BUSI 700 Week 5 Group DB Forum 3 Example

Group Discussion Board Forum 3: Thread, Group Presentation and Reply

BUSI 700 Doctor of Business Administration Orientation, Group Discussion Board Forum 3 (thread, group presentation and reply)

Thread by [Student Name]

The prompt, the group and the classmates are composites standing in for the course's own; McRaney's book and the research articles cited are real; the requirements follow Liberty's public course guide for BUSI 700.

Our prompt asked which bias from You Are Not So Smart poses the greatest danger to a doctoral student's research. My answer is confirmation bias. McRaney (2012) portrays it as the mind's habit of hunting for, noticing and recalling whatever backs a view already held, while evidence against that view slips past unexamined. Other biases in the book can mislead us about ourselves, but this one threatens the core task of a dissertation, which is to test a claim rather than defend it.

A DBA student usually arrives with a strong belief formed in practice, for example that servant leadership reduces turnover in their own company. That belief is what makes the research meaningful to them, and it is also the danger. Nickerson (1998) reviewed decades of evidence and found confirmation bias in many guises, from the way people choose which evidence to examine to the way they interpret ambiguous results, and he noted that scientists are not immune; they too favor hypotheses they hold and weigh supporting data more heavily.

The bias becomes most damaging when it joins another that McRaney describes, the Texas sharpshooter fallacy, in which someone finds a cluster in random data and treats it as a target hit. A student who runs dozens of analyses on a survey dataset will almost certainly find some significant relationships by chance. Confirmation bias then makes the ones that fit the student's belief feel important. Kerr (1998) named the final step HARKing: the researcher writes up an idea suggested by the results as though it were the prediction the study set out to test. The danger is not dishonesty; it is a sincere researcher who no longer remembers which hypothesis came first.

Awareness alone does not cure this, so I propose three safeguards. First, write and date the research questions and hypotheses before collecting data, and file them with the dissertation chair. Second, search deliberately for studies that found no effect or the opposite effect and give them a section of the literature review. Third, ask one committee member to argue against the hypothesis at the proposal defense. Proverbs observes that the first to state a case seems right until another comes and examines him (English Standard Version Bible, 2001/2016, Proverbs 18:17), which is exactly the role a skeptical reader plays.

The stakes are higher in a DBA than in many research settings because the student's employer is often the research site. A finding that flatters a program the student helped design is welcome news at work, which adds a professional incentive to the psychological one. Confirmation bias then has an ally in self-interest, and only procedures fixed in advance can hold both in check.

Group Presentation (posted to the class-wide forum)

Group 2 was asked which bias from McRaney's book most threatens doctoral research. Members proposed confirmation bias, the Texas sharpshooter fallacy, the Dunning-Kruger effect and the illusion of control; after discussion, we agreed on confirmation bias, because it works through the others, shaping which data we collect, which patterns we notice and how confident we feel. Research supports the concern: reviews of confirmation bias show it affects trained scientists, and the practice of hypothesizing after results are known turns chance findings into apparent support. We recommend three safeguards for DBA students: register hypotheses with the chair before collecting data, include contrary studies in every literature review and invite a committee member to argue the opposite position. Our group would welcome other safeguards from the class.

Reply to Group 1

Group 1, your case for hindsight bias as the greatest danger was convincing, especially your example of a manager who, after a failed product launch, insists the problems were obvious all along. McRaney (2012) describes how people reconstruct the past to make outcomes seem predictable, and I agree that a researcher interviewing managers about past decisions will hear many such reconstructions.

I would suggest that hindsight bias and confirmation bias work together in research. When a DBA student studies a past event, such as a merger that failed, hindsight bias in the interviewees makes the outcome look inevitable, and confirmation bias in the researcher makes the explanation that fits the student's theory seem the right one. Nickerson (1998) describes how people interpret ambiguous information to fit their existing view, and recollections of the past are about as ambiguous as evidence gets. One practical safeguard follows: wherever possible, check interview accounts against documents written at the time, such as board minutes or project plans, which record what people expected before they knew the result. Did your group consider how a researcher could separate what participants knew then from what they believe now?

References

English Standard Version Bible. (2016). Crossway. (Original work published 2001)

Kerr, N. L. (1998). HARKing: Hypothesizing after the results are known. Personality and Social Psychology Review, 2(3), 196-217. https://doi.org/10.1207/s15327957pspr0203_4

McRaney, D. (2012). You are not so smart: Why you have too many friends on Facebook, why your memory is mostly fiction, and 46 other ways you're deluding yourself. Gotham Books.

Nickerson, R. S. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2(2), 175-220. https://doi.org/10.1037/1089-2680.2.2.175