final version 30.9.03
Elina Haavio-Mannila, J.P. Roos, Anna Rotkirch
Do rich men have most sex?
Gender, capital and sexual activity in four countries
To Ulla Björnberg
As is well known, Darwin showed that male intrasexual competition and female heterosexual choice were the driving forces of sexual selection, which is a special variant of natural selection. Darwin stressed that males compete for sexual access to females by displaying signs of resourcefulness, fitness, and other valued abilities. In species where males do not participate in child rearing, male sexual seduction is all about ‘showing off’. The classic examples among animals are the male peacock=s tail and mammoth’s tusks, which both are extremely impractical and even hampering. They appear to contradict natural selection in the sense of adaptations to the ecological environment, but become understandable from the perspective of sexual selection. Females developed a preference for long tails or huge tusks and males consequently excelled in these attributes because they served as fitness indicators, instruments that help to judge the partner’s qualities. Such fitness indicators can be physical, psychological or material. They must be permanent, costly, difficult to acquire and hard to fake. (Darwin 1871; Buss 1996; Cronin 1991.)
For all species, the willingness to discriminate in the choice of sexual partners is dependent on the cost of reproduction (Trivers 1972). If the individual does not take part in the rearing effort, it has no need to be very selective. By contrast, if it will be rearing its offspring for several years and will preferably need a partner to participate in the effort, it will be more selective in the choice of the partner.
Human sexual choice is more complicated than the mammoths’, as men also participate in childrearing, albeit to a varying degree and usually significantly less than women (Björnberg 1998). When looking for long-term sexual partners, both men and women value kindness, reliability, and commitment. Respect gained from others is also a good sign of sexual value. Here, sexual selection is reciprocal – both partners strive to charm, seduce and eventually make the other commit him- or herself. Physical attributes are important, but so are signs of material well-being, good social networks, and emotional and intellectual abilities. It may even be that our brains originally developed as part of this two-way mating game. In this view, the brain’s ultimate reason of existence was not to work and produce, as the classical explanation goes, but to enjoy and seduce. You need big brains in order to tell jokes, make compliments, and confide in each other, as well as in order to evaluate and assess these jokes, compliments and confidences (Miller 2000).
Concerning short-term sexual relations, however, evolutionary theory predicts more gendered sexual strategies (a term that refers to general behavioural patterns but does not exclude personal, social and cultural variations.) Women are cross-culturally on average relatively selective and favour partners with impressive fitness indicators such as social status, strength and bravery. Being rich and famous increases the attractiveness of both sexes, but is more highly valued in men than in women. (Campbell 2002, 103 and 179.) Men are on average less selective than women and favour especially youth. This is true also in societies where contraceptives have cut the original evolutionary link between intercourse and pregnancy. Our evolved emotional dispositions continue to operate even today, and often unconsciously. For instance, women on average report feeling more sexual desire at the period of ovulation (Hrdy 1999, 220).
Humans form monogamous relationships but are also, as evolutionary theory puts it, a “mildly polygynous” species. In most human societies, men with social and economic power have had more wives and mistresses. Because our sex ratio is nearly 50-50, this meant that the men with least social and material assets ended up without any wives or mistresses. Evolutionary theory thus predicts a polarisation between men with regards to the number of sexual partners and marriages: some have much and some have none. Women have a more equal intrasexual distribution, as practically every woman can find a male who wants to mate with her if she accepts him, and most women also find a long-term heterosexual partner if they are not too choosy.
Already these traditional evolutionary premises of human sexual behavior - monogamy but some degree of polygyny – provide complicated sociological outcomes. Thus feminists disagree on which arrangement is more ‘pro-women’: monogamy, which is the best way to tie husbands (and his relatives) to marriage and child rearing (Hrdy 1999, 252), or polygyny in a liberal society, in which women choose their sexual partners more freely and can rear their children in a community of female friends and maternal relatives (Hurley 2002; for the loyalties of matri- vs. patrilineal relatives see Sarmaja 2003.) Contemporary Nordic societies show signs of both arrangements.
