I think there's another reason why conflict of interest from having made a successful career out of speaking engagements based on your research are not perceived as badly as those where you directly received payment.
In the former your bias is only beholden to yourself, in the latter you're litterally on somebody else's payroll.
I don't think it makes it less impactful in terms of harming your objectivity, but I can see the moral case for it being "less wrong"
What about researcher allegiance effects? I. E. Having a pro-certain therapeutic intervention motivation that influences their findings. There is some research that found similar findings as the research on conflicts of interests. For example a psychotherapist that is completely convinced of the therapeutic intervention they specialise in? Like trauma-focused therapies, or EMDR? I would presume this also goes for social constructionists, or evolutionary psychologist, I. E. They are convinced their foundational theoretical framework is the end-all be-all to explain everything. This falls under researcher allegiance effects and also biases the research towards what the researchers want to believe. This feels similar to conflicts of interests.
I am always late to the party but to me the dividing line for "soft" conflicts of interest is membership in an org. To take the Thatcher paper example: a socialist in general would not need to declare a COI but a member of a socialist party would. This extends out to other areas as well - if you are an advisor for the Trevor Project you should declare a COI on a paper about LGBT discrimination because you have a vested personal interest in an organization with a specific stance on your research topic. On the other hand if you are simply broadly pro-LGBT the "conflict" in the COI is purely a conflict with yourself and thus is not severe enough to mention.
I can only speak to the US on this but my impression is that a fair amount of published work is put out by people who are active members or advisors to orgs who then use that published work to advance their social mission. This is just as much a COI as any money.
so where is the limit? Are book royalties ok? Public lectures? Don’t those have the same incentives to stick with one’s prior results? Hard for me to see a clear line
Worth a mention: industry funded clinical trials tend to cherry pick participants, hence the need for real world studies once the drug has been approved and made available.
Devil's advocate: Couldn't the difference in odds ratios between for profits and non profits be explained by profit motivations causing companies to be more judicious with their spending and pickier about what studies they conduct?
I think it's an unlikely rationale, especially given the rest of the episode, but that particular point stood out to me
The discussion of researchers not publishing results that conflicts with their political interests immediately brings to mind the 2020 case where social science researchers retracted their own paper on racial bias in police shootings, despite standing by its methods and conclusions, because it was being cited by conservatives to support their position: https://retractionwatch.com/2020/07/06/authors-of-study-on-race-and-police-killings-ask-for-its-retraction-citing-continued-misuse-in-the-media/
Also the research into the effectiveness of puberty blockers for gender dysphoria that’s sitting in a file drawer because it didn’t show benefits and that could be ‘weaponised’: https://www.nytimes.com/2024/10/23/science/puberty-blockers-olson-kennedy.html
Sometimes it’s not subtle at all.
I think there's another reason why conflict of interest from having made a successful career out of speaking engagements based on your research are not perceived as badly as those where you directly received payment.
In the former your bias is only beholden to yourself, in the latter you're litterally on somebody else's payroll.
I don't think it makes it less impactful in terms of harming your objectivity, but I can see the moral case for it being "less wrong"
I had submitted this paper just before you dropped the podcast.
Did I go too far with my COI?
https://www.nature.com/articles/s41746-024-01408-x
What about researcher allegiance effects? I. E. Having a pro-certain therapeutic intervention motivation that influences their findings. There is some research that found similar findings as the research on conflicts of interests. For example a psychotherapist that is completely convinced of the therapeutic intervention they specialise in? Like trauma-focused therapies, or EMDR? I would presume this also goes for social constructionists, or evolutionary psychologist, I. E. They are convinced their foundational theoretical framework is the end-all be-all to explain everything. This falls under researcher allegiance effects and also biases the research towards what the researchers want to believe. This feels similar to conflicts of interests.
I am always late to the party but to me the dividing line for "soft" conflicts of interest is membership in an org. To take the Thatcher paper example: a socialist in general would not need to declare a COI but a member of a socialist party would. This extends out to other areas as well - if you are an advisor for the Trevor Project you should declare a COI on a paper about LGBT discrimination because you have a vested personal interest in an organization with a specific stance on your research topic. On the other hand if you are simply broadly pro-LGBT the "conflict" in the COI is purely a conflict with yourself and thus is not severe enough to mention.
I can only speak to the US on this but my impression is that a fair amount of published work is put out by people who are active members or advisors to orgs who then use that published work to advance their social mission. This is just as much a COI as any money.
Seems like commenting has been turned off for the most recent episodes? What's that about?
Oh - not deliberate! I’ll look into it
so where is the limit? Are book royalties ok? Public lectures? Don’t those have the same incentives to stick with one’s prior results? Hard for me to see a clear line
Worth a mention: industry funded clinical trials tend to cherry pick participants, hence the need for real world studies once the drug has been approved and made available.
Devil's advocate: Couldn't the difference in odds ratios between for profits and non profits be explained by profit motivations causing companies to be more judicious with their spending and pickier about what studies they conduct?
I think it's an unlikely rationale, especially given the rest of the episode, but that particular point stood out to me
I think that’s the point- they’re just picker in a way that biases studies to those leading to better results, not better studies.