Qualitative epidemiology II

Heterogeneity is a quantitative and qualitative issue in epidemiology in more than one way.  Two grown humans can have the same total height, but one may be long-legged and the other long in the trunk.  That is a quantitative matter, as are the different growth paths—one person may have shot up to their adult height in their early teens; the other a late bloomer.  On a more qualitative level, girls’ arrival at puberty when they stop growing much taller  may be influenced, although this picture is not at all clear, by diet, environmental hormones, a sense of insecurity in an unstable family context, and so on.  Moreover, the significance of height for life opportunities and experiences can depend on gender, sport participation, biases favoring the tall in employment, politics, relationships, sports, self-esteem, etc.  Of course, many of the factors just mentioned can be converted into a quantitative measure but life course epidemiology typically relies on larger data sets that lack such measures.

The possibility of hybrid qualitative-quantitative analysis in which heterogeneity is not suppressed is indicated by the work of the medical sociologists, George Brown and information sociologist, Brenda Dervin.  In both cases a variety of responses are coded in ways that permit quantitative analysis, but the interview format in each case allows variety to be expressed and explored.

For more than forty years, George Brown has studied social influences on mental illness by interviewing subjects about life events and experiences, tracing back from the recent past when there may have been a crisis to a longer span of possibly relevant antecedent conditions.  The interviewer is able to explore an event, such as, loss of a spouse, and inquire about its meaning—was this an unexpected loss of the primary breadwinner for the family or the end of a long period of decline in health?  The interviews are then coded by a different group of researchers who are not informed about the final status of the subject, that is, the coding is blind to whether the subject ended up with a clinical mental illness or not.  The coding can subsume a diversity of kinds of experience into categories of threat or disruption to one life.  In that way the heterogeneity of life experiences is explored through qualitative interviews but the coded outcomes are amenable to quantitative analysis.

The basic mode in Brenda Dervin’s research is to ask subjects to recount in fine detail what happened over the course of some information-seeking situation, such as going to a clinic to receive the results of a cancer biopsy.  For each point of the chronology the subject is asked, in effect, what helped and what hindered them.  Again the responses are later coded in a way that allows a diversity of experiences to be subsumed in a smaller set of synthetic categories amenable to quantitative analysis.


One thought on “Qualitative epidemiology II

  1. Pingback: Qualitative epidemiology III « Intersecting Processes

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