A short note on the scandal


I am glad this has been finally made public:

Academic who revealed Australian bus driver racism ‘victimised’

It has been a source of great concern for me because I care about the University of Queensland. It has been a source of great personal embarrassment because that was the place where I did both my undergraduate and graduate degrees.

The findings of the study when it first emerged were to some extent consistent with my, my family’s, and my friend’s experiences while living in Brisbane. I’m not sure that the university handled this well. This deserves nothing less than a parliamentary enquiry (not in the Queensland parliament).



11 Responses to "A short note on the scandal"
  1. The University of Queensland requires research involving deception to be reviewed at the institutional level: http://www.uq.edu.au/economics/ethics-committee-review-guidelines-and-instructions. From my reading of the news reports, the bus bias study was not reviewed at the institutional level and thus did not receive the proper ethical clearance. My understanding is that research cannot receive after-the-fact clearance, so I’m not sure what the University of Queensland can do other than not permit the study to be published.

    • I am not sure if you read the article carefully…but the fact is that University of Queensland representatives (i.e. the research coordinator/officer) within the school where the PhD student was at the time advised and approved/cleared the study to go ahead…hence why it did not go the institutional level. If it was decided differently, it would have had to gone to that level.

      So the University of Queensland made a decision and now are not admitting that they made that decision/approval themselves but are instead punishing the researchers.

      • My reading is that the school of economics officer has no authority to make that decision. See: “Academic staff research requires clearance from central university committees.” See also: “Irrespective of the level of risk involved, research of a kind falling under the following chapters of the National Statement must be reviewed by one of the institutional ethics committees: Deception and concealment (Chapter 2.3.4).” Those passages indicate that review at the institutional (university) level was required. The fact that a school of economics official made an error does not remove the fact that the research lacks the required institutional-level approval.

        • Well then I guess it has to be asked: why did the school of economics officer have the authority to make and advise on that decision?

          Which again points back to the procedures and roles assigned to the research coordinator by the hierarchy and management of the university. Their procedures do not seem to be very clear and well thought out…I mean, you can put whatever you like on an internal ethics website, but if you are doing and advising differently, then this is not right especially for students and your research staff.

          • I’m guessing that there might have been confusion because of the PhD student on the study, given that there appears to be different standards for students for some reason. I’d also guess that there might have been a lack of awareness that the study involved deception. In any event, like you indicated, the requirements are confusing and led to problems with what appears to be a valuable high-quality study.

  2. For all I can see, the study did not involve deception. I should mention that I have thought/written about this issue at length and have argued against the use of deception (e.g., http://spq.sagepub.com/content/71/3/222.refs) . My take: The Frijters – Mujcic testers had insufficiently funded cards (part of the design of the study obviously), they subsequently asked the driver to let them ride to a destination a kilometre away. That ride was part of the study. There is no false statements there. The one interesting methodological question is whether informed consent was needed ex ante (not an option here for the obvious reason), or ex post debriefing. People as prominent as List have argued, in Science no less, that for studies like this there is a case to be made that debriefing is not necessary.

  3. Hi Andreas. I think that the presence of deception in this case is something that reasonable people can disagree about, which seems to suggest the need for institutional review. I’d think of it this way: if I ask you for a favor *that I do not need* in order to study your reaction and then write a research article about what you and others did in response to my favor request — because I plan to test for possible racism in a specific group of you and your colleagues — then I certainly think it would be reasonable for you to feel deceived. Or think of it this way: the experiment would not have worked as successfully if the bus drivers had complete information and thus knew that the confederates were conducing research. But maybe there’s a formal definition of deception that ethics review committees use and I am unaware of.

    • L.J. I disagree. Lest the testers had indeed money on them and / or valid fare cards , then I just don’t see that deception was used. The question of ex-ante informed consent (here obviously not practical) or ex-post debriefing is another issue. As mentioned, people like List (likely Nobel Prize winner for his work) have argued that neither is needed under certain conditions. I would argue the Frijters-Mujcic study fit those criteria.

      • Hi Andreas. The tester told the bus driver: “I do not have any money, but I need to get to the [X] station.” I suppose in a strict sense it is true that the tester needed to get to the [X] station, but I’d argue that deception is involved because the bus driver did not appreciate the true reason why the tester needed to get to the [X] station.

        Suppose that the testers visit people’s homes and say: “I need to use your bathroom.” In reality, the testers need to use the bathroom only because the project manager wants the testers to list everything in the homeowner’s medicine cabinet. In there any deception in this bathroom experiment?

        I’d say that the bus and bathroom experiments are the same with regard to deception because in each case there is general expectation for each “need” but the implied reason for the “need” is not the true reason for the “need.”

        Of course, the bus and bathroom studies can be distinguished in that the bathroom study involves a privacy violation, but the presence of a privacy violation should not impact the presence of deception.

        I think that it is reasonable to disagree regarding deception in the broad parameters of the field experiment. However, some testers wore a replica of the Australian army uniform. Would you agree that the experiment would have involved deception if a replica army uniform had been worn by a tester who was not in the army?

  4. This is the piece by List that Andreas above is referring to- maybe it will make it more clear to some of those here unfamiliar with the exact aim and usefulness of a natural field experiment (where there is absolutely no added harm to anyone since you are replicating everyday life, and in the Mujcic/Frijters case; the topic being studied is arguably one of the most important social issues around):

  5. Hi NFE. I’m not sure why the usefulness of the study or the lack of harm caused by the study is relevant. The study was either conducted with or without the required ethics approval based on university policies. The points that you mention would be relevant for the subjective question of whether the study *should have* received ethics approval; however, the points that you mention do not appear relevant for the objective question of whether the study *did* receive the required ethics approval.

%d bloggers like this: