Jan-Erik Lönnqvist vs. Emil Kirkegaard on mental illness and the left

The Finnish war

Occasionally, academics launch an unprompted attack on me. Unprompted in the sense that I had never previously heard of this person as far as I know. Recently a Swedish professor Jan-Erik Lönnqvist was interviewed for the Finnish media Iltahleti. The title summarizes the article "The "nonsense study" of the liberal left began to spread like wildfire - The professor gives a blunt assessment" (original in Finnish). Apparently, my article "Mental illness and the left" upset the professor enough that he wanted to supply a rebuttal of sorts. I mean, I understand why, this article has 230k+ views on Researchgate (combined numbers from published and preprint versions). This is far more than Lönnqvist's combined read count for all his 80 articles, which stands at 59k (which is not bad!). Looking at his research it appears in general to be decent, large-sample based work.

I wrote the journalist to ask for the right to reply. He said an interview would be possible and sent me a list of questions by email. This reply was duly published as "The Danish "nonsense study" about the liberal left spread like wildfire - Now its creator is speaking" (Finnish version), he basically cut out parts out of my answers to make them more readable for laypeople. However, here I will post the complete version for those interested, plus additional commentary.

---

How do you response to the criticisms that your study has not been peer reviewed?  

The study in question was peer-reviewed, as it was published in the peer reviewed journal Mankind Quarterly. This is easy to verify, one can just look at the journal's website. It's possible that Jan-Erik Lönnqvist made a simple mistake because he read the preprint version (non-reviewed) instead of the published of the article. Peer review isn't some magic method that guarantees results are correct, it just means that some other researchers looked it over and made some criticism before it was published. This is a low bar to pass, which is why so much published research is false and unreliable (replication crisis).

But even if Jan-Erik Lönnqvist wants to ignore my study, there are many replications done by other researchers. In 2021, Luca Bernardi published a study showing leftists have higher rates of depression. Soyoung Kwon published another replication in 2022. Zach Goldberg posted his findings on Twitter in 2020 from a recent dataset collected by Pew Research, also showing the same results. And back in 2012, Schlenker and colleagues published a study showing the general higher levels of happiness of conservatives. My 2022 blogpost The conservative advantage in mental health keeps replicating summarizes all this research and more. If Jan-Erik Lönnqvist had read this, then he would already know that much research shows that conservatives are happier and less mentally ill.

You said that in our article the professor said some falsehood about you, Can you be more specific?

Are the following comments, that that vegans are mentally ill, and that you don’t believe in climate change true?

Jan-Erik Lönnqvist said I don't believe in climate change, but that is false. I have spent quite a number of years trying to convince libertarians and conservatives that human-caused climate change is real. This 2015 blogpost describes my efforts.

Vegans are higher in mental illness, though not all vegans are mentally ill. There is a meta-analysis (a summary of other research) from 2020 by Isabel Iguacel and colleagues who showed that vegans suffer from more depression than non-vegans. I have also replicated this result in my own study from 2022. With regards to education, I generally want to defund (remove the subsidies from the government) higher education for both sexes. Not completely, but to some extent. The reason for this is that higher education is generally wasteful, doesn't produce the benefits that proponents claim, and costs a lot of money. Bryan Caplan has reviewed all this evidence in his excellent book The Case Against Education from 2018.

How do you response to the criticism that your articles have only been published in pseudoscientific magazines?  Where in general your studies have been published? 

It is a false statement made by Jan-Erik Lönnqvist. My research has been published in 15 different academic journals run by many different organizations. A cursory search reveals this if one looks at my Google Scholar page. Jan-Erik Lönnqvist is correct that I have published a lot of studies in Mankind Quarterly, and that he doesn't like this journal because of its conservative history and willingness to publish controversial research about group differences. But attacking the outlet where work is published is a weak argument. If Jan-Erik Lönnqvist has some actual criticism of my research, he is welcome to publish a study saying why my research is wrong instead of making false statements about me or my work in a newspaper interview. I could equally well point out that Jan-Erik Lönnqvist has himself published 5 papers in Frontiers in Psychology, which is a journal where one has to pay a large fee (3000 USD) to publish (most journals are free to publish in), and which has received ample criticism for being a predatory journal.

---

Now that I have more space, I should also note that Lönnqvist does accept the happiness correlations that I mention above. However, he buys into the cope theory about them:

- Among other things, trans-hatred, opposition to abortion, climate change, gun policy and the destruction of nature are all good reasons for liberals to be unhappier with the society they have looked at. And on the other hand, liberals are less prejudiced about mental health problems, which lowers the threshold for seeking help, says Lönnqvist.

It is cope because we are living in some of the most leftist and woke countries ever to exist in the history of mankind with regards to the above issues. If leftists still cannot be happier, that speaks more about their psychology rather than some implausible interaction effect between society and themselves. As far as I know, the tendency of leftists to be less happy has been replicated across countries already, so it is hard to dismiss by some hand-waving about some particular western society not being leftist enough. As a matter of fact, the left vs. right gap in happiness and mental illness has grown larger in the USA as it became more woke during the great awokening, not smaller as his hypothesis would predict.

These gaps have been found whether one uses diagnoses or symptoms to study the effect, so it would appear that measurement invariance problems cannot explain the results. In fact, Brandt et al 2021 conducted a large-scale study of measurement invariance for political ideology. They studied two large samples for this question. First results:

Second set:

As can be seen, the effect sizes for various psychological constructs were pretty similar before and after correction for measurement invariance issues. A lot of items showed differential item functioning (DIF), which is not surprising given their large sample sizes (160k in the US study, 6k in the Dutch). Almost no questions can be expected to work exactly the same for two groups of people that differ in many ways. However, these differences should not in general go in any particular direction with regards to the score, and thus will cancel out at the scale level. In other words, omnipresent DIF is not a big concern unless it has a certain consistency of direction. In the figures, positive values indicate a value in the direction of conservatives. Thus, we can observe that satisfaction with life was about 0.4 d higher in Dutch conservatives vs. Dutch leftists both before and after correction for DIF. Similarly, agreeableness was about 0.5 d lower for Dutch leftists compared to Dutch conservatives. Strangely, the agreeableness gap was not seen in the US data. Both datasets showed a small association between neuroticism and leftism (about d = 0.15), similar to my study. As such, there is no particular reason to expect mental illness or happiness gaps are due to measurement invariance failure, and what evidence we have, shows they still exist accounting for this potential problem.

Another curiosity is that the gap is larger among Whites.

From his perspective, one would have to come up with some implausible theory of how despite increasing leftism in society, Whites in particular are getting less happy and more mentally ill due to the aforementioned factors that mostly don't affect them ("structural racism"). Instead, I conjecture the counter-hypothesis that the reason White leftists in particular are getting more depressed etc. is that their own ideology specifies their ethnic group as uniquely blameworthy and responsible for the plights of others. Conservatives are less deluded about this. In other words, White self-hatred affects White leftists particularly hard. This seems a rather mundane model, but it has the advantage of having a lot of plausibility.