The Obligation to Publish

Lately, I’ve been feeling a bit melancholy about the my obligations to speak publicly about what I know. This has affected both my contributions to this blog, and my work on my longstanding blog about academic writing. It’s not, of course, that I don’t know anything, nor that I don’t have anything I want to say; it’s just a sort of reticence about engaging with others. It will, of course, pass in due time, and it’s probably not something to worry about. But it does raise an interesting ethical question: do we have an ethical obligation to say publicly what’s on our mind?

The Code tells us that we have an obligation

2. To the advancement of managerial knowledge. Prudence in research design, human subject use, and confidentiality and reporting of results is essential. Proper attribution of work is a necessity. We accomplish these aims through:

  •   Conducting and reporting. It is the duty of AOM members conducting research to design, implement, analyze, report, and present their findings rigorously.

I imagine most people read this with an emphasis on “rigorously”, i.e., as a responsibility when we do conduct research and report it to do so rigorously. But I think we do well to keep in mind that if we spent our entire scholarly careers conducting no research at all, or not reporting whatever research we did conduct, we would in fact be shirking an important responsibility.

Reporting our research opens it to criticism by our peers. It allows us to be corrected in our views wherever they happen to be erroneous. One of the most important reasons to publish, that is, is to give our peers an opportunity to tell us where we have gone wrong, so we can stop misleading our students about it, for example. But it is also a way of informing others about results that might call their previously held views into question. If I know that something you think is true is actually false (or vice versa) then I have an obligation to share that knowledge with you. That’s part of what it means to be an academic.

There’s an interesting variation on this theme in the current discussion of the publication of “null results”. If 9 out of 10 studies show no significant effect of a particular managerial practice, but only the 1 out of 10 studies that shows an effect is published, then we are being systematically misled about the efficacy of that practice. And yet, in today’s publishing culture, authors and journals are much less incentivized to publish null results than significant ones.

The Code does say that it is the responsibility of “journal editors and reviewers [towards the larger goal of advancing managerial knowledge] to exercise the privilege of their positions in a confidential, unbiased, prompt, constructive, and sensitive manner.” Perhaps I’m once again grasping at straws, but it is possible to construe “unbiased” as requiring us to publish valid but insignificant findings, i.e., studies that show no effect where one was hypothesized.

This becomes a personal ethical concern for individual scholars when they don’t publish results that call their own favoured theory into question, always, of course, citing the unwillingness of journals to publish null results. But whether it’s the authors or the editors that are to blame, the overall effect is that the truth remains hidden. So, in a sense, it is a species of dishonesty.

For that reason alone, I hope this melancholy of mine soon passes and that I once again start doing the responsible thing, namely, putting my ideas out there for all to see.

16 thoughts on “The Obligation to Publish”

  1. I think that this is problem that many of us face. How much effort should we exert to make people aware of their misconceptions or of general problems in a literature? In a field as applied as ours errors can have real consequences, and while I think that we all have some ethical responsibility to try to correct errors and misconceptionsI also think that the responsibility rises with our own rise in influence. If we are handed a bullhorn in the form of tenure or a journal editorship (or the official blogging keys to “The Ethicist”) then we have an even greater responsibility to our field then when we are graduate students or junior faculty. I hope that your melancholy passes soon and that you are soon able to write about the specifics of this issue. Courage!

  2. “If I know that something you think is true is actually false (or vice versa) then I have an obligation to share that knowledge with you.”

    I’m not so sure. If I’m reasonably confident that you aren’t going to accept the truth [or, if you prefer, you aren’t going to accept what you see as (1) my version of the truth, (2) my misunderstanding of the truth, (3) my outright denial of the truth, or (4) my deliberate misrepresentation of the truth], do I really have that obligation?

    On the other hand, bless you for your willingness to put this on the table.

  3. That’s a good question. My first impulse is say that it’s a bad excuse, sort of like saying “I’m not going to submit this null result because no one will publish it anyway.” At some level it’s always going to be: if I think you are wrong I have an obligation to tell you. (As you just told me.) In any case, most publication is for an unknown reader; i.e., someone who *may* hold another view than you do, but you don’t know anything about their stubbornness in holding it.

    And thanks for the encouragement, also. I’ll continue on Tuesday.

  4. You raise important questions, but they go beyond the individual. I could spend all my time writing 10 excellent papers with null results and failing to get any of them published. In fact, I have actually been wrestling with just such a situation – one that has taken years and has yet to see the light (but I continue to be hopeful). My approach is to balance some of the “null” effects with other papers that are more likely to get published for other various reasons. Otherwise, my dean and colleagues would be wondering what I’ve been up to in the past few years. In other words, I maintain some level of pragmatism recognizing that our institutions have a bias, and that I current live with those institutions (and help run them – and write for them – and edit for them). The wonderful thing about being an academic is that if we are committed to one or another particular truth, at some point, we will have an opportunity to ‘get the word out’. Persistence, commitment, and passion are the critical ingredients that eclipse statistical significance.

  5. Benson, you sound almost … uhm … hopeful. 😉 That wonderful! You’re right. The idea is to keep plugging away at it, without investing too much time in “noble” but “losing” causes. That’s what makes ethics so difficult, and so important to discuss.

  6. Hmmmm. Not sure if I agree with that. Sometimes causes only appear to be “noble but losing” when they are really simply noble. Should we not do the “right thing” even if (or especially if) they appear to be losing causes? I am far more impressed by those who stand up to the status quo to do the right thing than those who do the right thing when it is a “winning” cause. Of course, perhaps there is a large contingent of us who all feel like we are alone and we simply need to find each other in order to turn losing causes into winning causes.

  7. In life or death situations, we rarely have the ethical obligation to do so something that will almost certainly kill us (even to save a life). Similarly, I would think we’re not obligated to do a “right thing” at the cost of our career in scholarship. Keep in mind we’re now talking about your obligation to “advancement of managerial knowledge”, not a situation in which you’re protecting an individual from personal harm. If it’s “merely” the truth that is at stake, we can pass up a promotion, perhaps even tenure, but we can’t be expected to pursue a fruitless struggle to bring some truth to light for years and years until our career is effectively derailed.

    If you stand up to the status quo and it simply gets rid of you, then you haven’t changed anything. So you have to do the “right-enough thing” long enough, perhaps, to discover your allies. After all, most truths have been hidden from us (as a species) for thousands of years. A few more decades won’t matter in the scheme of things.

  8. Luckily life-or-death situations are pretty rare in management. What is perhaps more common are outrageously unfounded claims by some of our practitioner friends or highly questionable statistical practices among our academic colleagues. What is our responsibility in the face of such behavior? Some of the worst offenders are very powerful and well connected. Do we simply stand by and wait a couple of decades until the latest insane fad passes? What if that same fad is costing organizations millions and doing the reputation of our field real harm?

  9. I share your concern, Andrew. My point with the life-or-death analogy is that our ethical obligations rarely (i.e., usually not even there) require us to lay down our lives for the cause. Thinking the analogy through: if the false claims is “merely academic” then I’m not required to sacrifice my academic career to correct it. The solution, then, (and this will be the subject of my next post), is to make sure that only a reasonable amount of courage is required of us to correct our peers’ errors.

    Now, it’s true that we can imagine cases where bad management ideas are justifying the needless suffering of thousands of people and (like you say) costing millions. This would be akin to an expensive medical treatment that doesn’t work, or even causes harm. It’s possible, in such cases, that you have “whistleblower” obligations. We’ve talked a little about that here (and here and here) at the ethicist in the past.

  10. I tend to think that some ethical responsibility lies with our professional associations – not just with our individual members. Why is it that the associations for medical professionals are so relatively good at putting out pretty strongly worded statements opposing quackery (like the debunked vaccine-autism link) but we never see AoM take a position on business quackery? I think that our reputation (ethical and practical) would be much stronger if AoM took opposition to some of the quackery and tomfoolery that characterizes our field – especially when we have very strong empirical evidence that something is nonsense.

  11. You raise a familiar and important issue, Andrew. In what sense is management a “profession” like law, engineering and medicine? In the paradigm cases, there is an accepted body of knowledge and a consensus about best practices. In medicine, for example, there is a consensus about vaccines that makes it relatively easy to identify the quacks.

