In-depth Interviewer Effects: Mitigating Interviewer Bias

The following is a modified excerpt from Applied Qualitative Research Design: A Total Quality Framework Approach (Roller & Lavrakas, 2015, pp. 83-84).

The outcome of a qualitative in-depth interview (IDI) study, regardless of mode, is greatly affected by the interviewer’s conscious or unconscious influence within the context of the IDIs—that is, the absence or presence of interviewer bias. The interviewer’s Interviewer Effects-Biasdemographic characteristics (e.g., age, race), physical appearance in face-to-face IDIs (e.g., manner of dress), voice in face-to-face and telephone IDIs (e.g., a regional accent), and personal values or presumptions are all potential triggers that may elicit false or inaccurate responses from interviewees. For example, imagine that an IDI study is being conducted with a group of public school teachers who are known to harbor negative feelings toward the district’s superintendent but who express ambivalent attitudes in the interviews as the result of the interviewers’ inappropriate interjection of their own personal positive opinions. In this way, the interviewers have caused the findings to be biased. In order to minimize this potential source of distortion in the data, the researcher can incorporate a number of quality enhancement measures into the IDI study design and interview protocol:

  • The IDI researcher should conduct a pretest phase during which each interviewer practices the interview and learns to anticipate what Sands and Krumer-Nevo (2006) call “master narratives” (i.e., the interviewer’s own predispositions) as well as “shocks” that may emerge from interviewees’ responses. Such an awareness of one’s own predispositions as an interviewer and possible responses from interviewees that might otherwise “jolt” the interviewer will more likely facilitate an uninterrupted interview that can smoothly diverge into other appropriate lines of questioning when the time presents itself. In this manner, the interviewer can build and maintain strong rapport with the interviewee as well as anticipate areas within the interview that might bias the outcome.

For example, Sands and Krumer-Nevo (2006) relate the story of a particular interview in a study among youth who, prior to the study, had been involved in drug use and other criminal behavior. Yami, the interviewer, approached one of the interviews with certain assumptions concerning the interviewee’s educational background and, specifically, the idea that a low-level education most likely contributed to the youth’s illicit activities. Because of these stereotypical expectations, Yami entered the interview with the goal of linking the interviewee’s “past school failures” to his current behavior and was not prepared for a line of questioning that was not aimed at making this connection. As a result Read Full Text

Qualitative Research: A Collection of Articles from 2016

qr-2016-collection-headerMany of the articles published in Research Design Review in 2016 were dedicated to qualitative research for the simple reason that qualitative researchers are faced with myriad issues when attempting to achieve quality outcomes, and yet there is relatively little discussion about the quality standards by which to guide their research.  RDR attempts to fill this void by focusing on the unique attributes of qualitative research and how they serve to define the optimal approaches to conducting qualitative research that is credible, analyzable, transparent, and useful.

Qualitative Research: A Collection of Articles from Research Design Review Published in 2016 is a compilation of the 17 RDR articles that were published in 2016 devoted to qualitative research.  These 17 articles include articles on:

Paying Attention to Bias in Qualitative Research: A Message to Marketing Researchers (& Clients)

Researchers of all ilk care about bias and how it may creep into their research designs resulting in measurement error.  This is true among quantitative researchers as well as among qualitative researchers who routinely head-in-the-sand-2demonstrate their sensitivity to potential bias in their data by way of building interviewer training, careful recruitment screening, and appropriate modes into their research designs.  It is these types of measures that acknowledge qualitative researchers’ concerns about quality data; and yet, there are many other ways to mitigate bias in qualitative research that are often overlooked.

Marketing researchers (and marketing clients) in particular could benefit from thinking more deeply about bias and measurement error.  In the interest of “faster, cheaper, better” research solutions, marketing researchers often lose sight of quality design issues, not the least of which concern bias and measurement error in the data.  If marketing researchers care enough about mitigating bias to train interviewers/moderators, develop screening questions that effectively target the appropriate participant, and carefully select the suitable mode for the population segment, then it is sensible to adopt broader design standards that more fully embrace the collecting of quality data.

An example of a tool that serves to raise the design standard is the reflexive journal.  The reflexive journal has been the subject (in whole or in part) of many articles in Research Design Review, most notably Read Full Text