Recent primatological findings have further complicated the picture, especially with regards to female sexual choice. Contrary to classic assumptions, dominant male apes do not always have greater sexual success (in terms of copulations) or greater overall reproductive success (in terms of numbers of surviving grandchildren). The empirical evidence of ape communities is so far inconclusive, ranging from no correlation to quite high correlation between dominance and paternity. One study claimed that the causal relationship may be the opposite, so that leaders do not become attractive but attractive males become leaders. Thus some female Macaque apes favoured young males before they reached the status of alpha males. (Campbell 2002, 66-67.) Female ‘infidelity’ is also much more common than assumed. A few years ago, the first DNA paternity tests among chimpanzees showed that over half of the infants were sired by males outside the community - a fact the human researchers and probably also the dominant males had been totally unaware of (Hrdy 1999, 85).
In our view, family sociology could engage with evolutionary theory through empirical studies of how sexuality and families are formed under various socio-economic and cultural conditions. Surely, the task is no longer to quarrel over what is ‘biological’ and what is ‘social’, but to describe the rich kaleidoscope of ways in which innate dispositions are activated, transformed and realised under various conditions (Laland and Gillian 2002; Roos and Rotkirch 2003).
It is also important to stress that even if something can be said to be ‘natural’ in the sense of being a genetically transmitted disposition, it does not follow that this would be socially or ethically justified. Evolution is in itself ‘witless’ and has no inherent teleological or moral agenda (for a feminist discussion of this, see Zuk 2000). Our evolved physiology also easily leads to maternal pain and death during childbirth, but although this is natural it is not ideal. The naturalist fallacy that equates natural with desirable is unfortunately very common among both lay people and sociologists (see e.g. the discussion on results showing that men on average desire more partners than women: AThe idea that male promiscuity is hardwired and therefore ‘normal’ drew swift and furious criticism@ (www.msnbc.com/news/946836.asp).
In this paper, we test the evolutionary prediction that power and status are important criteria of sexual selection of males, especially concerning short-term relationships in contemporary Northern European societies. We are also interested in whether increased gender equality in a society reflects itself in more similar gendered sexual strategies, so that female social status would affect women’s sexual selection. The main emotional dispositions that evolved during prehistoric times cannot be expected to have changed radically during the last hundred or fifty years. However, these dispositions may be more or less environmentally sensitive. For instance, the disposition of finding women in the beginning of their reproductive life to be most sexually attractive is apparently not very sensitive to environmental factors and is universally found in all cultures. By contrast, the disposition to discriminate between children according to sex differs with social prospects: in a society where boys have worse marriage prospects than girls, parents treat girls better (even if their articulated ideology may favour sons) (Cronk 1990). Thus the dispositions to correlate wealth and power with sexiness do not need to be strongly tied to the male sex. There may exist a general ‘power is sexy’-disposition, that easily accommodates to incorporate women when a society has more women in positions of power. However, the question is not so much ‘are women in power sexy?’ (we know many think they are) but ‘what kind of partners do women in power want?’ If a society’s gender equality leads to increased similarity in sexual strategies, then women in power should both want and find more sexual partners, or men in power should want less partners. Previous research suggests that women in power, just like other women, prefer partners who are socio-economically their equal or superior and thus select quite heavily among possible pretenders (Buss 1996). We will question this conclusion by looking at how gender is related to reported sexual strategies and power in four different regions.
The data: six surveys from four countries
This chapter is based on survey data gathered in the 1990s in four regions of the Baltic Sea Area: Sweden, Finland, Estonia and the city of St. Petersburg in Russia. The data consists of representative surveys of adult populations in Sweden in 1996 (N=2,810), Finland 1971 (2,152), Finland in 1992 (2,250) and 1999 (N=1,496; the two data sets are combined), Estonia (1,034) and St. Petersburg 1996 (2,080). People were interviewed face to face and/or filled a structured survey questionnaire (Kontula & Haavio-Mannila 1995; Lewin et al. 1998; Haavio-Mannila 2003).
One tumbling stone for empirical assessments of evolutionary predictions has been the difficulties in operationalising ‘dominance’ or ‘social status’ in modern societies. For instance, social status is usually measured as level of education or income (Freese 2000). This excludes e.g. the leaders of local communities or subcultures. Our article does not escape this problem, although we try to approach ‘power’ in a multidimensional way. We have operationalised sexual success as the number and quality of sexual partners and relationships and selected four variables to be explained: (1) number of sexual partners during lifetime, (2) having had parallel relationships, (3) finding intercourse pleasurable, and (4) satisfaction with sexual life as a whole. The variables which supposedly predict sexual activity are different types of capital or power: (1) economic, (2) social and (3) cultural capital.
Information on economic capital - monthly income after taxation - was unfortunately available only from Finland. In 1971, it consists of family income, in the 1990s of the individual income of the respondent. Cultural capital was defined on the basis of three categories of general education: low, medium and high. Social capital was measured on the basis of the present occupation and in Petersburg also on the basis of type of employer. Thus it here comes close to and partly intertwines with economic capital. The classifications varied in different countries.
In Sweden a three level social group classification was available (cf. Lewin et al. 1998).
I academics, free professions, higher functionaries, landowners etc.
II middle class, functionaries, craftsmen
III workers and assisting personnel
The Finnish social groups were in 1971 defined on the basis of the following question: “It is customary to say that people belong to different social groups or classes. Which of the following groups do you consider yourself to belong to?” The seven categories listed in the questionnaire were combined into four as follows:
I upper or leading social group, upper middle class
II lower middle class
III working class
IV farmer (classes III and IV are in some cases combined)
In Finland 1992, there was an open-ended question on present occupation. The replies were coded on the basis of the occupational classification by Statistics Finland 1987 into three social capital groups:
I Upper white collar employees, other entrepreneurs than farmers
II Lower white collar employees
III Workers, farmers
In 1999, the Finnish social grouping was based on the question: “Into which of the following groups do you consider yourself to belong on the basis of your occupation?” The six alternatives listed were combined into three in the same way as in 1992.
In Estonia, the economically active respondents were divided into three groups:
I White collar employees
II Other entrepreneurs than farmers
III Workers, farmers
The 22 occupational categories in the St. Petersburg study were combined into three occupational status classes as follows:
I Upper non-manual occupations: engineers, specialists in natural and humanistic sciences and fields, medical personnel (physicians, dentists, pharmacists, veterinarians), researchers, teachers, people working in arts, leisure, sports and media, directors, and entrepreneurs
II Lower non-manual occupations: technicians, nurses, office workers, workers in justice and administration (militia, solders, fire workers)
III Manual work occupations: sales, traffic, factory, service and farm workers
In addition, type of employer (state – private enterprise) was used as an indicator of social capital in St. Petersburg.
Age affects sexual activity in the obvious way that younger people are sexually more active than older people regardless of their social position. Age is also in itself related with income and social power. We have therefore checked the relationship between the number of partners and economic and social position in the age categories from 35 upwards, but the results remained the same as among all respondents. In the Figures shown here, data has been adjusted for age.
In the empirical part of this chapter, we look at the effects of economic (income), social (occupational status, in St. Petersburg also type of employer) and cultural (education) capital on the reported numbers of sexual partners, parallel relations, enjoying intercourse, and sexual satisfaction.
The examination of the number of partners in different income categories (Finland only) showed clear relationships between the number of partners and economic capital in the 1990s. The higher the income group, the more partners the respondent reported having, and this was especially clear concerning men. The average number of partners of male respondents ranged from twelve in the lowest income group to eighteen in the highest group (Figure 1). In the case of women, who reported fewer partners, there was a similar tendency. However, women in the highest income category reported fewer partners than those in the next highest. (But we should remember that men tend to exaggerate and women to underreport the number of their sexual partners, Kontula and Haavio-Mannila 1995, 93-94).
Figure 1. Number of Sexual Partners According to Income in Finland
In Finland, the influence of income was stronger in the 1990s than in 1971. This is probably partly explained by the fact that in 1971, the question referred to household income whereas in the 1990s it referred to personal income.
Effect of Social Capital
Next we look at the influence of social capital, here measured as occupational status, on the number of partners. In 1971 in Finland, workers reported most sexual partners (Figure 2). In the 1990s, by contrast, the number of partners of women rose systematically from the lower occupational status group to the higher status group. Among men there was a curvilinear relationship: lower white-collar men reported the highest number of partners.
