Open Science: Inside Peer-Review at PSPB

We submitted a ms. that showed problems with the validity of the race IAT as a measure of African Americans’ unconscious attitudes to PSPB (Schimmack & Howard, 2020). After waiting patiently for three months, we received the following decision letter from the acting editor Dr. Corinne Moss-Racusin at Personality and Social Psychology Bulletin. She assures us that she independently read our manuscript carefully – twice; once before and once after reading the reviews. This is admirable. Yet it is surprising that her independent reading of our manuscript places her in strong agreement with the reviewers. Somebody with less research experience might feel devastated by the independent evaluation by three experts that our work is “of low quality.” Fortunately, it is possible to evaluate the contribution of our manuscript from another independent perspective, namely the strength of the science.

The key claim of our ms. is simple. John Jost, Brian Nosek, and Mahzarin Banaji wrote a highly cited article that contained the claim that a large percentage of members of disadvantaged groups have an implicit preference for the out-group. As recently as 2019, Jost repeated this claim and used the term self-hatred to refer to implicit preferences for the in-group (Jost, 2019).

We expressed our doubt about this claim when the disadvantaged group are African Americans. Our main concern was that any claims about African Americans’ implicit preferences require a valid measure of African Americans’ preferences. The claim that a large number of African Americans have an implicit preference for the White outgroup rests entirely on results obtained with the Implicit Association Test (Jost, Nosek, & Banaji, 2004). However, since the 2004 publication, the validity of the race IAT as a measure of implicit preferences has been questioned in numerous publications, including my recent demonstration that implicit and explicit measures of prejudice lack discriminant validity (Schimmack, 2021). Even the author of the IAT is no longer supporting the claim that the race IAT is a measure of some implicit, hidden attitudes (Greenwald & Banaji, 2017). Aside from revisiting Jost et al.’s (2004) findings in light of doubts about the race IAT, we also conducted the first attempt at validating the race IAT for Black participants. Apparently, reading the article twice did not help the action editor of PSPB to notice this new empirical contribution, even though it is highlighted in Figure 2. The key finding here is that we were able to identify an in-group preference factor because several explicit and implicit measures showed convergent validity (factor ig). For example, the evaluative priming task showed some validity with a factor loading of .42 in the Black sample. However, the race IAT failed to show any relationship with the in-group factor (p > .05). It was also unrelated to the out-group factor. Thus, the race IAT lacks convergent validity as a measure of in-group and out-group preferences among African Americans in this sample. Neither the two reviewers, nor the acting editor challenge this finding. They do not even comment on it. Instead, they proclaim that this research is of low quality. I beg to differ. Based on any sensible understanding of the scientific method, it is unscientific to make claims about African Americans’ preferences based on a measure that has not been validated. It is even more unscientific to double down on a false claim when evidence is presented that the measure lacks validity.

Of course, one can question whether PSPB should publish this finding. After all, PSPB prides itself on being the flagship journal of the Society for Personality and Social Psychology (Robinson et al., 2021). Maybe valid measurement of African Americans’ attitudes is not relevant enough to meet the high standards of a 20% acceptance rate. However, Robinson et al. (2021) launched a diversity initiative in response to awareness that psychology has a diversity problem.

Maybe it will take some time before PSPB can find some associate editors to handle manuscripts that address diversity issues and are concerned with the well-being of African Americans. Meanwhile, we are going to find another outlet to publish our critique of Jost and colleagues unscientific claim that many African Americans hold negative views of their in-group that they are not aware of and can only be revealed by their scores on the race IAT.

Editorial Decision Letter from PSPB

Re: “The race Implicit Association Test is Biased: Most African Americans Have Positive Attitudes towards their In-Group” (MS # PSPB-21-365)

Dear Dr. Schimmack:

Thank you for submitting your manuscript for consideration to Personality and Social Psychology Bulletin. I would like to apologize for the slight delay in getting this decision out to you. Both of my very young children have been home with me for the past month, due to Covid exposures at their schools. As their primary caregiver, this has created considerable difficulties. I appreciate your understanding as we all work to navigate these difficult and unprecedented times.

I have now obtained evaluations of the paper from two experts who are well-qualified to review work in this area.  Furthermore, I read your paper carefully and independently, both before and after looking at the reviews.

