Outline
I. Overview
A. Organisms as reproductive thoroughbreds
B. Theories and puzzling sex differences
II. The Integrated Model: An evolutionary, biosocial
Approach
A. Folk psychology vs. scientific psychology
B. Instinct blindness
C. The naturalistic fallacy
D. Generalizations
1. Overlapping distributions
2. Sex differences in means: effect sizes
3. Sex differences in variability
III. Closing comments
Sex Differences
The Evolution, Proximate Mechanisms, and Cultural Variations of Sexually Dimorphic Adaptations
Michael E. Mills, Ph.D.
Copyright © 2002, Michael E. Mills
Chapter 1: Introduction
Overview
Mysteries in life abound. How does the brain store information? Why are we conscious? Why is there so much suffering? Perhaps even more cosmological: Why is there something instead of nothing? And, given that there is something, why is it governed by a set of rules? And, given that there are a set of rules, why is it this particular set of rules, instead of an infinite number of alternative sets?
Sex differences also provide us with a variety of mysteries to contemplate. Why did sexual reproduction evolve? Why are there two sexes? And why do the sexes seem, at times, so oddly different, and, at times, incompatible? In this book, we will set for ourselves the task of attempting to scientifically untangle the puzzles of sex differences.
It should be acknowledged here at the outset that females and males are far more similar than they are different. In this book, however, we will focus our attention on sex differences. We will attempt to identify the most fundamental, enduring and "robust" sex differences -- those average, between group differences that tend to be manifest across different cultures and in different times. In particular, we will use an evolutionary, or an adaptationist, perspective to identify sexually dimorphic (different) psychological adaptations. Such sexually dimorphic adaptations evolved due to different evolutionary pressures operating on males and females -- a process that Charles Darwin called sexual selection. Here we will explore the different "reproductive strategies" employed by males and females (including those of nonhuman species), and the resulting conflicts of interest between the sexes. Since most sexually dimorphic adaptations are intimately connected with reproduction, a particular focus of this book will be on sex differences in sexuality, courtship, jealousy, mating systems, and parenting.
Part I of this book (Chapters 1 - 4) will provide a review of basic evolutionary concepts as well as identify the set of fundamental evolved male-female sex differences. In Chapter 2 we will review the scientific literature regarding how males and females differ, including morphological, hormonal and brain differences. The rest of the book will focus on why males and females evolved different physical and behavioral adaptations.
In Part II (Chapter 5 - 7) we will review the dimorphic reproductive strategies and mating systems of non-human species. By identifying the various strategies that non-human male and female animals have evolved in response to particular ecological and social conditions we will see that there are often many ways to solve survival and reproductive problems.
Part III (Chapters 8 - 12) focuses on humans and the evolutionary psychology of sex differences. The history and current status of the nature-nurture controversy in the social sciences is reviewed in Chapter 8. Chapters 9 and 10 explore on human male and female reproductive strategies, respectively. How these sexually dimorphic adaptations are expressed differently in various cultures, and in specific circumstances, will be explored. Chapter 11 examines the resulting "battle of the sexes" in courtship, mating and parenting. The final chapter reviews how societies have sought to structure the relationships between the sexes. The influence of legislation, religion, and other social institutions will be examined. We will also entertain speculations regarding the future course of the "battle of the sexes."
Organisms as reproductive thoroughbreds
All the world's a stage, and all the men and women merely players.
-- Shakespeare
Shakespeare’s famous metaphor, if viewed from the perspective of evolutionary biology, can assume a surprisingly relevant interpretation. Imagine the environment as the stage; that the players are females and males, and that the climax of the plot is successful reproduction. The various plot twists involve themes of romantic and sexual attraction, courtship, mating and parenting. It is this "reproductive drama" between the sexes that we are about to explore. It is a drama that is reenacted each generation, by all sexually reproducing organisms, on various ecological stages. Although the choreography, tempo, costumes, roles, and scripts of organisms may vary across time, place, and species, the fundamental, underlying objective remains an evolutionary constant: successful reproduction. We will learn that evolution is blind to virtually everything except reproductive success. Genes that predispose traits that lead to successful reproduction survive through each generational sieve; less fecund organisms send their genes into oblivion.
If you have seen horse races you may know that the racing horses are called "thoroughbreds." Why? Because they have been "thoroughly" bred over many generations, artificially by humans, to do one thing well: run fast. Those who win races consistently command very high stud or birthing fees, and for good reason: any genes that may have predisposed their superior running speed are likely to be passed on to their progeny. We, too, are thoroughbreds, but in a different sense. Think how many of your direct ancestors did not become parents. Trace back through your family tree -- your parents, your grandparents, your great-grandparents, as far back as you can. How many can you think of?
But this is a trick question, isn't it? Although we may know virtually nothing about our distant ancestors, there is one thing we know for certain: they all became parents! You have billions of direct ancestors. None of them failed to survive to reproductive age and to reproduce. This represents a three or four billion year breeding program to design an organism effective at reproduction (and surviving at least long enough to do so). We are reproductive thoroughbreds, as are all organisms.
Evolutionary genealogies can be likened to forever branching lines of dominos. It takes only one ancestor who failed to reproduce to terminate a genealogical branch--rather like one misplaced domino that fails to contact the one after it. There are billions of such terminated branches in the tree of evolution. But the genealogical lines that produced you are unbroken lines, and they trace back in time and eventually converge to one point -- to the very beginning of life on this planet three to four billion years ago.
