Money and Sex, The Illusory Universal Sex Difference

Comment on Kanazawa

Bibliographic Data

ID2277199
AuthorsThomas W Volscho (University of Connecticut, corresponding author)
Year2005
Volume46
Issue4
Pages719-736
Publication date2005-09-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueSociological Quarterly (JOURNAL)
Journal identifiersISSN: 0038-0253 • E-ISSN: 1533-8525
PublisherInforma UK Limited (PUBLISHER • GB)
DOI10.1111/j.1533-8525.2005.00033.x
OpenAlexW2149316138
LanguageEN
Citations received1
References cited14

Kanazawa tests the hypothesis, derived from Evolutionary Psychology, that men's income enhances their ability to engage in copulation with more partners and at a greater frequency. However, the results presented in Kanazawa's article fail to appropriately test for interaction effects and some of the analyses may suffer from sample-selection bias. I reestimate the equations appearing in Kanazawa's study and find (given the author's original methodological decisions) evidence in support of the evolutionary prediction in only two of the four original analyses. Had the same methodological decisions been consistently applied in the original study, then only one of the four analyses provides very weak evidence in favor of a sex difference in the returns to income. Furthermore, I conduct cross-national analyses with International Social Survey Program data from four other industrialized nations: Australia, Bulgaria, Ireland, and Poland. In only one of the nations (Ireland) and for only two of the four dependent variables is there any compelling evidence that men with higher incomes have more sex partners. Since the term “sex partner” may be ambiguous, I also use data from the National Health and Social Life Survey where detailed questions were asked that may better measure evolutionarily significant forms of copulation. The data are not consistent with the evolutionary psychological theory of sexual behavior. ACKNOWLEDGMENTSI thank Satoshi Kanazawa for sharing his SPSS computer files. I thank Angie Beeman, Andy Fullerton, Bandana Purkayastha, Ronald Taylor, Gaye Tuchman, and David Weakliem for comments. A special thanks to Jeremy Freese for providing helpful suggestions. Warmest thanks to Angie K. Beeman for helping me to think clearly and helping me in many other ways.NOTESNotes1 Kanazawa also examines the impact of men's income on their probability of never being married. The bulk of Kanazawa's article examines the relation between men's income and their sexual behavior; therefore this comment focuses on this particular outcome.2 One might speculate that CitationKanazawa (2003) decided to do this because CitationPerusse (1993) reported a stronger correlation between status and number of sex partners among currently unmarried men between the ages of 30 and 39. This begs the question of why the author did not restrict the analyses to this subsample in the figures reported in Table 1 (CitationKanazawa 2003:296).3 In a reply to another scholar, Kanazawa noted (about gay men having high numbers of partners), “This is why gay men have significantly more sex partners and have sex significantly more frequently than straight men, because there are no women in their relationships to say no (CitationLaumann et al. 1994:313–16, Table 8.4)” (CitationKanazawa 2001).4 The wording of the survey question according to the NHSLS codebook: “Now I will ask you some questions about vaginal intercourse. By vaginal intercourse, we mean when a man's penis is inside a woman's vagina.”5 Some analysts may have made alternative decisions about how to code the data. For instance, the count of lifetime number of sex partners might be better assumed to follow a Poisson distribution or one might add a small constant and then take the logarithmic transformation to achieve a normal distribution. With respect to the independent variable of interest, instead of using the arbitrary integers that the National Opinion Research Center attaches to people who fall within a given income range, one might instead use the midpoints of the income categories and use a Pareto-curve multiplier to estimate the mean of the upper income category. Nonetheless, in this comment I follow Kanazawa's protocols.6 The ISSP surveys are made available to the ICPSR by the Central Archive for Empirical Social Research at the University of Cologne, Germany (Zentralarchiv fuer Empirische Sozialforschung).7 These variables are “v18rels1” through “v18rels9” and the measure used in this article is obtained by summing these variables after deleting cases with missing values

Demographic economics · Economics · Positive economics · Sample (material) · Sociology · Test (biology) · World Values Survey · Cultural Differences and Values · Culture, Economy, and Development Studies · Demography · Evolutionary Psychology and Human Behavior · Psychology · Social Psychology

Tools

Open DOISci-Hub