    There is no comparable consensus about which managerial practices are hokum. In my view, the reason for this is, in part, that we publish mainly positive studies by people who are promoting particular practices–everything from lean manufacturing to authentic leadership. The conversation about these practices is had mainly by people who are interested in seeing them widely disseminated, usually people who are perfectly sincere in the intention to improve the lives of managers and employees.

    That is, there is no systematic review of the efficacy of the practices that our management theories are about. So “theories” become essentially ideologies that validate what people are already doing.

    My favourite example from medicine is PSA screening for prostate cancer. It took a while to shift the consensus from its then-obvious value to its now-obvious harms. But it was possible because of the sort of science that medical professional practice is routinely based on. The rise and fall of “lean” and “authenticity” in practice and in theory is not determined by science in the same way, but, as is often noted, by ebbs and flows of fashion.

    Without a reliable means to form a scientific consensus, I don’t think an official pronouncement from a professional body like the AoM is possible. We’re more like the humanities that way. We don’t see the MLA issuing pronouncements about Shakespeare, etc. Or at least I hope this doesn’t happen!

  12. Good points. I agree that it would be difficult for AoM to make these kind of pronouncements but if making such judgment calls is not possible at all then can we really call ourselves a scientific discipline? We seem to want to think of ourselves as one – unlike our colleagues in the humanities. For example, if meta-analyses clearly show that a widely used management intervention has no effect then why not take a position on this matter? You mention authentic leadership: if the statistical results in the foundational article are demonstrably false then why not point this out?

    I don’t want AoM to take a position on most matters of true scientific dispute but we I would love to have my professional association take a position on some of the clearer issues. The field of management does not have the best reputation and this may help make us a little more respectable and trustworthy even if it causes some hurt feelings. Perhaps we need to establish clear guidelines as to how AoM would make such position statements – perhaps something like the processes used in health and medicine by the Cochrane Collaboration.

  13. I haven’t looked at it very closely, but my sense is that any meta-analysis or systematic review of the management literature on any particular kind of intervention will reveal either a consensus that it “works” or a controversy about its efficacy. In some cases, the review itself may simply be inconclusive. Also, it’s very unlikely that a paper showing that a foundational article uses bad statistics will be published in a top-journal.

    The problem is not, I suspect, to be found in the meta-analyses (or lack thereof) or in some pronouncement-forming process, but in the literature on which these would be based. It may be very clear to you that a particular intervention has no effect–the problem is that it’s not at all clear to the scholars who study it.

    I often think the best solution is just to stop presenting ourselves as “scientists” and discuss problems of management in a more humanistic spirit. But I should disclose that I’d say that about great many social so-called sciences.

  14. I feel that we have to agree to disagree on this matter. Many meta-analyses will show the moderation of an effect but there are conceivably meta-analyses that show no average effect for a treatment or relationship and no moderation effect either. There are also other settings where no meta-analysis is really necessary. Take, as an example, the use of ipsative personality tests for selection in employment settings. Most of us recognize this practice as sheer lunacy but there are consultants out there who make a fortune selling ipsative tests to organizations for selection purposes. This is pure quackery and yet AoM is silent on the matter.

    I would also hesitate to refer to someone as a scholar if they are entirely unwilling to change their position on something in the face of empirical evidence. When one of my students becomes overly enthusiastic about a particular theory or model that they have encountered in a class I always ask them “What piece of evidence or empirical finding would make you change your mind about the value of this theory?”. Almost all of them come up with a reasonable answer, reflecting a willingness to change their mind. This seems like a core requirement for any scholar whether they are working in a scientific field or not.
    Thank you for an interesting discussion – it has certainly made me think.

  15. Likewise. And thanks for the example of personality tests. I guess it’s like any “untested” alternative medical treatment. The official opinion of the medical profession has to be to dismiss any claims that it cures a condition as quackery. In medicine I do grant that kind of reasoning: until the carefully controlled study has been done, we can’t claim that a particular dietary supplement will cure or prevent a disease.

    Perhaps, what we disagree about is whether management can operate as a science-based profession in the same way. It’s certainly interesting to imagine what corporate life would look like if we had a kind of “FDA approval” process for management practices (like ipsative personality tests).

Leave a Reply