Wednesday, October 2, 2019

Why do People Find the Fundamental Attribution Error so Confusing?



Social psychologists have developed names for a host of biases in the ways people perceive behaviour. The fundamental attribution error is not only one of the most famous of these biases, but apparently one of the most frequently misunderstood. Many laypeople confuse the fundamental attribution error with distinctly different phenomena, such as the self-serving attribution bias. Such confusion is not limited to laypeople, however. A recent article by a sociologist making a misguided attempt to apply a sociological/social psychological analysis to the popular TV show Game of Thrones illustrates the same confusion, and perhaps illustrates a deeper confusion among those who would attempt to deny the importance of human individuality in the name of social science.

I have previously written about the fundamental attribution error in a pair of articles in which I argued that, despite the importance that some people claim for it, it is actually highly overrated. Even though its name is well-known, a lot of people seem to be confused about what this phenomenon is supposed to be, perhaps because it is a counter-intuitive concept. Hence, I will start with some definitions and distinctions. The fundamental attribution error was defined by psychologist Lee Ross as a tendency for people, when making attributions about the causes of behaviour "to underestimate the impact of situational factors and to overestimate the role of dispositional factors in controlling behaviour" (Reeder, 1982). In plainer language, this means that even when someone has very little choice in how they behave because they were pressured into doing something by external environmental demands (i.e., situational factors controlled their behaviour), other people will tend to assume that they behaved the way they did because of their own attributes, such as their personality, attitudes, and desires (i.e., dispositional factors). This phenomenon, which is supposed to be a ubiquitous and pervasive error, is said to occur because laypeople have causal theories of behaviour in which they believe that “situational factors have little impact on human behaviour” (Gawronski, 2004). That is, people supposedly underestimate the “power of the situation” because they are “intuitive dispositionalists” who mistakenly think that people act consistently with their personalities. Got all that?

At this point, some readers may object that what I have described is not the fundamental attribution error and that I don’t know what I am talking about. (Oh, ye of little faith!) Instead, they might think that the fundamental attribution error is something like the following:
We also have a bias for the individual as the locus of agency in interpreting our own everyday life and the behavior of others. We tend to seek internal, psychological explanations for the behavior of those around us while making situational excuses for our own. This is such a common way of looking at the world that social psychologists have a word for it: the fundamental attribution error.When someone wrongs us, we tend to think they are evil, misguided or selfish: a personalized explanation. But when we misbehave, we are better at recognizing the external pressures on us that shape our actions: a situational understanding. If you snap at a coworker, for example, you may rationalize your behavior by remembering that you had difficulty sleeping last night and had financial struggles this month. You’re not evil, just stressed! The coworker who snaps at you, however, is more likely to be interpreted as a jerk, without going through the same kind of rationalization. This is convenient for our peace of mind, and fits with our domain of knowledge, too. We know what pressures us, but not necessarily others.
This is a quote from an article by sociologist, Zeynep Tufekci, with the modest and unassuming title, “TheReal Reason Fans Hate the Last Season of Game of Thrones.” Like much else in her article, this explanation of the supposed fundamental attribution error is horribly garbled and inaccurate, and is part of an agenda to interpret this TV show through a situationist, sociological lens to promote her own theories. What Tufekci describes is not the fundamental attribution error at all but what social psychologists call the actor-observer asymmetry (AKA actor-observer bias). There are two main accounts of this latter phenomenon. In the strong or general account, people routinely explain their own behaviour in situational terms and other people’s behaviour in dispositional terms, regardless of whether the behaviour is good or bad. Social psychologists have claimed that the general version of the actor-observer asymmetry is a robust and pervasive effect. This is supposed to occur because people have different perspectives depending on whether they are actors (i.e. doing something) or observing another’s behaviour. That is, when performing an action, people are supposed to be more aware of how their environment affects their behaviour, whereas when they are observing an action, they are more focused on the person doing the action than their environment (Malle, 2006). However, there is another more limited version of this phenomenon, in which the way people judge their own vs. another’s behaviour depends on whether the outcome was favourable or unfavourable. Specifically, actors attribute their failures or negative behaviours to environmental, situational factors, and their successes or positive behaviours to their own personal characteristics. On the other hand, when observing others, they either do not show this bias or show the opposite effect: attribute other people’s successes to the environment (e.g., luck) and failures to their personal characteristics. This phenomenon is also known as the self-serving bias in attribution (self-serving bias for short).

Although the fundamental attribution error and these two versions of the actor-observer asymmetry might seem similar, there are nevertheless important conceptual differences between them. First, advocates of the theory of situationism, like Lee Ross and Richard Nisbett, have claimed that the fundamental attribution error occurs because people believe that situational factors have little impact on human behaviour, that is, that people simply do not understand the “power of the situation” because they prefer to explain behaviour in dispositional terms. Furthermore, this is supposed to be an error because situational factors are the central causes of behaviour while dispositions or personality traits are of little importance. On the other hand, the actor-observer asymmetry suggests that people do accept that situational factors influence their behaviour, at least when considering their own actions in the general account, or when it suits them in the self-serving account. Furthermore, if the self-serving account is correct, invoking situational causes is not necessarily a sign that one has grasped the true causes of behaviour, but that one is attempting to look at things in whatever way allows one to feel better about oneself. Hence, there seems to be a contradiction between what these alleged phenomena imply. Hence, it may be helpful to consider the current evidence for each of them.

First, let’s get the fundamental attribution error out of the way. I have criticised this at length elsewhere (here and here), so I will deal with it briefly by summarising the conclusion of a review (Gawronski, 2004) that I did not have a chance to cover in my previous articles. According to this review, although there is evidence that people do draw dispositional inferences from situationally constrained behaviour (which the author refers to as correspondence bias), there is no evidence that laypeople have causal theories in which “situational factors have little impact on human behaviour.” Hence, the author proclaimed that “the fundamental attribution error is dead.” Instead, the author argues that it is not that people underestimate the importance of situational causes in general, but that they sometimes misunderstand which specific situational factors may be acting on a person.

What about the actor-observer asymmetry? Regarding the general version of this phenomenon, the author (Malle, 2006) of a review of more than 170 studies spanning 35 years found that the mean effect size was indistinguishable from zero and concluded that “The actor–observer hypothesis appears to be a widely held yet false belief.” Additionally, he noted that “actors and observers do not notably differ in their person and situation explanations” of behaviour. On the other hand, the self-serving bias fared better, as there was evidence that this effect replicated across studies. Specifically, the expected actor-observer asymmetry occurred for negative events (i.e., people explained own failures situationally and another’s failure dispositionally), whereas the reverse occurred for positive events (i.e., people explained own successes dispositionally and another’s success situationally). Hence, the actor-observer asymmetry is not a general phenomenon, but does occur under certain circumstances, depending on the evaluation of what is being explained. Additionally, this effect is a fairly modest one, that is, one might say that people tend to engage in this bias some of them time, but not consistently and constantly.

Why the confusion?
Why do so many people confuse the fundamental attribution error with the actor-observer asymmetry? I’m not altogether sure, but I suspect it might be because the fundamental attribution error is a counter-intuitive, even weird idea that does not match people’s everyday experience, while the self-serving version of the actor-observer asymmetry is easier to relate to. Specifically, even though some social psychologists have claimed that most people are “intuitive dispositionalists” who mistakenly believe that “situational factors have little impact on human behaviour,” there is actual a lack of evidence that this is even true; on the contrary, there is evidence that most people accept an interactionist view, in which both situations and personality traits influence behaviour (Newman & Bakina, 2009). By the way, the interactionist view was endorsed by the eminent psychologists Kurt Lewin in 1938, and also happens to be in line with the empirical evidence. On the other hand, the idea that people tend to be self-serving and believe whatever happens to make them feel better is more intuitive and easier to understand as it matches people’s common experience.

Situations, dispositions: which face is more important?

One noted critic of the fundamental attribution error (Funder, 2001) argued that the really, really fundamental attribution error is committed by psychologists, and not by laypeople: believing that the causes of behaviour are simple and easily dichotomized. More specifically, proponents of situationism have tried to advance the fundamental attribution error as a phenomenon that illustrates their dichotomous theories that behaviour is really controlled by external situational factors vs. internal dispositional ones. Returning to Zeynep Tufekci’s article, it seems that she has tried to apply a similar dichotomous theory to pop culture in her Game of Thrones article, in which she attempts to apply a situational vs. dispositional analysis first to the TV show and then to society at large. This irked me, because it seemed to reflect an attempt to deny the importance of individuality in society, which has long been a theme of situationist discourse in social psychology.

Was Storytelling in Game of Thrones, Sociological, Psychological, or a Mix of Both?
Tufekci’s article makes several claims that she thinks explain why season eight of the show was unsatisfactory to many fans. She argues that the show switched from what she calls sociological storytelling to psychological storytelling. As she explains, the former involves “structural storytelling” in which “characters evolve in response to the broader institutional settings, incentives and norms that surround them.” Psychological storytelling, on the other hand, which is apparently the norm for most TV shows, focuses on “overly personal” stories about individuals and their distinctive personalities. She seems to imply that this switch began with season eight because the show ran ahead of the George R.R. Martin novels, which focused on sociological stories, and the show’s producers did not understand how to continue doing sociological story-telling and switched to the more familiar mode of psychological story-telling instead. (Actually, the show had completely ran ahead of the books by the beginning of season six, but Tufekci’s article does not acknowledge this). Apparently, “the psychological/internal genre leaves us unable to understand and react to social change.”

Tufekci argues that characters in sociological storytelling still “have personal stories and agency,” but “are also greatly shaped by institutions and events around them,” and act according to incentives that “come noticeably from these external forces, too, and even strongly influence their inner life.” However, despite her assertion that characters have their own agency, she goes on to argue that a character’s choices are basically driven by situational factors and implies that individuality is not that important:

The hallmark of sociological storytelling is if it can encourage us to put ourselves in the place of any character, not just the main hero/heroine, and imagine ourselves making similar choices. ‘Yeah, I can see myself doing that under such circumstances’ is a way into a broader, deeper understanding.”

Note that she says any character, no matter how depraved or shocking their behaviour, which seems to imply that people and characters are basically interchangeable. If it were true that people make similar choices based only on circumstances, then this would imply that a person’s values, desires, and inclinations, indeed their personality, is of no real importance. Using some salient examples from the show, if viewers can put themselves in the place of any character and imagine themselves behaving the same way under the circumstances, this would mean that, on the one hand, they could see themselves being treacherous and opportunistic if they were in the place of someone like Petyr Baelish, yet also imagine that they would be honourable to a fault, even if it meant their own downfall, in the shoes of someone like Ned Stark. This does not make any sense. The meaning of honour is that a person does not betray their friends when it happens to suit their incentives. In fact, it means making hard choices, even when it would be convenient to put personal interests first. Arguing that viewers can imagine that they would act willy-nilly depending on the whims of circumstances would mean that they lack a concept of having personal values that guide a character’s behaviour.
Tufekci particularly criticizes how the show handled the character arc of Daenerys Targaryen, especially her murderous turn in episode 7. She argues that this started out as a story of the corruption of power, and if it had been done sociologically, it would have provided an interesting study of “a leader who starts in opposition with the best of intentions, … and ends up acting brutally and turning into a tyrant if they take power.” Her argument is that Dany’s downfall was originally set up to illustrate how she became corrupted by external forces provided her with ruinous incentives, “and season by season, we have witnessed her, however reluctantly, being shaped by the tools that were available to her and that she embraced: war, dragons, fire.” Then, unfortunately, her storyline in season eight went from an interesting sociological one to being ruined by this psychological storytelling[1] that the producers seem to favour.

It’s not my intention to defend season eight or the handling of Daenerys’ storyline. (People could be arguing about this for years to come.) What I want to point out instead is that the distinction that Tufekci makes between sociological and psychological story-telling rests on a false dichotomy and that she twists the facts to fit her pet theory. Even a casual inspection of the plots from the early seasons of the show illustrates that Dany was portrayed as having striking psychological characteristics that foreshadowed exactly her later descent into genocide and tyranny, quite apart from any external sociological forces that may have shaped her decisions.

Consider an important scene from season two, episode 3 (Garden of Bones). (See the relevant clip here.) 

Dany and her retinue are at the gates of the city of Qarth seeking admission. She states frankly, that her people have no food and water, and will all die if not admitted. The person whom she needs to convince to let her in, requests to see her dragons, as some of his colleagues doubt they even exist. Even though this is a simple and reasonable request that she could easily accommodate, she refuses and is denied admittance. She then makes the following remarkable statement, that explicitly foreshadows what she will do in season eight: “When my dragons are grown, we will take back what was stolen from me and destroy those who wronged me! We will lay waste to armies and burn cities to the ground!”

So, even though she is in a weak, vulnerable position, she refuses to use the only bargaining chip she has, and instead makes threats that she cannot yet carry out. Hence, even at this early stage, when she has quite limited power and her dragons are only babies, she demonstrates grandiosity, entitlement, and vindictiveness, even hints of madness. Is this an example of sociological storytelling? What incentives exactly did Dany have to behave in such an irrational and potentially suicidal manner? How many viewers think they would act the same way under these circumstances? Despite Tufekci’s argument that Dany was gradually corrupted by external forces and incentives, her own peculiar personality is clearly an important factor in explaining her behaviour, as she acts in a way that is clearly contrary to the “incentives” afforded by the situation.