If we can anticipate that thoroughbred horses will be fast runners due to their selective breeding, it should come as no surprising that most organisms eventually mate, bear offspring, and (among many species) invest in parenting. For humans, the economic and psychological costs of raising children are not trivial. According to the Insurance Institute of America, today it will cost new parents over a quarter of a million (
Given these fiscal and psychological stresses, why do people continue to bear children? Probably due to the same reason thoroughbred horses run fast -- we have been bred to do so. We are literally doing "what comes naturally." Given our long breeding history, our bodies and brains have been designed help ensure our survival to reproductive age, and our successful reproduction. If there is a "goal" of life, at least in an evolutionary sense, it is reproductive success.
Theories and puzzling sex differences
Scientific theories are evaluated by their consistency with real-world data. Different theories that attempt to explain similar phenomena are in competition -- those that enjoy greater corroboration by empirical data are favored by scientists. When a new theory is developed that more concisely and thoroughly explains the phenomena under study, older theories may gradually (and in some cases, rapidly) fall out of favor in the scientific community, in what is called a "paradigm shift" (Lakatos, 1970). Often what initiates such paradigm shifts are particularly puzzling, or anomalous, phenomena that older theories have difficulty explaining. In fact, it is puzzling phenomena often provide the "acid tests" for theories.
For example, an early astronomer named Ptolemy developed a complex theory designed to explain an anomalous astronomical phenomena called retrograde motion. When viewed from earth, retrograde motion occurs when planets are observed to stop, and then back up briefly, in their successive nightly progression across the sky -- quite an unexpected observation! At the heart of Ptolemy's theory was an "earth-centered" perspective of the solar system -- the assumption that the planets and sun all revolved around the earth. Retrograde motion did not fit well with that view, yet Ptolemy managed to develop a very clever, albeit convoluted, theory to explain it.
Later, the astronomer Copernicus developed a competing theory: a "sun-centered" model of the solar system based on the assumption that all of the planets, including the Earth, revolve around the sun. Because planets further from the sun take longer to orbit it, retrograde motion, as seen from the Earth, was parsimoniously explained as our planet overtaking the orbit of plants further from the sun. Copernican theory more simply and elegantly explained this anomaly of retrograde motion. The result was a significant paradigm shift in astronomy: from an earth-centered to a sun-centered theory of the solar system. As is often the case in science, a puzzling phenomenon provided the impetus for the shift.
Sex differences also provide us with many puzzles. Such sex differences are the subject of private conversation among friends, as well as the topic of public discussion, jokes, and debates. The novelist Carolyn See (1987) commented on the explosion of popular books with provocative titles such as: "Men Who Can't Love"... "Women Men Love/Women Men Leave: Why Men Are Drawn to Women/What Makes Them Want to Stay"... "Women Who Love Too Much: When You Keep Wishing and Hoping He'll Change." She noted that:
...dozens of other, similar books are... about belief systems. ...(I am not suggesting) that the authors of these (popular books on sex differences) intentionally prey on women's insecurities, but rather that they are addressing themselves to a series of large and intractable clashes between facts and belief systems that cannot, at this chronological moment, be put in coherent order. Ptolemy, when he invented the Ptolemaic system of astronomy, was not trying to get rich quick, or pull the wool over anyone's eyes. He was trying, as best he could, to explain the phenomena that he perceived. The writers of these books are doing the same, but it just isn't good enough. They are missing something, even if it would take a Copernicus to know what it is.
Today there are two very different, competing paradigms that attempt to explain sex differences: a "culture-centered" view that is called the "Standard Social Science Model" (SSSM). The second view is a "nature-nurture centered" view called the "Integrated Model" (Tooby & Cosmides, 1992). The SSSM paradigm has been dominant in the social sciences for decades. It suggests that sex differences are "socially constructed" by culture, that sex roles are arbitrary, and there are few important biological differences between the sexes. The SSSM proposes that humans are essentially born a tabula rasa (a “blank slate”) -- it is assumed that there is no common "human nature" to be found across different cultures. It assumes that there are no underlying commonalities between separated cultures -- each could be different on any trait in any direction. Proponents of this approach suggest that biology has no, or at most a very insignificant, influence in the determination of sex differences. From the perspective of the SSSM, hypotheses suggesting that there may be a biologically based "male nature" and "female nature" are seen as extraordinary, and require extraordinarily convincing evidence before they need be taken seriously (Symons & Ellis, 1989).
The second, more recent perspective, the Integrated Model, suggests that sexually dimorphic psychological adaptations have evolved, and that these interact with socialization to produce sex differences. The key argument of this approach is that, like morphology (physical characteristics), psychological structures too, are subject to evolutionary selection pressures. Thus emotional, cognitive and behavioral predispositions evolve over many generations and result in a bundle of functional neurological predispositions, termed "mental mechanisms." Taken as a whole, this set of predispositions may be termed "human nature" -- the universal set of psychological adaptations that characterize the human species. Every species has evolved a unique morphology (physical characteristics); so too is each species different in terms of its unique set of behavioral predispositions -- its "nature." The Integrated Theory suggests that cross-culturally consistent sex differences have resulted from different evolutionary pressures operating on males and females. That is, like "human nature," there also exist a slightly different "male nature" and "female nature."
It is important to keep in mind that psychological adaptations are not the same as instincts. As will be discussed in more depth in Chapter 3, expressed behaviors are plastic -- subject to modification by social conditioning, (b) affected by developmental events during ontogeny (the life history of an individual), and often (c) facultative (conditional -- dependent on specific situations or contexts). Thus, evolved behavioral predispositions that are due to evolved psychological mechanisms may be expressed differently depending on the social and environmental context. An adaptive behavioral predisposition may be magnified, minimized, repressed and/or or displaced in various contexts and in different cultures.