I found the topic of your work to be important and timely—indeed, I read the current paper with great interest. Disentangling in-group and out-group racial biases, among both White and Black participants (within the broader context of exploring System Justification Theory) is a compelling goal. Further, I strongly agree with you that exploring whether Black participants’ in-group attitudes have been systematically misrepresented by the (majority White) scientific community is of critical importance.

Unfortunately, as you will see, both reviewers have significant, well-articulated concerns that prevent them from supporting publication of the manuscript. For example, reviewer 1 stated that “Overall, I found this article to be of low quality. It argues against an argument that researchers haven’t made and landed on conclusions that their data doesn’t support.” Further, reviewer 2 (whose review is appropriately signed) wrote clearly that, “The purpose of this submission, it seems to me, is not to illuminate anything, really, and indeed very little, if anything, is illuminated. The purpose of the paper, it seems, is to create the appearance of something scandalous and awful and perhaps even racist in the research literature when, in fact, the substantive results obtained here are very similar to what has been found before. And if the authors really want to declare that the race-based IAT is a completely useless measure, they have a lot more work to do than re-analyzing previously published data from one relatively small study.”

See Reviewer 2’s comments and my response here

My own reading of your paper places me in strong agreement with the reviewer’s evaluations. I am sorry to report that I will not be able to accept your paper for publication in PSPB.

The reviewers’ comments are, in my several years of experience as an editor, unusually thorough and detailed. Thus, I will not reiterate them here.  Nevertheless, issues of primary concern involved both conceptual and empirical aspects of the manuscript. Although some of these issues might be addressed, to some degree, with some considerable re-thinking and re-writing, many cannot be addressed without more data and theoretical overhaul.

I was struck by the degree to which claims appear to stray quite far from both the published literature and the data at hand. As just one example, the section on “African American’s Resilience in a Culture of Oppression” (pp. 5-6) cites no published work whatsoever. Rather, you note that your skepticism regarding key components of SJT is based on “the lived experience of the second author,” which you then summarize. While individual case studies such as this can certainly be compelling, there are clear questions pertaining to generalizability and scientific merit, and the inability to independently validate or confirm this anecdotal evidence. While you do briefly acknowledge this, you proceed to make broad claims—such as “No one in her family or among her Black friends showed signs that they preferred to be White or like White people more than Black people. In small towns, the lives of Black and White people are more similar than in big cities. Therefore, the White out-group was not all that different from the Black in-group,” again without citing any evidence. I found it problematic to ground these bold claims and critiques largely in anecdote. Further, this raises serious concerns—as reviewer 2 articulates in some detail—that the current work may distort the current state of the science by exaggerating or mischaracterizing the nature of existing claims.

Let me say this clearly: I am strongly in favor of work that attempts to refine existing theoretical perspectives, and/or critique established methods, measures, and paradigms. I am not an IAT “purist” by any stretch, nor has my own recent work consistently included implicit measures. Indeed, as noted above, I read the current work with great interest and openness. Unfortunately, like both reviewers, I cannot support its publication in the current form.

I would sincerely encourage you to consider whether the future of this line of work could involve 1. Additional experiments, 2. Larger and more diverse samples, 3. True and transparent collaboration (whether “adversarial” or not) with colleagues from different ideological/empirical perspectives, and 4. Ensuring that claims align much more closely to what is narrowly warranted by the data at hand. Unfortunately, as it stands, the potential contributions of this work appear to be far overshadowed by its problematic elements.

I understand that you will likely be disappointed by my decision, but I urge you to pay careful attention to the reviewers’ constructive comments, as they may help you revise this manuscript or design further research.  Please understand that my decision was rendered with the recognition that the page limitations of the journal dictate that only a small percentage of submitted manuscripts can be accepted.  PSPB receives more than 700 submissions per year, but only publishes approximately 125 papers each year.  Papers without major flaws are often not accepted by PSPB because the magnitude of the contribution is not sufficient to warrant publication.  With careful revision, I think this paper might be appropriate for a more specialized journal, and I wish you success in finding an appropriate outlet for your work.

I am sorry that I cannot provide a more favorable response to your submission.  However, I do hope that you will again consider PSPB as your research progresses.

P.S. I asked the acting editor to clarify her comments and her views about the validity of the race IAT as a measure of African Americans’ unconscious preferences. They declined to comment further.

2 thoughts on “Open Science: Inside Peer-Review at PSPB

Leave a Reply to Ulrich Schimmack Cancel reply