Tufekci’s argument about the supposed distinction between sociological and psychological storytelling, with the psychological mode being the dominant but misleading one, while the sociological mode tells the real story, mirrors the distinction made by situationists between situational and dispositional accounts of behaviour. Both Tufekci’s argument and the situationist argument misrepresent their subject matter and both present a lopsided view of things that tries to deny the importance of human individuality. This is not to say that sociological factors are not important in good storytelling or in real life. An alternative to the view that sees personality dispositions and situational influences on behaviour as opposing forces, is that a person’s circumstances can reveal what their personality is really like. For example, facing danger provides a test of whether one will be cowardly or courageous, treacherous or loyal, depending on one’s individual strengths and weaknesses, virtues and vices. Similarly, when at its best, Game of Thrones was able to tell stories in which interesting characters faced trying circumstances that required them to make difficult choices, and these choices revealed their true selves. Tufekci argued that being given power corrupts even well-meaning leaders, but an alternative view is that power provides a test of the strength of a person’s character, that is, whether one has the capacity to act with moral integrity or whether one sees power as a path to self-aggrandizement. This question of choice was actually raised in the final episode of season eight in a scene in which Tyrion asks Jon Snow whether he would have acted the same way as Daenerys if he had opportunity to do so. (See this clip.) 

Tyrion reminds him that he has ridden on a dragon’s back, that he had that power, and asks him if he would have burned a city down? This question brings into focus the crucial importance of personal responsibility and choice. Situationist accounts of behavior tend to downplay personal responsibility and imply that people are controlled by external forces, which also seems to be Tufekci’s argument. This is a disempowering view of human nature and one that is based on discredited ideas, like the fundamental attribution error.

Tufekci argues that sociological storytelling is important because “the dominance of the psychological and hero/antihero narrative” may also be “the reason we are having such a difficult time dealing with the current historic technology transition.” She goes on to talk about the need to change the structures, incentives and forces that shape how people and their companies behave, rather than trying to “dethrone antiheros and replace them with good people.” This is all very well, but is worthwhile to remember that structural changes to society are often brought about by individuals who have the vision and the means to do so. For example, I noted in a previous post on American presidents that individual personality traits were substantially related to a president’s effectiveness while in office. I agree with her that it is important for society to build good institutions and provide incentives for people to behave well. But let’s not overlook the importance of individuals in shaping society.

This post was previously published on Psychology Today in two parts: Part 1 and Part 2.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted as long as a link to the original article is provided.



[1] Tufekci also complains that the scriptwriters explain Daenerys’ destructive choices in terms of “genetic determinism,” because the other characters note that madness runs in Dany’s family, as if this is a radical departure from the show’s previous sociological emphasis. On the contrary, from season 1, one of the show’s recurring themes was that incest has negative consequences for the children of such unions, which includes Daenerys and her family, with its long history of inbreeding. Hence, so-called “genetic determinism,” or at least recognition that heredity matters, has been an integral component of the storylines, even when it was supposed to be a “sociological” show.

References

Funder, D. C. (2001). The Really, Really Fundamental Attribution Error. Psychological Inquiry12(1), 21–23. 

Gawronski, B. (2004). Theory-based bias correction in dispositional inference: The fundamental attribution error is dead, long live the correspondence bias. European Review of Social Psychology15(1), 183–217. https://doi.org/10.1080/10463280440000026

Malle, B. F. (2006). The actor-observer asymmetry in attribution: A (surprising) meta-analysis. Psychological Bulletin132(6), 895–919. https://doi.org/10.1037/0033-2909.132.6.895

Newman, L. S., & Bakina, D. A. (2009). Do people resist social‐psychological perspectives on wrongdoing? Reactions to dispositional, situational, and interactionist explanations. Social Influence4(4), 256–273. https://doi.org/10.1080/15534510802674292

Reeder, G. D. (1982). Let’s give the fundamental attribution error another chance. Journal of Personality and Social Psychology43(2), 341–344. https://doi.org/10.1037/0022-3514.43.2.341

Sunday, January 15, 2017

What Differential-K Theory gets Wrong about Race Differences in Sexuality


In two previous posts, I discussed a study (Dutton, van der Linden, & Lynn, 2016) that aimed to test predictions from the highly controversial differential-K theory. Among other things, this theory proposes that there are racial differences in sexual attitudes and behavior, so that people of sub-Saharan African descent are supposed to be the most sexually permissive, while people of East Asian descent are supposed to be the most sexually restrained, and Caucasian descended people are said to be in-between. The authors used data from a sex survey by Durex, the condom manufacturer, to provide evidence of racial differences in sexual behavior that they claimed supported their theory. This method has problems because the survey lacked data from African nations, and its methodology was not up to rigorous scientific standards. Fortunately, more rigorous scientific research is available using data from 48 countries, including several African ones, that can shed some light on this subject. As will become clear, the evidence from this research contradicts the predictions of differential-K theory in several major respects. Furthermore, cross-cultural differences in sexual attitudes and behavior are related to a variety of environmental, social and cultural factors that need to be taken into account before making sweeping generalisations about race.



I have discussed differential-K theory before in some detail, so I will recap it only briefly here. According to this theory, over millennia, various racial groups have developed different reproductive strategies in order to adapt to differences in their local environments. Sub-Saharan African people developed a fast life history strategy entailing sexual permissiveness and high fertility. Caucasian and to an even greater extent Asian peoples developed slower life history strategies, characterised by greater sexual restraint, and more intensive parental investment in a smaller number of children. Dutton et al. (2016) proposed that these racial characteristics are related to differences in levels of androgens (male hormones including testosterone), with higher androgen levels being related to faster life history strategies. The evidence they presented did not wholly support this theory, as they actually found that African peoples were more similar to Asian ones in some respects. Dutton et al. argued that since sexual behavior is influenced by androgen levels, racial differences should be evident in this respect. Using data from the Durex sex survey, they found that survey respondents from Caucasian nations reported a higher annual frequency of sex and more lifetime sex partners than respondents from Asian nations. Unfortunately, the only African nation sampled in the survey was South Africa, so it was not possible to compare Africans with the other two groups. Nevertheless, Dutton et al. concluded that the results were broadly in line with the prediction of differential-K theory that Asians would be more sexually restrained than Caucasians.

Apart from the lack of African data, the survey has problems with lack of scientific rigour. It was conducted mainly as a promotional activity for Durex and it is not clear how representative the respondents are of the general population from which they were drawn. Furthermore, the survey made no attempt to account for important differences between the nations sampled that could influence sexual behavior, such as social and economic conditions. Hence, the survey provides a crude comparison between nations and it is unclear how much the results are due to biological factors related to the respondents’ race as Dutton et al. propose or to other salient factors.