Since the basic reproductive strategies of females and males may be expressed somewhat differently in different cultures, it can be easy to overlook the underlying commonalities. For example, the human biological predisposition to learn to talk is manifested differently in different cultures as different languages, dialects or and accents. Despite this cultural variation, it is clear that human language is an expression of a common evolved psychological adaptation (which may is also termed a mental mechanism) (Pinker, 1997). As we shall see, the same can be said about some sex differences. Despite the apparent cultural and temporal variability, certain sexually dimorphic adaptations are human universals.
As with the earth- versus sun-centered theories of the solar system, one way to evaluate the relative merits of the SSSM and the Integrated Model is to see how well each can explain sex difference anomalies. Can you think of some puzzling differences between women and men -- especially for "robust" male-female differences (those that may have remained fairly consistent across time and cultures)? What puzzles can you think of? For example, have you ever wondered about the following questions:
... Why are men taller than women?
... Why do women live longer than men?
... Why do women, but not men, have menopause (the loss of the ability to reproduce)?
... Why have virtually all soldiers throughout history been male?
... Why are most wives younger than their husbands?
... Why do most women prefer men who are taller than they are?
These sex-difference puzzles are intriguing because the reasons for them are not immediately obvious. Sex difference puzzles such as these, and those listed in
================================================================
START BOX START BOX START BOX START BOX
A few sex and gender puzzles
Some Biological Puzzles:
- Why is there sex? Surprisingly, sex is not required to reproduce. Both single and multiple celled organisms (even larger insects, such as aphids, can reproduce asexually by cloning (ref).
- Why then do many organisms reproduce sexually? As we shall see, the costs of sexual reproduction seem high when compared to the relatively more simple method of cloning.
- Why are there only two sexes? Why not three? Or four? Surprisingly, sexual reproduction does not require that there exist two different sexes. Some bacteria can reproduce sexually -- yet there are no males or females (ref).
- Around the world, women are generally physically shorter than men. They also tend to live longer than men -- about seven years in the
- Why do men have nipples? They serve no function -- men cannot lactate for their offspring.
- Why do females have the capacity to orgasm? The muscular contractions of orgasm serve a very important reproductive function for males: ejaculating semen. Unlike males, females need not orgasm in order to reproduce.
Some Psychological and Sociological Puzzles:
- The probability that a husband will commit adultery is highly positively correlated with his income -- this is true around the world.
- The first sexual and relationship initiations are almost always made by men. These are the initiatives that are unambiguous (cannot be interpreted as anything other than a relationship or sexual initiative) and are “risky” (in the sense that one is vulnerable to a clear rejection), such as asking for a date, holding hands or kissing for the first time, or initiating sexual contact for the first time.
- Single men in their teens and twenties are many times more likely than any other demographic group to engage in physically violent altercations that lead to severe injury or death, often over apparently very trivial conflicts. Around the world, men populate prisons 9 to 1 compared to women.
- The vast majority (upward of 98%) of romance novel readers are women. About forty percent of all paperback novels sold in the
- The majority of photographic models are females in their late teens or early twenties. For example, if you scan the covers or pages of most any magazine, regardless of whether it caters to men or women, there are generally far more photographs of women than of men. Further, many of these women are not famous -- we have no idea of their identity. Men, when they are presented, typically are famous, wealthy, powerful, or a sports hero -- someone that we are likely to be able to identify by name.
- Virtually all combat soldiers are young men. This has been the case in all cultures throughout all recorded history. (As noted by (ref), the tribe of South American Amazon women warriors is a mythical story, not anthropological fact.)
- In monogamous and polygynous cultures, wives are virtually invariably treated by their husbands as sexual "property" and jealously guarded from potential and actual advances of other men. Cross-culturally, penalties for adultery are typically far more severe for wives than for husbands.
- Consumers of explicit pornography, and the services of prostitutes (whether the prostitute is female or male), are virtually always men.
- Virtually all rapes are committed by men. In fact, the vast majority of all sexual crimes are committed by males.
- Many male homosexuals living near gay communities have had hundreds or even thousands of different sexual partners. In contrast, lesbians typically have had no more sexual partners in their lifetime than heterosexual females.
- In virtually all societies, there is a tendency for females to marry upward in social class ("hypergamy"), while males tend to marry downward in social class. There are two demographic groups that often complain about the difficulties of finding a suitable marriage partner: older, financially secure women, and younger, poor men.
- The probability that a girl will be physically or sexually abused in the home increases substantially if she is living with a step-father instead of her biological father.
- Cross-culturally, women say that they want to marry men who are slightly older than themselves (and they generally do); men say that they want to marry women who are younger than themselves (and they generally do)
- Cross-culturally, women prefer men who are slightly taller than themselves.
- During courtship, it is generally expected that the male will give more "token" as well as truly valuable gifts to his girlfriend, rather than vice versa.
- In wealthy countries, such as the
- Women consistently prefer to marry men who are older than they are (some times even twice their age or more). And very few men prefer to marry women much older than themselves. Why?
- Reactions in the workplace to a sexual invitation by a co-worker differ markedly by sex. Sixty-three percent of women responding to a survey said that they would be insulted, while 17% said they would feel flattered. When men were surveyed with the same question, about the same percentages were found, but in reverse. Sixty-seven percent said they would feel flattered, while 15% of men said they would feel insulted.
