Noting patterns and themes (repetitive or recurring patterns among many separate pieces of data)Ģ. I use this blog to explore topics in research, with particular focus on qualitative and mixed methods.List the following 13 tactics for drawing meaning from the data (pp. I am also the current Editor-in-Chief of the open access Ohio Journal of Public Health. I teach and do research in the College of Public Health at Kent State University and I am the Co-coordinator of the Kent State Graduate Certificate in Qualitative Research. Additionally, I have had some experiences with order of entry in regression and in some cases the difference is there but is minimal it seems like the themes that run through things float to the top one way or another. On the other hand, my practical nature argues for parsimony (as did one of my favorite statistics professors, and as did Creswell in his "Five Approaches" book (Sage Pubs), and to retain the focus on data reduction. But what about covariance? Am I missing some of it? Hyperresearch coding code#So obviously with these excerpts, I am first looking over my reduced code list for a match - before creating something new - and when there is a match, I'm assigning that category. This can create the illusion of greater importance. I was cautioning my students about order of entry of variables because the earliest entered variables will 'gobble up' as much of the variance as possible, even when there is substantial covariance with other variables. This thought process led me to the idea of statistical (regression) model building, which I lectured on just a few weeks ago. This led me to question whether the patterns really are there or whether I am just more likely to see patterns. It was in doing these that I was struck by how well the excerpts seem to match in one of the existing dozen (actually 13) categories. However, there was one set of first cycle codes left that needed to be placed into the second cycle categories. After going through this process, I was left with about a dozen categories, most of which had child or subcategories. I learned quickly enough (well, not quite quickly enough because I eliminated many first cycle codes through 'merging' in Dedoose) that I could drag the codes and drop into another to create a parent/child/grandchild, etc. I started to cluster the codes and actually went back to my default work style with Word. (This reminded me of the people who post on some of the qualitative listservs about having hundreds of themes.) However, I believe the very nature of a theme prevents the occurrence of hundreds or even dozens in most projects a theme in my opinion should be relatively broad. We made a very large number of categories or second cycle codes - well more than 100. We started with lots of first cycle codes - pages and pages - that we uploaded into Dedoose and treated as excerpts.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |