They see themselves as just as free and feckless as anyone from that underclass they’ll never have to meet. That’s how neatly they’ve arranged their lives, their careers, their families, their investments, and their estates.

They just don’t, as Charles Murray says, preach what they practice.

No worries :).

TLDR – The Big Five Personality Index, even when administered as a self-report survey, is actually fairly predictive of a lot of objectively observable behavior.

I’ve always thought personality tests don’t so much tell us about one’s personality as how one wants others to view one’s personality.

Most personality tests rest on self-report. Analysis based on self-report in general is subject to the weaknesses you describe. The question is whether this has adequately been accounted for in the Big 5 personality tests.

We can start by asking ourselves whether people would have an incentive to self-report dishonestly or inaccurately. We would certainly expect this to be the case with self-reports of, say, IQ, because of the positive associations with high IQ. With personality traits it’s not so clear because each trait can be described using both positive and negative adjectives.

(It’s worth noting that the Big 5 was developed by simply giving people a randomly sampling of adjectives, asking them how well each adjective described themselves, and then looking at what adjectives clumped together. The theory was developed afterward as an explanation to this clumping).

You make the sound point that the emotional valence of adjectives can vary depending upon one’s social context, including the political values of one’s community. So the question becomes, do these personality traits correlate with other traits that are not obviously related to political values? We do know that they correlate with media genre preferences[1], which is at least not obviously politically motivated.

But, as you rightly point out, the real question is whether self-reports of these personality traits are also consistent with objectively observable behavior, as that isn’t so easily falsifiable. While I don’t have studies handy, my understanding is that we do have evidence of such correlations. For instance, Openness is correlated with creative ability (as found in artists and entrepreneurs). Conscientiousness is correlated with behaviors like keeping a schedule, keeping a clean living space, and success in highly structured, rules-oriented environments.

Now, there is the meta-question about how reliable these results are. Given the general state of the social sciences at present, a good deal of skepticism is warranted. On this I will say two things. The first is that the current social science establishments tend to be less critical of results that conform to the politically correct/social-constructionist consensus, whereas many of the results from personality research are decidedly not politically correct. The second is that those who have done the most to criticize the lack of statistical rigor in the social sciences have also tended to stand by the results of psychometric research (IQ and the Big 5).

The million-dollar question though, is this: how do we explain the authoritarian behavior we have seen from the supposedly “open-minded” Left? That question deserves a whole other essay, but for now I would just like to observe that at least in our present circumstance this is in part due to an ongoing political re-alignment.

[1] http://www.overcomingbias.com/2015/10/media-genre-more-basic-than-politics-or-personality.html