- Extremely few women ever steal another woman's infant to raise it as their own. But virtually no men do
BOX END BOX END BOX END BOX END
====================================================
The Integrated Model: An Evolutionary, Biosocial Approach
Despite an avalanche of popular books about sex differences, virtually none have provided satisfactory (and certainly not scientifically sophisticated) explanations. The SSSM theories have not fared much better -- especially in explaining the puzzles listed in
This recent paradigm shift surmounts interdisciplinary boundaries. This approach is broadly termed, as noted earlier, as the Integrated Model. Titles for specific subfields include "evolutionary psychology" and "evolutionary anthropology." As you will learn, this relatively new perspective offers some surprisingly parsimonious solutions for each of the sex difference puzzles noted above and in
Folk Psychology vs. Scientific Psychology
No one would expect you to be able to understand physics without years of study and a mastery of advanced mathematics. Similarly, you could not be expected to develop a sophisticated understanding of chemistry or biology without years of dedicated study, and a mastery of a technical terminology. Yet many people seem to think that an understanding sex differences can come from personal observations in everyday life. Especially egregious examples are seen daily on many daytime talk shows and in some popular books. It is tempting to think that, via casual observation, reflection, and conversation with friends, we can develop accurate theories that will explain the behaviors of others, ourselves, and the opposite sex. Indeed, we all have "intuitive" or "informal" psychological theories in many areas of life -- some of which serve us well. Is there anything wrong with such personal, intuitive theories? After all, we have spent most of our lives interacting with others, and observing the behavior of the two sexes. Why shouldn't our personal, informal theories be accurate?
No doubt, we do have much intuitive knowledge about people -- this is termed "folk psychology." And some of it is quite accurate. For example, many commonly believed stereotypes about men and women are surprisingly consistent with findings from empirical research on sex differences (Eagley, 1995). Research suggests that people very often correctly estimate (and sometimes under estimate) the magnitude of sex differences (Swim, 1994), as we shall see shortly.
Although most people generally accurately estimate how females and males differ, folk psychology fails to provide an accurate understanding of why the sexes differ. For example, many people may inaccurately believe that (a) men are taller than women because young boys were encouraged to exercise more than girls, (b) young girls generally engage in less "rough and tumble" play than boys because only boys are praised for doing so, (c) males generally die younger than females primarily because they experience greater job stress, or (d) legalized pornography and prostitution increases the incidence of rape. All of these intuitive theories about sex differences are wrong.
Our "intuitive theories" of physics ("folk physics") often fail us, too. Although we all have had many experiences with falling objects, many people incorrectly believe that a ball twirled in a circle by an attached string will, if the string breaks, continue to fly in a curved trajectory (McClosky, 1983). This betrays a lack of understanding of the underlying causality. Similarly, our years of interaction with other people do not necessarily provide us with accurate theories of human behavior. For example, is "out of sight, out mind" correct, or does "absence make the heart grow fonder"? Do "birds of a feather flock together" or do "opposites attract"?
According to Stanovich (1989),
...our intuitive theories are often a rag bag of general principles, homilies, and cliches... The enormous appeal of cliches like these is that, taken together as implicit 'explanations' of behavior, they cannot be refuted. No matter what happens, one of these 'explanations' will be cited to cover it. No wonder we all think we are such excellent judges of human behavior. ...psychology's threat to folk wisdom disturbs some people... Science seeks conceptual change. Scientists try to describe the world as it really is, as opposed to what our prior beliefs dictate it should be like. The dangerous trend in modern thought is the idea that people must be shielded from the nature of the world... (p. 31)
Another factor that contributes to our preference for "folk psychology" over potentially more accurate scientific theories and evidence is that folk psychology speaks our language -- every day language. We don't have to invest the effort to learn precisely defined technical terms that are used in scientific analyses and theories. But such well defined terminology (or "jargon") often help to communicate new concepts.
Instinct Blindness
If intuitive physics and folk psychology often fail us, we should not expect that the reasons for sex differences will be intuitively obvious, either. Evolutionary psychologists suggest that our brains and bodies were designed to help us each to survive to reproductive age, to reproduce, and to help ensure that our children reproduce. However, we did not evolve a "meta-consciousness" -- a conscious awareness of the functions of our own mental design. Our ancestors needed not consciously want to have children. Rather, they evolved the physical sensations, emotions, and cognitive predispositions that historically contributed to their reproductive success. We have inherited those designs. Richard Alexander (1987) noted that
...even though people are usually following their (survival and reproductive) interests, they do not themselves know precisely... what those interests are. ...such information is not a part of their conscious knowledge, and if you asked people what the Ythink their interests are they would usually give wrong answers... People are not generally aware of what their lifetimes have been evolved to accomplish, and, even if they are roughly aware of this, they do not easily accept that their everyday activities are in any sense means to that end. (p. 36)
This "instinct blindness" was noted more than a century ago, in 1890, by one of the first psychologists, William James. He suggested that we take our "instincts" for granted -- we feel they need no explanation, they just "are." Studying what is natural, then, is likely to seem strange. Why do people smile when they are happy? Why do we experience jealousy? Why do we fall in love? James suggested that such "instincts" require extra effort to understand. Natural feelings work so well that we don't often appreciate the complexity of their underlying mental machinery, or the evolutionary reasons why such machinery exists in the first place. As we will learn, an evolutionary approach can provide a powerful "lens" through which we can look to help correct our instinct blindness (Tooby & Cosmides, 1997).
The Naturalistic Fallacy
The "naturalistic fallacy" is the presumption that what is natural is morally good. For example, the fear of many people who believe in sexual equality is that if it is discovered that men and women are by nature different, some people might argue that these differences are "good" (because they are "natural"), and that this "naturalness", in turn, would serve as a justification for sexual discrimination.