Figure 2. Number of Sexual Partners According to Social or Occupational Status in Finland
In Sweden, the relationship between social group and the number of partners was linear (Figure 3). In the lowest social group Swedish men reported on average 13 and women 6 partners, in the middle group 14 and 7 and in the highest group 17 and 8 partners, respectively.
Figure 3. Number of Sexual Partners According to Social Group in Sweden
In Estonia, there was a slight difference in the reported number of partners between male workers and entrepreneurs, on the one hand, and male white collar workers, on the other (13 and 16 partners) (Figure 4). Female entrepreneurs reported a higher number of partners (8) than female workers and employees (6). In St. Petersburg, there were no statistically significant differences in the number of sexual partners between the three occupational groups. The tendency was that the number of partners increased with growing occupational status (Figure 5).
Figure 4. Number of Sexual Partners Figure 5. Number of Sexual Partners
According to Occupational Status According to Occupational Status
in Estonia in St. Petersburg
This homogeneity between occupational groups in St. Petersburg was elaborated by classifying the economically active respondents on the basis of type of employer: (1) people working for state, governmental or municipal enterprise or office, (2) people working in a private enterprise or company previously owned by the state and (3) private enterprises never owned by the state. This gave a clear correlation as people working in the private sector reported more sexual partners than those working for the state (Figure 6). Being employed in the Russian private sector can be presumed to indicate both higher wages and higher social prestige than having a state employer.
Figure 6. Number of Sexual Partners According Type of Employer in St. Petersburg
Effect of Cultural Capital
Next we explore the effect of cultural capital, measured as level of education, on the number of partners in the four regions. The number of sexual partners increased linearly with education in Sweden and among Finnish women in the 1990s (Figure 7). Finnish men with middle level of education (especially those who had attended vocational college) reported the highest number of partners. In Estonia and St. Petersburg, there were no significant differences between the educational groups.
Figure 7. Number of Sexual Partners According to Education
In sum, the number of sexual partners correlated clearly with direct indicators of economic capital (personal income in Finland or type of employer in Petersburg). It also correlated to some degree with occupational and educational groups, with the exception of the Finnish surveys where workers or middle class men reported the highest numbers of sexual partners. There were clear gender differences, but more capital increased the number of sexual partners for both men and women, although sometimes with the exception for the women with most capital.
Economic resources also clearly correlated with reported parallel (extramarital) sexual relations, especially those of men. The experiences of having had parallel relations were studied in slightly different ways in the different research sites. From Sweden and from Finland in 1971 we have data on parallel relationships during lifetime. The Finnish data from the 1990s and the Estonia and St. Petersburg data cover reported infidelity during the present marriage or cohabitation.
Higher income entails more parallel relationships both for men and women. In the 1990s, the proportion of reported infidelity of Finnish men rose from twenty percent in the lowest income group to fifty in the highest one. For women it rose from ten to twenty, respectively (Figure 8). In 1971, the trend was very strong for men but also visible among women.
Figure 8. Parallel Relations during the Present Marriage or Cohabitation According to Income in Finland
Effect of social capital
Of the different occupational groups in Finland, white collar men reported more parallel relations than workers and farmers (Figure 9). Among women, the differences between occupations were much smaller and a clear rising trend was evident only in the 1990s.
Figure 9. Parallel Relations According to Social or Occupational Status in Finland
In Sweden, social group differences in reported parallel relations were also significant both among men and women (Figure 10). Only 32 percent of men and 17 percent of women in the lowest group reported having been unfaithful in their lifetime, in the highest group, the respective percentages were 48 and 31. In Estonia, occupational status did not predict statistically significantly parallel relations. Male employees and entrepreneurs had parallel relations slightly more often than male workers and farmers (Figure 11). Among women, entrepreneurs again scored higher than the others.
Figure 10. Parallel Relations during Figure 11. Parallel Relations
Lifetime According to Social during the Present Marriage
Group in Sweden or Cohabitation According
to Occupational Status in Estonia
In St. Petersburg, occupation (worker, lower and upper employee) was not statistically significantly related to having had parallel relations (Figure 12). Women of the upper occupational groups tended to report extramarital affairs slightly more often than the lower status women but this association was reversed among men. When we looked at type of employer, there was a clear correlation, as people employed in private businesses reported most parallel relations (Figure 13).
