The survey defined socialism as "what is done in Scandinavia", which for me is social-democracy : free-market capitalism with wealth redistribution, strong social safety net, regulation to protect workers, customers or the environment, and some critical sectors (like education) more or less directly handled by the state.

My own definition of socialism is "socialized ownership of the means of production", which can take many shapes : government ownership is a form socialism, but cooperatives or mutualism are other forms of socialism. Socialism doesn't necessarily means centralized planning, even if it is usually described as such.

But anyway, since the survey defined socialism in the first meaning, I used under it the first meaning to answer "socialism" in the survey, and I think we can safely assume most people who answered "socialism" used it under the first meaning.

Or rationality doesn't much impact people's views on politics. Good thinkers seem to me to be all over the place politically.

"either socialism is more interesting than it looks or people in Europe define 'socialism' as liberalism".

I wouldn't be surprised.

In my country a "liberal" is basically "classical liberal", some kind of conservative or libertaria, but I've grown accustomed to American usage over the past three years.

We need data on what the proportion of Americans is.

Statistical significance is a mental disease.

Only if you misuse it! Statistical significance is a good shorthand for how likely it is that a result is a fluke, which is helpful to know. Concluding from even astronomically precise p-values to some particular non-null hypothesis without other evidence in its favor, like "reading the Sequences causes people to be less socialist," is another kettle of tea.

Or to go from a different angle: I don't think much of anything can be concluded about communists and the sequences, even though there's a non-tiny effect size, because, like, there are only five of us. (Probably the same applies to conservatives, just slightly less so; I'm too lazy to do the math.) One's better off with reasonable priors: Sequences probably don't impact politics that much, communists are probably like socialists in their likelihood of reading a blog by a George Mason economist, the terms probably aren't cutting reality at its joints, and so on.

or people in Europe define 'socialism' as liberalism"

"Socialism" is a really imprecise word aside from various specialized contexts and discourses; all you can really conclude is that the person identifying as such values equality relatively highly. (In this sense it's a poor descriptor but perhaps a less mind-killing label than a good descriptor of policy preferences would be, although "left" is probably better at this still, so it's not clear to me that there's an actual sweet spot that would justify continued use of the term.) Per the definitions offered in the survey, socialism ("socially permissive, high taxes, major redistribution of wealth") and liberalism ("socially permissive, more taxes, more redistribution of wealth") are basically the same thing, aside from being defined absolutely or relatively (such that liberalism would technically be to socialism's left if you think your scope polity has high taxes and major redistribution of wealth.) Since there's a lot of left ideological space between "a bigger welfare state" and (meaningfully narrow interpretations of) "state control of many aspects of life," it's difficult to say how many self-reported socialists are social democrats and how many are one flavor of "hard" left or another. Similarly, it's hard to guess how many of the Moldbug set classified themselves as libertarian versus conservative - is "socially permissive"/"traditional values" about religion or race and gender? - and hence what the paleo/"liberaltarian" split is among libertarians (though you could probably catch the Objectivist and ancap cluster with virtue ethics and deontology.)

It seems pretty clear that the political mainstream here, as in almost all educated Western sets, is some form of cosmopolitan democratic capitalism, but it's difficult to see how large that mainstream is. Maybe you'd see some interesting correlations with being some flavor of weirdo or a weirdo at all - so if you have a strong prior that reading the Sequences will draw you into that mainstream, the low effect size of reported socialism on sequence reading probably shouldn't affect it much! - I dunno.

As a Scandinavian socialist I support things like:

• High taxes and public spending - around 50% of GDP seems about right.
• Cradle-to-grave socialized medicine.
• Publicly funded education - up to and including the university level.
• An elaborate social safety net.
• Extensive feminist social engineering schemes (long state-funded paternity leave and so on).

Why do I support things like that? Because I'm used to them and my first-hand experience tells me they work quite well. Also because the Scandinavian countries look good in international comparisons of various things. Though of course we can think of alternative explanations for that - maybe Scandinavia works well because it is populated by Scandinavians (a Steve Sailer type explanation) - I'm open to persuasion.

If you define socialism as "a system with state ownership of the means of production" then I'm not a socialist but some other type of collectivist.

IIRC, liberalism itself has different meanings on the two sides of the Atlantic, even though I can't remember what either of them was.

Since old-timers tend to be Overcoming Bias fans the libertarian-leaning nature of that blog explains the difference.