One prediction of differential-K theory is that populations with a predominantly fast life history strategy should have more interest in short-term mating, whereas those with a predominantly slow life history strategy should be more focused on long-term mating. One way to assess interest in short-term versus long-term mating is to look at sociosexuality. i.e. attitudes towards sex without commitment to a long-term relationship. People high in sociosexuality feel comfortable with uncommitted sex and are interested in having many sexual partners. Conversely, people low in sociosexuality are generally unwilling to have sex outside of a committed relationship and consequently desire fewer partners. If differential-K theory is correct, African nations should have the highest rates of sociosexuality, Asian ones should have the lowest, while Caucasian ones should be intermediate. There has been a study on national levels of sociosexuality (Schmitt, 2005) which includes enough nations from each of these three groups to allow the necessary comparisons. This study found that both African and Asian nations on the whole had significantly lower levels of sociosexuality compared to Caucasian nations from Europe, North America and Australia. This result contradicts differential-K theory and is also consistent with Dutton et al.’s other findings that people from African nations were more similar to those from Asian nations in some respects than they were to Caucasians.

Looking more closely into Schmitt’s results reveals more findings that are inconsistent with differential-K theory. National sociosexuality was positively correlated with the level of human development, including life expectancy, and negatively correlated with infant mortality, teen pregnancy rate, prevalence of child malnutrition, prevalence of low birth weight, and fertility. According to life history theory, environments in which life expectancy is short and infant mortality is high should foster a fast life history strategy, involving early marriage and high levels of fertility. Differential-K theory predicts that fast life history strategies should be associated with sexual permissiveness, yet in countries with shorter life expectancy, high infant mortality and high fertility, people actually tend to be more sexually restrained compared to people in more developed countries with better life expectancy and so on. Schmitt’s results are more consistent with an alternative theory, known as strategic pluralism, that low sociosexuality and a preference for monogamy are more adaptive in harsh, difficult environments because infants have better chances of survival when bi-parental care is more prevalent. Conversely, in resource-rich environments, such as in developed nations, single-parenting becomes more viable and higher sociosexuality becomes more common.
Another environmental factor associated with sociosexuality is imbalanced sex-ratio (Barber, 2008). For example, in societies where there are more men than women available for marriage, levels of sociosexuality tend to be lower. In this situation, marriageable women are highly in demand, and they can demand a higher level of relationship exclusivity from prospective partners, and are more likely to delay intercourse until after marriage. Conversely, when there are fewer men than women available for marriage, levels of sociosexuality tend to be higher. In this situation, women must compete more intensively for mates; consequently, sex outside of marriage and more permissive sexual attitudes are more common. Imbalanced sex ratios featuring more men than women occur today in a number of East Asian countries and this might help explain why these countries tend to be more sexually conservative (Schmitt & Project members, 2003). Additionally, societies where it is common for women to delay marriage in order to pursue a career also tend to be more sexually permissive. This might help explain why higher levels of economic development in a country are associated with higher levels of sociosexuality (Barber, 2008).

International differences in sociosexuality might also be related to changes in societal values as countries become more developed. Improvements in economic development are not only marked by increased life expectancy and reduced infant mortality, but by changes in societal values. Poorer countries tend to be marked by survival values, which emphasise tradition and obedience to authority. Economically developed societies tend to shift away from survival values towards self-expression values emphasising individual freedom. (I have discussed national values in more detail in a previous post.) An emphasis on personal freedom may also bring about greater sexual permissiveness.

Truth will come out in the end. 

Conclusion
Differential-K theory makes predictions about racial differences in sexual attitudes that are contradicted by research evidence. Differential-K theory assumes that a fast life history strategy should be associated with greater sexual permissiveness, but this does not appear to be the case. Harsh environments that foster a fast life history strategy due to shorter life expectancy and higher infant mortality are more likely to be associated with sexual restraint and monogamy. Conversely, sexual permissiveness may become less risky in resource-rich environments conducive to a slow life history strategy. Furthermore, differential-K theory is based on the premise that cultural differences in sexual attitudes and behavior are based on inherent differences between racial groups. Specifically, Dutton et al. propose that racial differences in androgen levels are responsible for these variations. However, there is considerable evidence that sexual attitudes and behavior may be linked to social and environmental factors that may be distinct from biological racial differences.

Please consider following me on Facebook, Google Plus, or Twitter.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted as long as a link to the original article is provided. 

This article also appears on Psychology Today on my blog Unique - Like Everybody Else.

Image credits
Venus and an Organist and a Little Dog, 1550 by Titian.
Truth rising out of her well to shame mankind, by Jean-Léon Gérôme, 1896. 

References
Barber, N. (2008). Cross-National Variation in the Motivation for Uncommitted Sex: The Role of Disease and Social Risks. Evolutionary Psychology, 6(2).
Dutton, E., van der Linden, D., & Lynn, R. (2016). Population differences in androgen levels: A test of the Differential K theory Personality and Individual Differences, 289-295
Schmitt, D. P. (2005). Sociosexuality from Argentina to Zimbabwe: a 48-nation study of sex, culture, and strategies of human mating. Behavioral and Brain Sciences, 28, 247-311.
Schmitt, D. P., & 118 Members of the International Sexuality Description Project. (2003). Universal Sex Differences in the Desire for Sexual Variety: Tests From 52 Nations, 6 Continents, and 13 Islands. Journal of Personality & Social Psychology, 85(1), 85-104.

ResearchBlogging.org

Population Differences in Androgens Fail to Validate Richard Lynn's Claims about Racial Differences in Penis Size


In the first part of this post, I discussed a recent study (Dutton, van der Linden, & Lynn, 2016) that attempted to empirically test the predictions of differential-K theory. To recap briefly, this theory proposes that racial groups differ in their preferred reproductive strategies and that as a result they differ in a wide range of physical and psychological characteristics including intelligence, personality, sexual behavior and attitudes, and even penis length. These differences are thought to derive from group differences in androgens (male hormones such as testosterone). According to the theory, African populations should have the highest androgen levels, followed by Caucasians, then Asians. The study tested five presumed markers of androgens: CAG repeats on the AR gene; amount of androgenic body hair, specifically, mid-phalangeal hair (i.e. on the middle digit of the fingers); national prostate cancer incidence; and two measures of sexual behavior, specifically, number of partners, and annual frequency of sex. The results found were not entirely in accord with the theory’s predictions, as Africans were not significantly higher than Caucasians in CAG repeats on the AR gene, and were lower than Caucasians in androgenic body hair and prostate cancer incidence. The authors were not able to compare Africans with the other groups on sexual behavior, as they had no African data. They did find that Caucasians were higher than Asians in all androgen markers, which they took in support of their theory. In this post, I will discuss how the authors of the study interpreted their anomalous results, then respond to Dutton’s claims that his results support the validity of Lynn’s penis data and explain why his results actually contradict this.

Dutton et al. acknowledged that the results for androgenic hair and prostate cancer do not support their hypothesis, and offered some tentative explanations. In relation to prostate cancer, they suggest that differences in general diet might be a factor, as people in Western countries consume more dairy products and these have been linked to prostate cancer. Additionally, people in Western countries tend to be more obese, another risk factor for cancer. They point out that African American men have a 30% higher rate of prostate cancer Caucasians even after taking into account obesity, while Asian Americans have lower rates than these two groups. They also cite a 1986 study that found that African American men have 15% levels of freely circulating testosterone than Caucasian Americans and that this might contribute to their higher prostate cancer rates. They acknowledge that differential access to medical care in the US might affect these results and that African American men might not be fully representative of current Sub-Saharan Africans.

Prostate cancer incidence and its relationship with testosterone is a complex subject that I am not able to review fully. However, I would like to briefly point out some research evidence that I am aware of that may be of interest. Environmental factors including conflict with the law can influence testosterone levels. Studies that take into account conflict with the law comparing African American and Caucasian American men have found that these groups had similar testosterone levels (Zitzmann & Nieschlag, 2001). This same paper stated that higher levels of prostate cancer in African Americans do not appear to be due to differences in testosterone levels. Additionally, although Dutton et al. report a substantial non-significant correlation between AR CAG length and prostate cancer incidence across nations, prior research has found little evidence for an association between CAG length and risk of prostate cancer at an individual level (Lange et al., 2008). Although prostate cancer may well be a marker of androgen levels, the causes of variations between ethnic groups in incidence rates are complicated by local environmental factors, which makes it difficult to interpret these ethnic differences. This raises doubts about the usefulness of treating national prostate cancer incidence as a marker of reproductive strategies.

Moving on to androgenic hair, Dutton et al. offer a highly speculative explanation of why Caucasian men have higher rates of body hair than other populations. This was actually my favorite part of the paper as I found it wryly amusing. They state:

The anomaly that Caucasians have the highest levels of androgenic hair and Africans the lowest can only be speculated upon. It has been found that Caucasians, in contrast to the other two populations, retain a small percentage (2–4%) of Neanderthal genes. It has been argued that this may be one of the reasons why Caucasians are unexpectedly hairy (e.g. Sankararaman, et al., 2014).

Note the reference cited in the last statement. There is a slight problem with this explanation, specifically, that the facts stated are completely wrong. Here is what Sankararaman, et al. (2014) actually had to say on the subject:

the proportion of the genome with confidently inferred Neanderthal ancestry has a mean of 1.38% in east-Asian and 1.15% in European populations consistent with previous reports of more Neanderthal ancestry in east-Asian than in European populations... 
We do not detect tissue-specific expression patterns; however genes involved in keratin filament formation and some other biological pathways are significantly enriched in Neanderthal ancestry in European populations, east-Asian populations, or both. Thus, Neanderthal alleles that affect skin and hair may have helped modern humans to adapt to non-African environments. [Emphasis added.]

Well, so much for that theory about Caucasians being hairy cavemen. It is not clear at all how the data for androgenic hair are supposed to fit in with differential-K theory.



Dutton et al. argued that the pattern of correlations between the androgen indicators provides evidence of their validity. In his conference paper, Dutton also stated that these androgen indicators are correlated with the penis length data used in Lynn's (2013) paper. He claims that this supports the trustworthiness of Lynn’s data. Dutton’s argument might possibly have some merit if certain assumptions were met. In principle, two variables that are closely related to each other should be correlated and this is known as convergent validity. Hence, a strong correlation between two variables is often taken as evidence of convergent validity. However, correlational analyses are based on the assumption that two variables have a linear relationship (i.e. the data points should form a roughly straight line pattern). When variables have a non-linear relationship, use of correlations may give a misleading account of how they are related. As I will show, Dutton’s data has some problems with non-linearity that affects its convergent validity.     

Furthermore, Dutton groups national data into three racial categories which are supposed to form a distinct hierarchy of African > Caucasian > Asian. Hence, in order to argue that the data on androgen markers provides convergent evidence of the trustworthiness of Lynn’s data, both sets of data should follow the same hierarchical pattern. However, they clearly do not.

Dutton et al. report an impressively large correlation between androgenic hair and prostate cancer. However, an inspection of the scatterplot provided (reproduced below) shows that the relationship between the two variables appears decidedly non-linear. Among Caucasian countries there is not much variance in the percentage of androgenic hair yet there is much more variance in prostate cancer incidence. This means that among these countries there is basically no correlation between the two variables. Among the African and Asian nations, prostate cancer incidence has a more restricted range, but there is somewhat more variance in the percentage of androgenic hair. So again, there is basically no correlation between the two variables among these countries either. Essentially, Caucasian countries have higher rates of both androgenic hair and prostate cancer compared to the other countries, but this does not mean that there is a linear correlation between these two variables, even though they are both supposed to be markers of androgens. This suggests that whatever underlies differences in national rates of prostate cancer is not related in a simple linear way to whatever underlies national differences in androgenic hair.

Click to enlarge

According to Lynn’s paper, data he derived from the worldpenis site indicated that men from African nations had the longest penises, followed by Caucasians, followed by Asians. Lynn argued that this was in accord with the predictions of differential-K theory. In my rebuttal, I argued that because the data on this site was compiled by an anonymous source and it is not clear how it was derived or how valid it is, I could not see why a serious scholar would rely on such a source of information or draw any conclusions from it. In his paper, Lynn himself admits that the site is not a peer-reviewed source, so I can only wonder why he used it in the first place rather than drawing on primary research papers published in reputable journals. In his conference presentation, Dutton states (see slide 5) that I pointed out ‘minor mistakes’ on the website in order to ‘ridicule’ Lynn. Perhaps, he and I have different opinions about what makes a mistake ‘minor’. Without going into all my original criticisms here, among these ‘minor’ mistakes I noted that some of the research papers the website cites as sources do not even exist.[1] Hence, I find it natural to wonder what else on this site has been simply made up. 

Additionally, some of the penis sizes stated for particular countries do not match the numbers provided in the genuine papers cited as data sources.[2] For some other countries it is not clear what sources there even are. Personally, I think these issues are cause for serious concern, but perhaps I am overstating how serious they are? Dutton stated that the data from his study on androgen markers correlates with the penis length data, indicating that the latter can be trusted. The argument here seems to be that penis length should be correlated with other androgen markers, and that if the penis length data can be statistically predicted from the androgen data then the former is probably valid, or at least in the right general direction. Dutton did find moderate statistical correlations between androgen markers and Lynn’s penis data, therefore, he argues, Lynn’s data passes muster.

However, there is a problem with this argument. As explained earlier, Lynn’s results and two of Dutton et al.’s results (prostate cancer incidence and androgenic hair) are incompatible.
For penis length, Lynn found: African > Caucasian > Asian.
For androgenic hair, Dutton et al. found: Caucasian > Asian > African.
For prostate cancer incidence, Dutton et al. found: Caucasian > Asian = African.
I do not see how the pattern for the latter two results can be used to validate the first one. More specifically, if prostate cancer and androgenic hair accurately predicted penis length, then we would expect African men and Asian men to both have smaller penises than Caucasian men, contrary to what Lynn claimed. (Please note, I am not asserting anything at all about actual differences in penis length between races, because I do not have sufficient data. I am commenting on the methodology used to support such claims.) To put this another way, although Lynn’s penis data is correlated with the other two variables, the actual relationship between the former and the latter is non-linear, so the correlations do not allow valid predictions.

Scientists should use reliable data sources lest they get caught up in flights of fancy or end up groping in the dark like jackasses. 

It might be helpful to illustrate this with a chart showing the relationship between national prostate cancer rates and the penis data. Dutton does not provide relevant scatterplots, so I created one in Excel using prostate cancer data from Haas, Delongchamps, Brawley, Wang, & de la Roza (2008) and 2011 penis length data from the website Lynn used. 


There is a linear correlation of r = .34 between prostate cancer incidence and penis length. This is not statistically significant due to the low sample size (only 21 countries), but Dutton has argued that correlations of this value are ‘substantial’ (slide 7). Does this mean that prostate cancer incidence can predict average national penis length? If there was a clear linear relationship between the two it might, albeit crudely. But look at the scatterplot above. The Asian countries cluster on the lower left, the Caucasian in the middle, and the African on the lower right. The relationship between the two variables is clearly non-linear. This is because the countries with the highest incidence of prostate cancer are Caucasian, yet these countries are mostly intermediate in penis length. A similar result would appear if one were to graph the relationship between androgenic hair and penis length, as Caucasians also had the highest rates of the former. At the risk of belabouring the obvious, it simply does not make any sense to say that because Lynn’s penis length data is correlated with these other variables that this means that the former can be trusted. Dutton’s and Lynn’s measures do not follow the same patterns so correlations between them can be explained as statistical artifacts of using linear tests on non-linear data. If scholars really want to know if there are differences between these groups in penis length, the sensible thing to do would be to find better data sources rather than trying to validate an anonymous database using crude predictions based on loosely related variables.

Conclusion
Differential-K is a grand theory that aims to explain a very wide range of human population differences. However, it appears to be out of step with the available data in some respects. One prediction of this theory is that racial groups should differ from each other in consistent ways on a number of factors that are supposed to be markers of androgen levels. Yet, an empirical test of this prediction was not able to confirm the expected racial hierarchy. The relationship between things that are supposed to be indicators of androgen levels, such as androgenic hair and prostate cancer might be too complex to be represented in a simple linear way. Furthermore, how these things are supposed to be related to reproductive strategies is not entirely clear. Dutton et al. found differences in sexual behavior between Caucasian and Asian populations, but provided no data on Africans. The data source they used, an internet survey by Durex, is not a peer-reviewed source and may be of questionable validity. There are more scientific sources of information that could shed light on cross-national differences in sexual behavior, and I make some relevant observations about these in a follow-up post. Richard Lynn (2013) argued that racial differences in penis length provide evidence of corresponding racial differences in sexual restraint, although critics consider this view to be naïve. Although Edward Dutton argued that his research results support the validity of Lynn’s earlier claims about racial differences in penis size, a closer examination of his own data contradicts this. Differential-K theory has not fared well and may be an overly simplistic theory that attempts to explain too much with too little.

Please consider following me on Facebook, Google Plus, or Twitter.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted as long as a link to the original article is provided. 

This article also appears on Psychology Today on my blog Unique - Like Everybody Else.





Footnotes
[1] Since writing my original article, the penis website has added measurement data from various countries to the sidebar. Most of these cite genuine research, but there is one on ‘Ecuadorian measurement data,’ that cites a paper called ‘Study of penile dimensions in healthy Ecuadorian men of multiple ethnicities’ supposedly published in Andrologia. Included is a professional looking Abstract along with a pair of impressive looking graphs. However, Andrologia has never published anything by this name, and the paper appears not to exist. The whole citation looks like an elaborate hoax. What does this say about the scholarly value of this website?
[2] See the blog Ethnic Muse for more detailed information.

References
Dutton, E., van der Linden, D., & Lynn, R. (2016). Population differences in androgen levels: A test of the Differential K theory Personality and Individual Differences, 289-295

Haas, G. P., Delongchamps, N., Brawley, O. W., Wang, C. Y., & de la Roza, G. (2008). The Worldwide Epidemiology of Prostate Cancer: Perspectives from Autopsy Studies. The Canadian journal of urology, 15(1), 3866-3871.
Lange, E. M., Sarma, A. V., Ray, A., Wang, Y., Ho, L. A., Anderson, S. A., . . . Cooney, K. A. (2008). The androgen receptor CAG and GGN repeat polymorphisms and prostate cancer susceptibility in African-American men: results from the Flint Men's Health Study. J Hum Genet, 53(3), 220-226.
Lynn, R. (2013). Rushton’s r–K life history theory of race differences in penis length and circumference examined in 113 populations. Personality and Individual Differences, 55(3), 261-266. doi: http://dx.doi.org/10.1016/j.paid.2012.02.016
Sankararaman, S., Mallick, S., Dannemann, M., Prufer, K., Kelso, J., Paabo, S., . . . Reich, D. (2014). The genomic landscape of Neanderthal ancestry in present-day humans. [Letter]. Nature, 507(7492), 354-357. doi: 10.1038/nature12961
Zitzmann, M., & Nieschlag, E. (2001). Testosterone levels in healthy men and the relation to behavioural and physical characteristics: facts and constructs. European Journal of Endocrinology, 144(3), 183-197. doi: 10.1530/eje.0.1440183

ResearchBlogging.org

Saturday, December 31, 2016

Population Differences in Androgens Fail to Support Differential-K Theory

According to a controversial theory, Differential-K, major human populations differ in a systematic way along a continuum of psychological and physical characteristics, based on their preferred reproductive strategies. These presumed characteristics include intelligence, personality, sexual behavior and attitudes, and even, according to Richard Lynn, penis length. (I discuss this last topic in detail here.)


Discussion of racial differences can become heated, but hopefully cooler heads will prevail. 

Differential-K is a very bold theory, sweeping in scope, developed by J.P. Rushton that aims to connect a wide range of apparently unrelated human variables in a single theoretical framework (Meisenberg & Woodley, 2013). According to this theory, variations in the generally preferred reproductive strategy of particular human groups have affected the during the course of their evolution. A fast life history strategy involves high mating effort and the production of a larger number of children with less intensive investment in each one. This kind of strategy is suited to conditions in which life expectancy is fairly short and infant mortality is high, and so people must aim to reproduce fairly quickly and frequently to ensure they pass on their genes to the next generation. In contrast, a slow life history strategy involves less mating effort and more parenting effort, with the production of fewer children and more intensive investment in each one. Differential-K theory assumes that fast and slow strategies are each associated with a whole suite of human characteristics that differ not only among individuals but between whole populations. Specifically, East Asian peoples are assumed to have the slowest strategy, while sub-Saharan African peoples have the fastest strategy, and Caucasian peoples are in-between, although closer to Asians than Africans. Naturally this theory has been the focus of intensely heated controversy due to its assumptions about racial differences. Supporters of this theory argue that it has the virtue of being parsimonious, as it purports to explain a wide range of disparate phenomena with a single principle (Meisenberg & Woodley, 2013). Critics have argued that it is makes arbitrary assumptions about what characteristics are supposed to be associated with fast and slow strategies respectively in order to create a hierarchy of humanness, and that Rushton and colleagues have used cherry-picked and distorted selections of evidence in support of this theory (Weizmann, Wiener, Wiesenthal, & Ziegler, 1990). Due to the breadth and complexity of the topic, I will not attempt anything like a comprehensive review here. Instead I will focus on a recent paper (Dutton, van der Linden, & Lynn, 2016) that attempts to test if racial differences in androgen levels follow a pattern predicted by differential-K theory.

According to Richard Lynn, life history strategy might be regulated by androgens (i.e. masculine hormones like testosterone) such that higher androgen levels are associated with a faster life history strategy. Higher androgen levels are associated with greater aggression and competitiveness, as well as with increased short-term mating. Populations with a slower life history strategy have historically had a greater need for cooperation among males due to the harsher environments in which they lived and this led to reduced androgen levels. Hence, Lynn proposed racial differences in androgen levels, with sub-Saharan Africans having the highest levels, followed by Caucasians, followed by East Asians. This theory implies that high androgen populations should have higher levels of interest in short-term sexual relationships (a marker of mating effort) while lower androgen populations should be more sexually restrained. Dutton, et al. (2016) aimed to test this theory by examining data on androgen indicators in a range of representative countries. The lead author of this paper, Edward Dutton, has also provided information about this research in a conference presentation that can be viewed here. I was personally interested to note that Dutton mentions my name in this presentation in regard to an article I wrote a few years ago critiquing a paper by Richard Lynn (2013) in which the latter used data from an anonymous website as evidence for race differences in penis length (see slide 5). According to Dutton, I 'ridiculed' this paper because of ‘minor mistakes’ on the website that I argued invalidated Lynn’s claims. Dutton goes on to assert that the results of his research demonstrate that ‘Lynn’s penis data can indeed be trusted.’ I still stand behind my original criticisms and respond to Dutton’s comments in the follow up to this article. In the meantime, I will see if I can keep the ridicule to a minimum, or at least stay within the bounds of civil discussion.

To test their theory, Dutton et al. identified five androgen-level indicators for which national-level data was available. One of these was CAG repeats on the AR. According to a review (Minkov & Bond, 2015), more CAG repeats are associated with androgen insensitivity, while fewer repeats are linked to more sexual partners and violent and impulsive behavior.[1] The other indicators were: amount of body hair, specifically, hair on the middle digit of the fingers (mid-phalangeal hair); national prostate cancer incidence; and two measures of sexual behavior, specifically, number of partners, and annual frequency of sex. Data on sexual behavior derived from a 2005 internet survey by Durex, the condom manufacturer. Unfortunately, because the Durex survey included only one African country[2], it was only possible to make comparisons between Caucasians and Asians. I would like to point out that this survey is not scientific and is not peer-reviewed, so the quality of its methodology is unclear. Internet surveys are not necessarily representative of the population they are derived from[3], so the results may not be valid indicators of rates of sexual behavior in the respective countries. Racial categories were decided based on the main group within each country. For Dutton et al.'s purposes, Northeast Asian (e.g. China) and Southeast Asian (e.g. Malaysia) countries were classified as East Asian (or just Asian for brevity), while European, North African and several South-Asian (e.g. India) countries were classified as Caucasian. Sub-Saharan African countries were classified as such, and simply referred to as African for brevity.

I will briefly summarise the results and then provide some comments. The five androgen indicators were correlated in the expected directions with each other, and 7 out of 10 of the correlations were significant. The authors argue that these inter-correlations support their hypothesis that these are actually manifestations of androgen levels. Although not mentioned in the published paper, Dutton’s conference presentation notes that the five androgen indicators were correlated with the penis length data used by Lynn. He asserts that because all these measures are correlated with each other that this ‘demonstrates that Lynn’s penis data can indeed be trusted’ (slide 7).

To show the outcomes for the group comparisons I have adapted the authors’ results into the table below.

Group Comparison of the Androgen Indicators
East Asians
Caucasians
Sub-Saharan Africans
M
(+SD)
M
(+SD)
M
(+SD)
AR CAG length
23.1
0.35
21.31
1.28
20.2
0.93
N countries
8
24
4
Androgenic hair
38.71
5.72
62.49
10.04
13.63
7.69
N countries
14
71
22
Prostate cancer
5.4
4.9
64.4
19.36
16.88
10.41
N countries
4
10
12
Sex frequency
80.75
16.67
107.18
11.95
NA
23.22
N countries
8
22
Sex partners
6.48
2.83
9.21
2.81
NA
5.53
N countries
8
22

In line with the authors’ expectations, statistical tests showed that East Asian populations had lower androgen markers than Caucasians on all five indicators. However, the remaining results were not in line with differential-K theory because African populations did not have significantly higher androgen markers than Caucasian ones on any measure. For AR CAG length, the difference between Caucasians and Africans was in the expected direction but was not statistically significant, although both groups had greater CAG lengths than Asians. Africans had significantly lower rates of prostate cancer than Caucasians (and did not significantly differ from Asians) and the lowest percentage of androgenic hair.   

To recap, differential-K theory predicts that African populations should have the highest levels of androgens, followed by Caucasians, followed by Asians, and that Caucasians should be closer to Asians than Africans. Of the three comparisons that included Africans, only the results for AR CAG repeats come close to this pattern, although the difference between Africans and Caucasians was not significant, and Caucasians were actually slightly closer to Africans than to Asians. To be fair the non-significance of this result might be attributable to the small number (only four) of African nations in the analysis. A number of previous studies have actually found that people of African descent on average did have shorter CAG repeats than other peoples (Ackerman et al., 2012; Esteban et al., 2005; Kittles et al., 2001; Lange et al., 2008). However, whether this actually indicates anything about the life history strategy of different populations remains questionable. The two other androgen indicators for which African data was available follow a completely different pattern. For androgenic hair, Caucasians have the highest rate, followed by Asians, then Africans. For prostate cancer, Caucasians have the highest rate, followed by Asians and Africans, who do not significantly differ.