Figure 12. Parallel Relations during the Figure 13. Parallel Relations during the
Present Marriage or Cohabitation Present Marriage or Cohabitation
According to Occupational Status According to Type of Employer
in St. Petersburg in St. Petersburg
In the Russian data, there is an interesting gender discrepancy concerning parallel relations. Men with the lowest and women with the highest amounts of social capital reported most parallel relationships. The gender gap in parallel relations is thus largest in the lowest occupational group and smallest in the highest one. Higher amounts of social capital appeared to favor more similar behaviour of men and women.
As in the case of sexual partners, women working in private enterprises in St. Petersburg most frequently reported parallel relations. Working class women in originally private enterprises (which had never been a state-owned) exceptionally often reported extramarital relations, but also many upper employees working in these firms had had parallel relations.
In most areas studied, a higher level of education increased the likelihood of reporting parallel relations (Figure 14). The exceptions are St. Petersburg and Finland in 1971, where men with middle level of education most often reported parallel relations.
Figure 14. Parallel Relations during the Present Marriage or Cohabitation (Sweden: During Lifetime) According to Cultural Capital
In sum, the reported number of parallel relationships was clearly related to economic, social and educational indicators in most cases. Like with the number of partners, direct economic indicators showed clearer correlation between status and sexual activity than our occupational and educational indicators did. This was especially evident in Russia, where the level of education often refers to achieved social situation during the Soviet times, while the current type of employer revealed more about the respondents’ actual social situation.
Pleasure from Intercourse and Satisfaction with Sexual Life
With regard to sexual satisfaction, the country differences were considerable. People in Estonia and St. Petersburg found intercourse less pleasurable and were much less satisfied with sexual life than people in Finland and in Sweden (where only overall satisfaction was studied). This is due to differences in living standards and in the access to sexual education and contraceptives between the two Nordic and the two post-socialist countries (Haavio-Mannila and Rotkirch 1998).
In Finland, experiencing intercourse as pleasurable and being satisfied with sexual life as a whole was most characteristic in the higher middle income groups of both men and women (Figure 15). Thus the richest people were not always as pleased with sexual life as the next highest groups.
Figure 15. Sexual Satisfaction According to Income in Finland
Effect of Social Capital
In Finland in the 1990s, male white-collar employees enjoyed intercourse and particularly sexual life in general more than male manual workers and farmers (Figure 16). Interestingly, the 1971 data gave no male occupational difference for reported pleasure in intercourse (all three occupational groups estimated 4.4). The women gave slightly lower estimates in 1971 than in the 1990s (3.9, 4.1 and 4.3 and 4.0, 4.2 and 4.3, respectively). Women white collar employees found intercourse more pleasurable than women workers while there was no significant occupational difference in the satisfaction with sexual life as a whole among the Finnish women. In 1971, the results for overall satisfaction with sexual life were identical with those in the 1990s.
Figure 16. Sexual Satisfaction According to Occupational Status in Finland in the 1990s
In Sweden, satisfaction with sexual life was lowest in the middle social group and in all groups slightly higher for women than for men (Figure 17). (Experiences of intercourse were not studied in Sweden.)
Figure 17. Sexual Satisfaction According to Social Group in Sweden
Figure 18. Sexual Satisfaction According to Occupational Status in Estonia
In St. Petersburg, occupational differences in sexual satisfaction were very small but clearer when we checked for type of employer. Russian men working in private business reported more pleasure from intercourse and were more satisfied with sexual life than those in state employment (Figure 19).
Figure 19. Sexual Satisfaction According to Type of Employer in St. Petersburg
Generally, finding intercourse pleasurable increased with education (Figure 20). Nevertheless, in Finland in 1971, men in the middle educational group were most pleased with intercourse. The overall satisfaction with sexual life of men did not show any clear correlation in Sweden but grew linearly with education in Estonia and St. Petersburg. The reported sexual satisfaction of St. Petersburg women decreased with higher levels of education (for a discussion of this surprising finding, see Haavio-Mannila and Rotkirch 1998).
