Psychology professors in the US are mostly liberal and rarely conservative. For example, one survey found that 84% identified as liberal, whereas just 8% identified as conservative. That is a ratio of 10.5 to 1 (Duarte et al, 2015). This overwhelming dominance of liberals in the field of psychology has lead to a lot of bad science, propelled by questionable assumptions and faulty logic. Put simply, many researchers assume something is wrong with conservatives and that therefore conservatism is best explained as the outcome of psychological processes. Clark and Winegard (2020) put it nicely:
“Because most researchers are liberal, they view liberalism as “normal” and conservatism as “deviant” (in some sense) and therefore spend less time attempting to explain liberalism (except as the opposite of conservatism, which is viewed as mysterious and in need of explanation).”
Explaining political beliefs as the outcome of psychology is a form of reductionist reasoning. Some examples:
“…a four-year old’s openness to a new toy predicts how open she’ll be as an adult to, say, the US forging new relations with Iran and Cuba.”
“If you really want to understand someone’s politics…understand how they feel about novelty, ambiguity, empathy, hygiene, disease, and dis-ease, and whether things used to be better and the future is a scary place.” [italics in the original]
[Political orientation] “is typically merely a manifestation of a package of cognitive and affective styles.”
— Robert Sapolsky (2017) Behave: The Biology of Humans at Our Best and Worst
An alternative approach to understanding political beliefs would be to consider the extent to which these beliefs capture some reality outside the individual - that is, the extent to which they are reasonable extrapolations from a person’s experience. Want to understand why business owners want fewer regulations? Look at the effect of regulations on their businesses. Want to understand why soldiers say it’s their duty to obey superiors? Consider the realities of warfare.
Of course, the reductionists could be partly right in some cases under some conditions. This is ultimately an empirical matter. The challenge is to separate sound studies from junk studies. Take Sapolsky’s assertion that a young child’s personality predicts her politics as an adult. What is his basis for such a claim? One measly study (Block & Block, 2006), which reported not a single statistically significant finding. We’re talking not one reported finding that yielded a p-value of at least .05, which in itself is considered at the borderline of statistical significance*. What the authors did find is that extroverted and uninhibited preschoolers tended to endorse liberal political beliefs in their early 20s, while reserved and compliant children tended to endorse conservative beliefs as young adults.
Actually I was somewhat surprised that this study didn’t identify any statistically significant correlations between childhood behavior and the politics of young adults, given that childhood behavior reflects parental values to some degree and the children of politically engaged conservatives tend to be conservative young adults.** Conservatives value modesty and self-control more than liberals so I would expect their young children to be more reserved than the children of liberals - and I’d expect the adult children of conservative parents to be more conservative (on average) than those of liberal parents. Unfortunately, the Block & Block study did not collect information on the political attitudes of the preschoolers’ parents.
This study was also small - just 95 participants, “with relatively few participants tilting toward conservatism”. How few? The authors don’t say, but that doesn’t stop them, or Robert Sapolsky, from overstating the significance of their findings: a hallmark of junk social science.
—
* Instead, the authors reported correlations at the .10 significance level, although they hint that some correlations may have been stronger, saying “all findings reach at least the .10 significance level”. So why didn’t they report those stronger correlations?
** These are families that talk a lot about politics, so the children are exposed to conservative opinions and arguments. The parent-child correlation weakens as the adult children get older and are exposed to other viewpoints and experiences.
—
Note: This is a substantial rewriting of an earlier post, Junk Social Science: Ferreting Out Future Conservatives in Preschool.
Next: Qualities of Good and Bad Research
References:
Block, J. and J. H. Block (2006). "Nursery school personality and political orientation two decades later." Journal of Research in Personality 40(5): 734-749. https://doi.org/10.1016/j.jrp.2005.09.005
Cory J. Clark & Bo M. Winegard (2020) Tribalism in War and Peace: The Nature and Evolution of Ideological Epistemology and Its Significance for Modern Social Science, Psychological Inquiry, 31:1, 1-22, DOI: 10.1080/1047840X.2020.1721233
Duarte, J. L., Crawford, J. T., Stern, C., Haidt, J., Jussim, L., & Tetlock, P. E. (2015). Political diversity will improve social psychological science. Behavioral and Brain Sciences, 38, 1-13.DOI: http://dx.doi.org/10.1017/S0140525X14000430
Inbar, Y., & Lammers, J. (2012). Political Diversity in Social and Personality Psychology. Perspectives on Psychological Science, 7(5), 496–503. https://doi.org/10.1177/1745691612448792
Jennings, M. K., L. Stoker, et al. (2009). "Politics across Generations: Family Transmission Reexamined." The Journal of Politics 71(3): 782-799. https://www.journals.uchicago.edu/doi/abs/10.1017/S0022381609090719
Sapolsky, R.M. (2017) Behave: The Biology of Humans at Our Best and Worst. Kindle Edition. New York, NY: Penguin Press
Suhler, C. L. and P. Churchland (2011). "Can Innate, modular "foundations" explain morality? Challenges for Haidt's Moral Foundations Theory." Journal of cognitive neuroscience 23(9): 2103-2116; discussion 2117-2122. DOI: 10.1162/jocn.2011.21637