Backroom Research

It's clear from this week's reading that there are pros and cons for both highly structured and lightly structured inquiry methods.  Highly structured methods often result in unambiguous data collection, making processing and analyzing more straightforward and less time-consuming, a definite benefit for the researcher. However, as Knight expresses in chapter three, highly structured methods also risk a limited scope that can lock respondents into the researcher's theory. Conversely, lightly structured methods can solve some of these problems; the informant is able to speak freely without as many methodological constraints, and this could potentially produce responses rich in context. Of course, the data is difficult to analyze and responses are still potentially incomplete.

A particular research question is likely better suited to one method over the other, and so in certain circumstances, researchers would not have to grapple with selection. Still, I would argue that most research that requires hard statistics could do with a little context, something that lightly structured methods seem to enable. An example: the recent census over-reported the number of married gay couples because a high number of same-sex, platonic roommates mistakenly indicated they were married (CBC article here). This is probably due to faulty design of the questionnaire, but a simple open-ended question regarding the nature of the relationship would have helped to provide some context. Not the best example, I know, but it illustrates the point.

Not being too familiar with social science research, I'm unsure whether structural hybridity is practiced. Do researchers often combine highly and lightly structured elements (for instance, a questionnaire that requires the participant to check appropriate boxes and then answer open-ended follow-up questions)? This seems like it would bridge both approaches and result in hard data with some additional context.