Figure 20. Sexual Satisfaction According to Cultural Capital
Thus reported sexual satisfaction correlated with educational level in Estonia and St. Petersburg but not in Sweden and Finland. In all areas, social status increased sexual pleasure but not overall satisfaction for women.
Do rich men have more, or even most, sex? On the basis of this analysis, the answer is a clear ‘yes’. Also in contemporary Northern European societies, sexual success correlates well with male economic and social power. This was very clear in the case of reported parallel relations and overall sexual satisfaction and to a lesser extent evident with regards to reported numbers of sexual partners and reported pleasure in intercourse. In some cases, the middle or upper middle groups reported more sexual activity. Especially entrepreneurs proved to be quite entrepreneurial also in their sexual lives. The Estonian and especially the St. Petersburg material showed less socio-economic differences overall than the Swedish and Finnish data did. In the Russian case, this prompted us to look at the type of employer, which proved to be much more revealing than stated occupation and education.
In this article, we wanted to take a first step in opening the discussion between Nordic sex research and evolutionary theory. It is worth noting that the results are not as self-evident as one may think (‘of course rich people have more of everything’). Classical sociological theory does not predict that power and sexual activity go hand in hand. Quite the contrary – according to the ‘promiscuity thesis’ which already Edvard Westermarck (1912) criticised but which remains wide-spread, primitive people and less educated classes are characterised by unrestrained, wild sexual behaviour, while culture and civilisation teaches the upper classes sophistication and measure. Also psychoanalytical theory, which is often been incorporated into sociological views of the human being, predicts that talented and successful men would sublimate their desires, that is, work more and have less sex. True, feminist theory has been more sensitive to the idea that social and sexual power are somehow intimately linked, and has especially focussed on the abuses of socio-sexual power by male elites. With the aid of evolutionary theory, we can broaden the scope to include also consensual and mutually enjoyable sex and also better grasp how crucial female sexual choice has been for male power aspirations.
Obviously, the comparisons made here would all deserve deeper exploration and methodological considerations. Even if the evolutionary predictions made in the beginning were generally confirmed, we have not taken into account possible alternative explanations, concerning for instance class-specific sexual morality and occupationally related travel opportunities. Our results do not contradict evolutionary theory but they are not enough to prove it, either. Such a task would require more sophisticated empirical data. Darwinian theory actually distinguishes between lack of power and elite positions, not with gradual median differences. In ape communities the difference between alpha males and others is very great. The data most appropriate for our purposes should actually come from samples in which small elites (leading businessmen and -women, politicians, intellectuals etc.) are overrepresented and in which they can be expected to honestly tell about their sexual escapades. Now we can assume that we have too few representatives of real elites and the elite respondents may belittle their sexual activities.
The variables occupation and education were not always correlating with sexual success. This may be a problem of operationalisation, as mentioned in the beginning, but may also pose a substantial challenge to evolutionary predictions. In several cases, middle and lower middle status people scored quite high points of sexual activity. This may be due to social and life course factors – for instance, we know that the most educated people start their sexual life later than those with less education (Kontula 1991). It also points to the need of more ethnographically oriented research, where social status is really measured in its live setting.
What about the relations between gender equality and gender differences in sexual strategies? The differences in gendered sexual strategies were significant in all measured variables, as predicted by evolutionary theory. However, they were clearly smaller in the Nordic countries, while they were much bigger in the private than in the state sector in Russia. This supports the thesis that general social equality increases similarity of sexual behaviour among women and men (cf Kontula and Haavio-Mannila 1995).
Classic evolutionary theory predicts that highest status females would have less sexual partners and parallel relations than middle-range females, as there are simply very few men who meet their selection criteria, and also as female sexual activity is supposed to be less tied to social status (and more to e.g. outlook and personality). In some cases, our results supported this view. For instance, men’s general satisfaction with sexual life was more dependent on income and social group than women’s, and women’s income stopped correlating with their numbers of sexual partners for the richest. But contrary to this assumption, status often increased women’s sexual activity proportionally about as much as that of men. For instance, with the exception of Estonia, higher status women also reported having more parallel relations. The gender differences thus follow two distinct patterns: either the male curve is linearly growing while the female curve is ‘curbed’ at the upper end; or the male and the female curves are both linearly growing. The second pattern resembles gender differences in aggressive behaviour, where men and women have similar curves, only with consequently higher average rates for men than for women. Anne Campbell (1999) has explained this finding by stipulating general human dispositions for competitive and aggressive behaviour, which in women are inhibited and tempered by specific psychological mechanisms. To the extent that our results show similar gender differences for sexual behaviour, it suggests that gender in sexual emotions and strategies may also be more a question of degree than of kind.