Virtually every evolutionary psychologist disagrees strongly with this. What is natural is not necessarily good -- and this becomes increasingly obvious as one learns more about the cruelty that is inherent in much of nature. In his essay "Nature," the philosopher John Stuart Mill (1874) noted that "what is" does not imply "what ought." He wrote that, far from being benevolent, nature impales, devours, burns, crushes, starves, freezes, poisons and "has hundreds of other hideous deaths in reserve." Darwin himself remarked: "What a book a devil's chaplain might write on the clumsy, wasteful, blundering, low and horribly cruel works of nature!" (quote in Pinker, 1997, p. 50).
In this book, we will at times investigate some of the more cruel aspects of human nature, including jealousy, aggression, sexual harassment, rape, and infidelity. Why study such unpleasant phenomena? By analogy, if one wants to discover a cure for cancer or pneumonia, one must be willing to expose oneself to some gruesome autopsies and embark on long study of unpleasant diseases. None of our explorations of negative human traits is meant to offer any moral justification for these undesirable behaviors. In this book we will not focus a great deal on issues of morality, equality, politics, or "what ought" to be. Rather we will focus on the "what is" of robust gender differences, and why these differences evolved. The rationale for this is that a realistic and sophisticated understanding of "what is" is a prerequisite to make informed judgments about "what ought" to be, and how to get there (Dennett, 1995).
In his book The Evolution of Human Sexuality, Symons (1979) noted that the evolutionary perspective might help us not only understand the perspectives of the opposite sex, but also to help to refute the idea that one sex can been seen as a "defective" version of the other.
...an evolutionary approach to (studying gender differences) may counteract the frequently stated or implied notion that a difference between human males and females inevitably indicates that one sex is inferior or defective. For example, since it is commonly assumed to be "natural" for humans to be sexually aroused by viewing an attractive, unclothed member of the opposite sex, the evidence that women are less likely to be sexually aroused by the sight of naked men than men are by the sight of naked women often is attributed to systematic early repression of female sexuality, and women thus are seen as inhibited or defective. On the other hand, Shulasmith Firestone (1970) argues that, unlike women, men are not able to love. She implies that although the sexes are identical at birth, males are crippled emotionally by early Oedipal experiences that females escape; hence, insofar as they differ from women in their romantic feelings, men are seen as defective. But the view ...that selection has produced marked sex differences in sexuality -- implies that neither sex can be usefully considered to be merely a defective version of the other. (p. 4)
Measuring Sex Differences
Few things are harder to put up with than the annoyance of a good example.
-- Mark Twain
Every statement in this book about females or males is a generalization. Misused or misunderstood, generalizations can lead to inappropriate, inaccurate, and even dangerous conclusions. It is very easy to misinterpret generalizations because the human mind is not predisposed to think in terms of statistical averages and overlapping statistical distributions. Instead, we tend to think categorically -- to sort items into discrete and mutually exclusive slots (Pinker, 1997). For example, consider the following generalization: "Men are taller than women." Most people understand that this generalization does not imply that all men are taller than all women. But it is not always easy to keep this in mind. For example, imagine that you are a man who is 5'2" tall, or a woman who is 6'1". The generalization that "men are taller than women" does not apply to you as an individual. Unless you kept the concept of what a generalization is clearly in mind, if you were to hear the statement repeatedly that "men are taller than women" it might begin to wear a little thin. You might be inclined to wonder: "What about me? Am I abnormal?" Well, in one sense -- statistical abnormality -- you are! But as we will see in Chapter 2, statistical abnormality is not necessarily associated with any other kind of abnormality (e.g., social or functional abnormality). Also, generalizations do not indicate what is morally "correct." There is nothing wrong with being a short man or a tall woman. Similarly, there is nothing wrong if any other sex difference discussed in this book does not seem to apply specifically to you. Another problem with generalizations is that, since it is so easy to forget that they are statistical averages, some people erroneously think that the generalization can be falsified if one counter-example can be found. For example, presented with the generalization that "men are taller than women" one may be inclined to say: "Wait a minute. I know of several women who are over 6' tall!"
If generalizations are easily misunderstood, why use them -- especially when discussing a politically charged subject like sex differences? From a scientific perspective, when average group differences between the sexes are found to exist it is often enlightening to try to find out why they exist. While some may argue that a little knowledge (especially in the form of a generalization) is worse than none, the search for deeper knowledge must begin somewhere. Second, generalizations can useful when we know nothing about a specific individual. An accurate generalization, as a statistical probability, often can offer a best guess estimate when more information is not available. Lastly, sometimes people confuse the objectives of nomothetic and idiographic theories. Nomothetic theories attempt to explain behavior using general principles (generalizations) that apply to groups as statistical averages. Idiographic theories attempt to explain the behavior of a particular individual. "Why do people like sugar" is a nomothetic question; "Why does Sam not like cotton candy" is an idiographic question. Keep in mind that we are presenting nomothetic, not idiographic, theories in this book. Different researchers in the same discipline may choose to focus their work on either nomothetic or idiographic questions. For examples, early physiologists researched the commonalities between human bodies (everyone was found to a liver, a heart, and lungs); other physiologists became interested in the differences between individuals (Why is it that some individuals have type A blood, others type B? Why are the shapes of lungs different in different people?). The nomothetic physiologist investigates "human morphological nature;" the idiographic physiologist studies the genetic and environmental processes by which human bodies come to differ. Both nomothetic and idiographic approaches are useful, but to simplify matters, in this book we will attempt to explain sex differences using nomothetic methods, using a "men in general" and "women in general" perspective. We will not explore in much detail the behavioral variability within one sex, although that is certainly a valid area of study. Keep in mind that since nomothetic theories use general principles to explain statistical averages, finding a few exceptions to a general rule is not unexpected, and is not fatal to a nomothetic theory. This is captured by the phrase “an exception that proves the rule.” In addition, it is often easier to predict and explain the average behavior of a group than it is to accurately predict and explain the behavior a particular person.
Overlapping Distributions
People differ on many traits; the distribution of these differences is generally well described by a "normal" or "bell-shaped" curve. Height is normally distributed, so is intelligence. When the two sexes differ those differences can be visualized as two overlapping normal distributions. For example, if we examine the distributions of the height of adult Americans, the female mean is 64.3 inches and the male mean is 69.7 inches (McGraw & Wong, 1992). The standard deviations for these two distributions are 2.6 and 2.8, respectively.
Adult Male and Female Height Distributions
(In inches)
Note the extreme differences in the relative proportions of males and females near the tails of the distributions. At the middle of the distributions, the proportions of males and females are similar. For example, in the table below, of the people that are 67 inches tall (5'7") about 46% are women and 54% are men.
TABLE 1.1
Percentage of People at Various Heights by Sex
Height (in inches) | Percent
| |
| FEMALES | MALES |
61 | 99 | 1 |
62 | 98 | 2 |
64 | 95 | 5 |
65 | 88 | 12 |
67 | 46 | 54 |
68 | 23 | 77 |
69 | 8 | 92 |
71 | 2 | 98 |
72 | 1 | 99 |
However, near the extremes of the distributions, we find extreme differences in the proportion of each sex. Of the people who are 6 feet tall, 99 percent are men; of the people who are 5'1" tall, 9 percent are women.
Web Resource: See http://bellarmine.lmu.edu/faculty/mmills_fp/software.htm You can enter the data for any group difference (that is normally distributed) to see the overlapping distributions, the proportion of each group at each level, and the calculated effect size).
Sex Differences in Means: Effect Sizes
The difference, or offset, between two overlapping distributions is called an effect size. The term can refer to the effect of some intervention; for example, the average difference in rated headache pain between an experimental group of people who got aspirin compared to a control group that received a placebo. The "effect" can also refer simply to the magnitude of the (average) difference between two pre-existing groups.
In the table above, the "effect" refers to the average effect of sex on height. An effect size (or d) is mathematically expressed as the distance, in standard deviation units, between the mean of the first distribution (females in the example above) and the mean of the second distribution (males in the example above). (If you are unfamilar the term "standard deviation" -- see the appendix. In brief, it is about the same as the "average deviation" from the mean for a set of scores -- the average distance from the mean of the distribution.) The effect size for sex and height is about 2.0. That is, the mean height for females is 2.0 standard deviations distant from the male average.
For those who do not readily understand the meaning of standard deviations, the Common Language Effect Size was developed as an alternative index by McGraw and Wong (1992). In our example, this is the probability that a randomly selected male will be taller than a randomly selected female -- in this case, it is about .92. That is, in about 92 out of 100 blind dates, the male will be taller than the female. Or, another way to say this, about 92 percent of men are taller than the average female height of 5'4". However, because it is commonly used among researchers, here we will use the effect size (d) statistic to express the magnitude of various sex differences. Using this index, the magnitude of effect sizes can be somewhat arbitrarily classified as follows: 0.20 is a small effect size, 0.50 is a moderate effect size, and 0.80 is a large effect -- one that would be clearly noticeable (Cohen, 1977). The height difference between men and women is obvious, and its effect size of 2.0 is large indeed. In fact, it is substantially larger than the effects of most interventions in psychology, which, on average, have effect sizes of about d = 0.50. That is, treatment groups in experimental (or clinical) psychology change, on average, by about one half of a standard deviation (d = .5) compared to the control group (Lipsey & Wilson 1993). Although this would be classified as a moderate effect size, such an effect size can reflect group differences of both theoretical as well as practical import.
With respect to sex differences, even relatively small effect sizes can result in extreme differences in the proportion of males and females at the tails of a distribution, and, as noted above, this can often result in significant real-world effects. For example, Martell, Lane and Emrich (1996) performed a computer simulation of competitive promotions in a corporate setting with eight management levels. There were progressively fewer positions at top levels, and previous advancements were prerequisite to future promotions. Starting with an equal proportion of males and females at the bottom level, the computer simulation found that even a very small sex difference in average performance led to a highly unequal sex ratio at the top levels of the organization. For example, a sex difference of only 1% of variance between the sexes resulted in a 65 : 35% sex ratio at the top management level.
Swim (1994) compared the effect sizes (derived from several empirical research studies) of various sex differences with people's estimates of the magnitude of these sex differences, as shown in Table 1.2
Table 1.2
Some sex difference effect sizes and people's estimates of them.
(Adapted from Swim, 1994, Table 3)
Characteristic | Effect Size | People's Estimate of Effect Size | General Accuracy |
Restless | 0.72 | 0.20 | Underestimate |
Emerge as a leader | 0.49 | 0.29 | Underestimate |
Math tests | 0.41 | 0.28 | Underestimate |
Help in a group | 0.42 | 0.14 | Underestimate |
Aggression | 0.29 | 0.51 | Overestimate |
Influenced by Group Pressure | -0.32 | -0.06 | Underestimate |
Involved in conversations | -0.32 | -0.30 | Accurate |
Decode nonverbal cues | -0.43 | -0.39 | Accurate |
Gazing during conversations | -0.68 | -0.16 | Underestimate |
NOTE: A positive effect size indicates a higher average score for men. The assessments of accuracy are those of Swim (1994).
In Table 1.2, a positive effect size indicates a higher average score for men; a negative effect size indicates a higher average for women. Contrary to the assumptions of many people (including many social scientists) that sex stereotypes are exaggerations of imagined or small sex differences, Swim found that people are more likely to underestimate the magnitude of sex differences than to overestimate them. Despite this, Eagly (1995) notes that many social scientists "...have worked energetically to preserve the 1970s scientific consensus that sex-related differences are null or small. ...Indeed, this view has become quite widely accepted... and strongly influences contemporary textbook presentations of gender research" (p. 150). In her review of the empirical research literature, Eagly concluded that sex differences are not small, are stable across studies, and tend to corroborate many popularly held sex-role stereotypes. Most sex differences are smaller than the male-female height difference, however, some of the "robust" differences predicted by evolutionary theory (discussed later in this book) are very large indeed, a few are much larger than the effect size of the male-female height difference.
Sex Differences in Variability
The sexes can differ, not only on group mean scores, but in the variability of each distribution. It has been suggested that, in general, there is greater variability among men than women for a variety traits because women have two X chromsomes (XX) while men only have one (XY). In women, the second X chromosome may moderate, or have an averaging effect, on the influence of the first X chromosome (Lehrke, 1978, 1997; Turner, 1996). A few weeks after conception, one of the two X chromosomomes in each cell in a girl's body is randomly deactivated and shunted aside (it is then called a Barr body). All subsequent decendants of that cell have the same X chromosome activated. Thus, each cell in a woman's body has only one active X chromosome, but whether it is the one inherited her mother or her father varies between different clumps of cells. An X-linked gene inherited from the mother that coded for an extreme trait (either high or low) may be moderated an alternate gene inherited from her father that has a less extreme effect. However, a male who inherited an X-linked gene that coded for an extreme trait would have the same gene in all of his cells (because he doesn't have a second X chromosome) and thus there would be no moderating effect. For example, ocular albinism (a genetic disorder of the retina) is a X-linked disorder. A woman who inherited this gene from her mother, but not her father, would have a retina with a combination, or a "mosaic," of clumps of normal cells (derived from the normal gene from her father) intermixed with clumps of non-functioning albino cells (derived from the defective gene from her mother) (Lehrke, 1997). Although her vision would be deficient, she would not be blind. However, because a male with this gene would have all his retinal cells derived from the defective X-linked gene from this mother, all of his renital cells would be albino, and he would be blind.
If some genes that code for intelligence are X-linked, males would inherit their intellectual potential only from their mothers -- and any genes for extremely low or high intellectual potential would not be moderated (Lehrke, 1978, 1997; Turner, 1996). Females would inherit their intellectual potential from both parents, and the effects of each would be averaged. As can be seen in Diagram 1.1, the distributions of male and female IQ scores indicate greater intellectual variability for males. Compared to females, there are more males who are intellectually deficient, and more who are intellectually gifted.
Male and Female Distributions for Sex-Linked Genetic Traits
If, in general, there is greater variability among males, we would expect to see more males than females at both the low and high extremes of a distribution for many different traits (at least for those traits that are X-linked). This is often the case (Feingold, 1995; Hedges and Nowell, 1995), and it can lead to unequal representation of males and females in certain occupations (Hedges & Nowell, 1995). It may also help explain some sex differences we see in school, as noted in Table 1.3.
Table 1.3. Sex ratios (male/female) for
various problems (data from Lehrke, 1997; Fombonne, 1999)
Behavior problems | 9.4 : 1 |
8.0 : 1 | |
Delinquency | 4.5 : 1 |
Autism | 3.8 : 1 |
Personality disorders | 2.6 : 1 |
School failure | 2.6 : 1 |
Mental disease | 2.3 : 1 |
Special class placement | 1.7 : 1 |
Mental subnormality | 1.5 : 1 |
The unpaired, un-moderated X of males may also help explain why males suffer proportionally more often from certain disorders, such as hemophilia and color blindness. Only females that have the predisposing gene on each of the X chromosomes inherited from their mother and their father will have the disorder. All males that have the unpaired X-linked gene from their mother will have these disorders.
The Interaction of Evolved Adaptations with the Environment and Culture
We often hear, especially with respect to sex differences, “Is it due to biology or culture?” The question is based on a misguided premise, one that suggests that the causal influences of biology and culture can be separated – they cannot. Behavior is always caused by an interaction of biological and cultural factors.
Perhaps a better way to phrase the question might be: is the trait a sexually monomorphic adaptation (males and females both have the same adaptation) or is the trait a sexually dimorphic adaptation (males and females have evolved somewhat different adaptations with respect to the trait). In addition, we need to know if the sex role socialization in the culture is sexually monomorphic or dimorphic, that is, are males and females differentially rewarded for the same behavior?
When can we expect to find an average sex difference in a particular trait? To answer this question, we need to know two things: (1) Is the adaptation sexually monomorphic or dimorphic, and, (2) is the sex role socialization related to the adaptation monomorphic or dimorphic? If the socialization is dimorphic, we also need to know if the differential socialization will generally tend to either increase or decrease the magnitude of a sex difference. (You may wonder how differential socialization could decrease a sex difference. This could happen when the trait is a sexually dimorphic adaptation, and socialization is working to reduce the sex difference. For example, if physical aggression is a sexually dimorphic adaptation, dimorphic socialization could reduce the sex difference if boys were rewarded for non-aggressive behavior while girls were rewarded for aggressive behavior).