According to differential-K theory, Africans and Asians are supposed to be at opposite ends of the life history strategy continuum. However, if androgens are a marker of life history strategy, then based on two of the indicators Africans and Asians would appear to be at the slow end of the continuum and Caucasians at the fast end. This is very difficult to reconcile with differential-K theory. Dutton et al. provide no African data on sexual behavior, so they cannot say whether Africans are less restrained than Caucasians as predicted by their theory. However, they found that Africans were more similar to Asians than Caucasians on two of the androgen indicators. According to the logic used by Dutton et al., if sexual behavior is correlated with androgen levels, then it would be reasonable to expect Africans to be more like Asians in respect to sexual behavior as well. However, such a result would also be contrary to the predictions of their theory. On the other hand, the data source Dutton et al. used for sexual behavior is not a scientific one, so it might not even be valid anyway.

In my follow-up post, I discuss how Dutton et al. interpret their anomalous results, then respond to Dutton’s claims that his results support the validity of Lynn’s penis data and explain why his own results actually contradict this. My third post in this series address the issue of differences in sexual attitudes across major human populations, and shows that these do not fit the predictions of Differential-K theory.

ResearchBlogging.org

Please consider following me on Facebook, Google Plus, or Twitter.

© Scott McGreal. Please do not reproduce without permission. Brief excerpts may be quoted as long as a link to the original article is provided. 

This article also appears on Psychology Today on my blog Unique - Like Everybody Else.

References
Ackerman, C. M., Lowe, L. P., Lee, H., Hayes, M. G., Dyer, A. R., Metzger, B. E., . . . The Hapo Study Cooperative Research, G. (2012). Ethnic Variation in Allele Distribution of the Androgen Receptor (AR) (CAG)n Repeat. Journal of Andrology, 33(2), 210-215. doi: 10.2164/jandrol.111.013391
Aluja, A., García, L. F., Blanch, A., & Fibla, J. (2011). Association of androgen receptor gene, CAG and GGN repeat length polymorphism and impulsive-disinhibited personality traits in inmates: the role of short-long haplotype. Psychiatric Genetics, 21(5), 229-239.
Cheng, D., Hong, C.-J., Liao, D.-L., & Tsai, S.-J. (2006). Association study of androgen receptor CAG repeat polymorphism and male violent criminal activity. Psychoneuroendocrinology, 31(4), 548-552. doi: 10.1016/j.psyneuen.2005.11.004
Comings, D. E., Muhleman, D., Johnson, J. P., & MacMurray, J. P. (2002). Parent–Daughter Transmission of the Androgen Receptor Gene as an Explanation of the Effect of Father Absence on Age of Menarche. Child Development, 73(4), 1046-1051. doi: 10.1111/1467-8624.00456

Dutton, E., van der Linden, D., & Lynn, R. (2016). Population differences in androgen levels: A test of the Differential K theory Personality and Individual Differences, 90, 289-295

Esteban, E., Rodon, N., Via, M., Gonzalez-Perez, E., Santamaria, J., Dugoujon, J.-M., . . . Moral, P. (2005). Androgen receptor CAG and GGC polymorphisms in Mediterraneans: repeat dynamics and population relationships. J Hum Genet, 51(2), 129-136.
Kittles, R., Young, D., Weinrich, S., Hudson, J., Argyropoulos, G., Ukoli, F., . . . Dunston, G. (2001). Extent of linkage disequilibrium between the androgen receptor gene CAG and GGC repeats in human populations: implications for prostate cancer risk. Human Genetics, 109(3), 253-261. doi: 10.1007/s004390100576
Lange, E. M., Sarma, A. V., Ray, A., Wang, Y., Ho, L. A., Anderson, S. A., . . . Cooney, K. A. (2008). The androgen receptor CAG and GGN repeat polymorphisms and prostate cancer susceptibility in African-American men: results from the Flint Men's Health Study. J Hum Genet, 53(3), 220-226.
Lynn, R. (2013). Rushton’s r–K life history theory of race differences in penis length and circumference examined in 113 populations. Personality and Individual Differences, 55(3), 261-266. doi: http://dx.doi.org/10.1016/j.paid.2012.02.016
Meisenberg, G., & Woodley, M. A. (2013). Global behavioral variation: A test of differential-K. Personality and Individual Differences, 55(3), 273-278. doi: http://dx.doi.org/10.1016/j.paid.2012.04.016
Minkov, M., & Bond, M. H. (2015). Genetic polymorphisms predict national differences in life history strategy and time orientation. Personality and Individual Differences, 76, 204-215. doi: http://dx.doi.org/10.1016/j.paid.2014.12.014
Rajender, S., Pandu, G., Sharma, J. D., Gandhi, K. P. C., Singh, L., & Thangaraj, K. (2008). Reduced CAG repeats length in androgen receptor gene is associated with violent criminal behavior. International Journal of Legal Medicine, 122(5), 367-372. doi: 10.1007/s00414-008-0225-7
Weizmann, F., Wiener, N. I., Wiesenthal, D. L., & Ziegler, M. (1990). Differential K theory and racial hierarchies. Canadian Psychology, 31(1), 1-13. doi: 10.1037/h0078934




Footnotes
[1] Actually, the evidence reviewed by Minkov and Bond is somewhat equivocal. One of the two studies comparing violent offenders with a community sample, found no difference between the two groups in CAG repeat length (Cheng, Hong, Liao, & Tsai, 2006) while the other one did (Rajender et al., 2008). A study looking at impulsive personality traits (Aluja, García, Blanch, & Fibla, 2011) found that although an inmate sample was higher on a range of impulsive personality traits compared to a community control group, the two groups did not differ in CAG repeat lengths. A study (Comings, Muhleman, Johnson, & MacMurray, 2002) cited by Minkov and Bond as evidence linking CAG repeats to lifetime number of sexual partners did not actually assess CAG repeats at all, but a separate structure of the AR gene called the GGC polymorphism. Additionally, the study sample consisted of men being treated for substance abuse and did not have a healthy control group. 
[2] South Africa
[3] For example, countries differ in their levels of internet access, which can affect who responds to the survey. Additionally, people who choose to respond to internet surveys, especially ones about sex, and in particular one that is hosted on the website of a condom manufacturer, may not be typical of people in the general population. The Durex website does not provide any information addressing these issues. (Thanks to Petra Boynton for highlighting these concerns.)