Importance of Nonresponse in Qualitative Research

Nonresponse and non-response error is more than a quantitative issue.  While qualitative researchers may shudder at the thoubiasght, the typically-ignored impact of nonresponse is just as important in the qualitative realm.  Why is nonresponse in qualitative research important?  Because we are conducting qualitative research.  Not qualitative let’s get a few warm bodies around the table for our face-to-face focus group, but actually research methods that, like all research, demand certain protocols that address potential biasing effects.  One of these is nonresponse.  The warm bodies in our group discussion may make the moderator and client observers feel great – Thank goodness, someone showed up! – but the uncomfortable reality is that the people who chose not to participate – or were never contacted by a recruiter and asked to participate in the first place – greatly affect our research outcomes.  Indeed, the trajectory of a group discussion has as much to do with the people sitting around a table as it does with those who aren’t there.

Who are the folks that the recruiter glossed over and never contacted on our sample list?  How are the people who elect to attend our group discussion the same or different than those who didn’t?  And, without knowing the answers to these questions, how can qualitative researchers put their research findings into any perspective?  How do they know what they think they know?

This is why, among other things, researchers need to pay more attention to gaining cooperation in their qualitative studies.  Fortunately for the qualitative researcher, there are many ways to improve cooperation and thereby decrease the threat from nonresponse.  For example, the design of a focus group study needs to carefully consider the personal circumstances of potential participants, i.e., where they are, what they do in their day, how they communicate, etc.  This should lead to design considerations such as:

  • Location, e.g., Does the researcher need to give participants a choice of location, such as an uptown or downtown facility?
  • Time, e.g., Is 6:00 p.m. really the best time for participants to meet?
  • Incentive, e.g., What type of material and/or non-material incentive “makes sense” for this participant segment?
  • Mode of contact, e.g., Are participants most likely to respond to phone, email, or the U.S. mail as way of contact and how should the researcher combine these modes to gain greater cooperation?

And, at a minimum, the recruiting screener should:

  • Communicate the purpose of the study to arouse interest in participating without introducing details that may influence participants’ feedback in the discussion;
  • Communicate a personal benefit or non-material motivation (e.g., telling suburban women that their participation in a discussion concerning the local shopping mall is their opportunity to contribute to the creation of an improved shopping environment);
  • Mention the material incentives, such as the nature of the incentive (e.g., cash, a gift card, prized tickets to a sporting event, donation to a favorite charity, etc.) and the value;
  • Identify the study’s sponsor; and
  • Provide logistical details so that showing up for the focus group – face-to-face, on the telephone, or online – is easy and uncomplicated.

4 comments

  1. Representatitiveness is always something I grapple with in qualitative research. Whenever I ask the question to other qualitative researchers, I get the response “representative of what?” In quantitative studies, the answer to this is usually very clear. And in Qualitative studies that are meant to do the groundwork for quantitative studies or where the population of interest is predefined, the answer is also relatively clear. But then there is the big qualitative gem of exploratory research, where going deep is more useful than being representative. These kinds of studies serve a different purpose, but an important one, and questions of representativeness don’t work as well for them. In these cases, the best ethics seem to be the “it is what it is and isn’t what it isn’t” ethics…

    1. Thanks, Cheryl, great comment. I would beg to differ, however, in the idea that representativeness works in some but not all cases. In all research we are interested in at least attempting to tap into people who represent our target population. Even in qual work with highly-articulate or particularly creative people (e.g., for the purpose of ideation), there is some standard we set that identifies what is representative of this group and we fully document that in our final report so the reader is clear how we defined our research ‘universe’.

      1. Margaret, I want you to know that I haven’t neglected your response. I’m working through it in the only way a blogger can!

        Here it is, in part:
        http://freerangeresearch.com/2013/06/05/representativeness-qual-quant-and-big-data-lost-in-translation/

        But it’s been a theme ever since. It’s a BIG question! Part of it, I think, is the unit of analysis. I’ve been trained as a linguist, so I’m focusing on the level of language units, and not people…

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