When Can We Expect to See a Sex Difference?
| THE EVOLVED ADAPTATION IS SEXUALLY: | |
THE “SEX ROLE” SOCIALIZATION IS: | MONOMORPHIC | DIMORPHIC |
Monomorphic | No sex difference | Sex Difference |
Dimorphic – socialization that tends to increase a sex difference | Slight sex difference | Sex Difference -- possibly large |
Dimorphic – socialization that tends to reduce a sex difference | No sex difference | Sex difference – variable in size depending on the degree of the dimorphism of the adaptation |
To help to understand the interaction of these genetic and environmental factors, let us first consider physical traits that are sexually dimorphic, and that are either sexually dimorphic or monomorphic adaptations.
EXAMPLE: Height (a strongly sexually dimorphic physical adaptation)
| ADAPTATION IS SEXUALLY: | |
“Sex Role” Socialization: | Monomorphic | Dimorphic |
Monomorphic |
| A sex difference is expected (and observed) |
Dimorphic – socialization that tends to increase a sex difference |
| A sex difference possibly larger than is typical in most cultures |
Dimorphic – socialization that tends to reduce a sex difference |
| Sex Difference -- possibly smaller than is typical in most cultures |
If a sex difference is based on a dimorphic adaptation, then the magnitude of the sex difference will vary somewhat across cultures. However, the direction of the sex difference will tend to remain (e.g., males > females, or vice versa), and very few cultures will be found in which the sex difference is reversed.
EXAMPLE: Head hair length (a sexually monomorphic physical adaptation)
| ADAPTATION IS SEXUALLY: | |
“Sex Role” Socialization Is: | Monomorphic | Dimorphic |
Monomorphic | No sex difference in hair length |
|
Dimorphic – socialization that tends to increase a sex difference | Sex difference in hair length |
|
Dimorphic – socialization that tends to reduce a sex difference | No sex difference in hair length |
|
If a sex difference is based on a monomorphic adaptation, then the magnitude of sex difference will vary across cultures, and it will likely vary in both directions (both positively and negatively), depending on sex role socialization. In this situation, one can expect to see many sex difference “reversals” across cultures, perhaps with close to 50% of the cultures “reversed” on the sex difference.
EXAMPLES: Hair length (sexually monomorphic) and sex drive (sexually dimorphic)
| PSYCHOLOGICAL ADAPTATION IS LIKELY TO BE SEXUALLY: | |
“Sex Role” Socialization: | Monomorphic FOOD PREFERENCES | Dimorphic PHYSICAL RISK TAKING |
Monomorphic | No sex difference in food preferences | Sex difference, M > F |
Dimorphic – socialization that tends to increase a sex difference | Sex difference in food preferences (probably small) | Large sex difference, M > F |
Dimorphic – socialization that tends to reduce a sex difference | No sex difference in food preferences | Smaller sex difference, M > F, or no sex difference (M = F) |
Closing Comments
Heretofore, most textbooks on sex and gender differences have assumed that males and females have evolved sexually monomorphic psychological adaptations. Given this assumption, socialization has often been presented as the only, or at least the primary, cause of sex differences. Biology and evolution could be safely ignored in this paradigm. However, as you can see from the tables above, if dimorphic psychological adaptations have evolved, this “socialization-only” approach will not help to produce an accurate picture of the underlying causality of sex differences.
Here we will focus our attention on what theory and/or empirical evidence suggest are likely to be sexually dimorphic psychological adaptations. In this way, we hope to identify what is most fundamental about “male nature” and “female nature,” and how they differ.
Of course, we also must examine the effects socialization can have on sexually dimorphic adaptations. We will explore how socialization and biology interact in a variety of cultures particularly in Chapters 8 though 12. However, keep in mind that we will keep our focus on the relatively few robust, biologically based sex differences, rather than the great many similarities between the two sexes.
Evolutionary psychology, is a relatively new scientific paradigm -- it is a new theoretical "lens." It has potential to bring gender differences "into focus" -- a far better focus than has been achieved by competing theories. You will find that the intellectual journey you are about to take will be a fascinating one. Since evolutionary psychology has not had time to permeate into the larger, popular culture, many of the concepts, terminology and analyses will be new, and some concepts may initially strike you as somewhat strange.
New paradigms are often highly controversial. At first they typically sound patently untrue to those who have yet to invest the time to fully understand them. For example, it is ironic that, instead of being congratulated for their efforts and insight, Copernicus and Darwin were badly mistreated by their contemporaries. A similar reception was given to many pioneers who have applied evolutionary thinking to the study of human behavior. Some of the first theoreticians to apply new evolutionary approaches to the study of sex differences were Richard Dawkins (The Selfish Gene, 1976), E. O. Wilson (On Human Nature, 1978), Symons (The Evolution of Human Sexuality, 1979), Daly and Wilson (Sex, Evolution and Behavior, 1981). Decades later, the idea that evolutionary theory can help illuminate our understanding of fundamental sex differences is still controversial in the social sciences.
Before we use an evolutionary perspective to explore why males and females differ, let's first explore how males and females differ, both in their brains and their bodies. We will do this in the next chapter.
Aron, A. & Aron, E.N. (1986). Love and the
Expansion of Self: Understanding Attraction and
Satisfaction. New York: Hemisphere.
Dennett, D. C. (1995). Darwin's dangerous idea: Evolution
and the meanings of life. NY: Simon & Schuster.
Eagley, A. H. (1995). The science and politics of
comparing
women and men. American Psychologist, 50, 3, 145-158.
Feingold, A. (1995). The additive effects of differences
in central
tendency and variability are important in comparisions
between
groups. American Psychologist, 50, 1, 5 - 13.