Björnberg, Ulla (1998) Family orientation among men: A process of change in Sweden. In Drew, R. and E. Mahon (eds) Reconciling family and working life. London, Routledge.
Buss, David (1996) The evolution of desire. Strategies of human mating. New York, Basic Books.
Campbell, Anne (2002) A mind of her own. The evolutionary psychology of women. Oxford, Oxford University Press.
Cronin, Helen (1991) The ant and the peacock. Altruism and sexual selection from Darwin to Today. Cambridge, Cambridge University Press.
Cronk, Lee (1999) That complex whole. Culture and the evolution of human behavio. Westview Press, Colorado.
Darwin, Charles (1871) The descent of man: Selection in relation to sex. London, John Murray.
Freese, Jeremy (2000) What should sociology do about Darwin? Evaluating some potential contributions of sociobiology and evolutionary psychology to sociology. PhD, Dept of Sociology, University of Indiana. http://www.ssc.wisc.edu/~jfreese /freese_dissertation.pdf
Haavio-Mannila, Elina and Anna Rotkirch (1998) Generational and gender differences in sexual life in St. Petersburg and urban Finland. Yearbook of population research in Finland XXXIV 1997. Helsinki, The Population Research Institute, The Family Federation of Finland, 133-160.
Haavio-Mannila, Elina (2003) Gender gaps in multiple sexual partnerships. Poster at the annual meeting of The International Academy of Sex Research, Bloomington, Indiana, July 16-19.
Haavio-Mannila, Elina and Anna Rotkirch (2000) Gender polarisation and liberalisation: Comparing sexuality in St. Petersburg, Finland and Sweden. Idäntutkimus – Finnish Journal of Russian and Eastern European Studies, No 3-4, 4-26.
Hrdy, Sarah Blaffer (1999) Mother nature. Natural selection and the female of the species. Chatto & Windus, London.
Hurley, Susan L. (2002) “Feminism and evolutionary psychology: Can they be reconciled?” http://www.warwick.ac.uk/staff/S.L.Hurley/papers/fep.pdf
Kontula, Osmo (1991) Sukupuolielämän aloittamisen yhteiskunnallisista ehdoista (On the social conditions of starting sexual life). Sosiaali- ja Terveyshallitus. Tutkimuksia 14/1991. Helsinki, Valtion painatuskeskus.
Kontula, Osmo and Elina Haavio-Mannila (1995) Sexual pleasures. Enhancement of sex life in Finland, 1971-1992. Aldershot, Dartmouth.
Laland, Kevin and Gillian R. Brown (2002) Sense and nonsense. Evolutionary perspectives on human behaviour. Oxford, Oxford University Press
Lewin, Bo (ed.) (1998) Sex i Sverige. Om sexuallivet i Sverige 1996 (Sex in Sweden. About sexual life in Sweden 1996). Stockholm, Folkhälsoinstitutet 1998:11.
Miller, Geoffrey (2000) The mating mind. How sexual choice shaped the evolution of human nature. London, Random House.
Roos, J.P. and Anna Rotkirch (2003): Habitus, nature or nurture? Towards a paradigm of evolutionary sociology. Paper presented at the European Sociological Association Conference, Murcia, September 23-28, 2003.
 While human sexual selection in a limited sense concerns only heterosexual and possibly reproductive sex, our evolved sexual preferences, dispositions and sexual strategies are obviously involved also in other types of love and sexual relationships such as homosexuality. The evolutionary origins of homoerotic relations may be related to sexual selection in a larger sense, involving social and kin selection. (Miller 2000, 217-219.)
 The income classification refers to the 1990s. The Finnish marks used in the 1971 survey are roughly equivalent in their real value with
the Finnish marks in the 1990s.
 The percentages in the Figures are calculated from the married or cohabiting respondents. If single persons are included in the base, the results are about the same.