New paper out: Honesty, Intelligence, and Race

Honesty as a surprisingly strong marker of intelligence

Together with Sebastian Jensen, I have a new paper out:

• Jensen, S., & Kirkegaard, E. O. W. (2023). Honesty, Intelligence, and Race. OpenPsych, 1(1). https://doi.org/10.26775/OP.2023.07.17

Research shows that honesty correlates positively with intelligence. Similarly, there are racial differences in honesty, with Europeans being more honest than various other ethnic groups. It is currently unknown to what degree race differences in intelligence can explain the differences in honesty. We investigated this question using data from the National Longitudinal Study of Youth 1997 (NLSY97), a large American longitudinal dataset. We replicate prior findings that honesty correlated with measures of intelligence (r = .38, 95% CI [.34, .41]) and that Blacks (d = -0.67, 95% CI [-.76, -.59]) and Hispanics (d = -0.4, 95% CI [-.50, -.31]) are less honest than Whites, and this holds whether honesty is measured by self-reports, interviewer-reports or by parent-reports. In addition, race differences in honesty remained between Blacks and Whites but not between Whites and Hispanics after controlling for intelligence. Differences between Blacks and Whites but not Whites and Hispanics were noticeably lower in self-reports (Blacks: d = -0.18 [-0.24, -0.11], Hispanics: d = -0.24 [-0.31, -0.17]) than parent-reports (Blacks: d = -0.43 [-0.52, -0.35], Hispanics: -0.24 [-0.33, -0.15]) and interviewer-reports (Blacks: d = -0.7 [-0.75, -0.64], Hispanics: -0.3 [-0.36, -0.25]). Cross-national comparisons were made using national IQ data and Hofstede’s cultural dimensions. Bayesian model averaging suggests that Hofstede’s individualism dimension (β = .64, PIP = 100%), national IQs (β = .25, PIP = 73.6%), and masculinity (β = -.35, PIP = 100%) predict differences in honesty between countries. Parking violations per diplomat were only predicted by national IQs (r = -0.28, p < .001), given that no other variable reached a posterior inclusion probability above 0% besides national IQs. Implications and theories concerning these findings are discussed.

It is well known that there are race differences in crime rates, including for fraud and similar crimes that involve deception. Similarly, there are great national differences in the level of corruption, with northern Europeans being the least corrupt:

Another approach to looking at country level honesty is to run experiments. A 2019 study had informants pretend to have found a wallet and hand it in to receptionists in hotels to see how many were returned. Results look like this:

With regards to the differences in this wallet study, we sought to explain them using national intelligence and Hofstede's cultural dimensions:

Here we used Bayesian model averaging (BMA) because our number of predictors was small for the number of countries (n = 35). The most important predictor was individualism (beta = 0.64), which is basically what sets apart northern Europeans from the rest of the world (WEIRDest people in the world stuff, but see also Kevin MacDonald's work). There was also a plausible role for national intelligence (beta 0.25) and masculinity (beta -0.35, think Russian-style macho behavior). It should be noted that with this small sample size, these conclusions are not very firm.

American data

Because of the known group differences, one might wonder if there are differences in more general day-to-day life honesty. We examined this question using a data from a large American survey (National Longitudinal Study of Youth 1997 (NLSY97)).

Measuring honesty in a survey is tricky. You can't just ask a potentially dishonest person whether they are honest or not and expect an honest answer, as they might lie on that question too. They might also not lie, for instance, if they don't even care about lying in general. Alternatively, one can ask parents to rate their children. Again, this is not perfect because parents have some level of positive bias towards their children ("he was a good boy"). A more impartial account can be given by an interviewer, who neither a friend nor a relative. The limitation with them is that they haven't known the person for long, but just for the purposes of interviewing the person for the survey. Data for all three approaches were available for study. Here's the race gaps:

The results line up with expectations based on the amount of partiality for Blacks: self-reports produce the smallest gaps to Whites and have the most bias, parental report larger gaps and have weaker bias, and interviewer ratings show the largest gaps with the least amount of bias. For Hispanics, there was no such pattern, which is odd given the results for Blacks. Perhaps a statistical fluke? It can't be ruled out entirely given the sample sizes.

From a psychometric perspective, we would expect that the three measures of honesty correlate positively, as they are all indicators of a general honesty factor. In fact, they do correlate, but not much:

The correlation between the 3 different honesty variables was fairly low. The tetrachoric correlation between self-reported honesty and interviewer-reported honesty was 0.07, the tetrachoric correlation between self-reported honesty and parent-reported honesty was 0.32, and the tetrachoric correlation between parent-reported honesty and interviewer-reported honesty was 0.18.

Still, one can do research with the imperfect scores on this general honesty factor which is (hopefully) less biased and more reliable than each of the indicators:

For Hispanics, the gap is larger on the general honesty score than each individual measure, in line with expectations; but for Blacks, the gap is slightly smaller than for the interviewer-report measure only. Maybe the biases from Black self- and parent report are strong enough to offset the increased reliability from combining them. The distributions of honesty look like this:

Most people of all races are pretty honest, but there is a far left tail of very dishonest people, and this tail is a lot larger for Blacks than for Whites and Hispanics. This is where we expect to find most of the career criminals, serial welfare fraudsters etc.

Looking at the national rankings above, we saw that the more intelligent countries are more honest in general, with China being a notable and large negative outlier (presumably related to communism). Similarly, we know from decades of research that criminality correlates negatively with intelligence, so we would also expect that honesty correlates positively with intelligence. Indeed, this is what we find:

A correlation of about .40 is rather large! Larger that I had expected. Theoretically, this is still underestimated because of the imperfect reliability of the general honesty score, and because of the ceiling effect for the honesty scores. In fact, based on the low correlations among the measures of honesty, the reliability of the general score should be quite low -- maybe .50 -- so the adjusted correlation between intelligence and honesty should be quite high, .60 perhaps. Such a high value is difficult to believe, so this requires further study.

On a similar note, one might want to know whether the race gaps in honesty can be accounted for by parental socioeconomic status (SES):

Looking at model 1 and model 4, we see that parental SES doesn't predict honesty after controlling for intelligence. But even controlling for intelligence, Blacks are still somewhat more dishonest than Whites, an effect which was not seen for Hispanics. The same pattern of results were found for each of the three honesty measurement methods, though with weaker results (beta for intelligence on self-reported honesty 0.09, for parent-report 0.24, for interviewer-report 0.30).

Conclusions

• Honesty differs by race, probably by a substantial amount.

• Honesty correlates with intelligence, probably not below .40, even if that seems difficult to believe.

• Intelligence explains most of the racial differences in honesty, but race probably predicts honesty independently of intelligence. Probably this is because of differences in personality traits and culture, mainly individualism.

• The results are potentially of relevance to explain the prevalence of scientific dishonesty seen outside of the western world, and by foreigners in the western world. Given the ease at which scientific dishonesty can be discovered if someone cares to look at papers with a critical eye, I predict that AI capable of automatic detection of scientific dishonesty will soon be created. They will then check all published studies in the world and identify thousands of fraudulent or grossly incompetent researchers. There will be a large over-representation of non-Europeans among these.