It's a Girl! Gender Differences
Ages: Birth to Eighteen Years
Three-year-olds take gender roles as seriously as drag queens do. One of our colleagues, who was dedicated to freeing her kids from traditional gender expectations, bought a doll for her son and trucks for her daughter. She gave up her quest after she found the boy using the doll to pound in a nail and the girl pretending that the trucks were talking to each other.
Many puzzled parents have wondered where this highly stereotyped behavior comes from, especially in households where Mom wouldn't be caught dead in a pink frilly dress and Dad would rather cook dinner than watch sports. All over the world, a phase of intense adherence to a sex role seems to be important for the development of a solid gender identity. This stubbornness reminds us of the early stage of grammatical learning, another area where young kids apply newly learned rules more broadly than necessary ("That hurted my foots" instead of "That hurt my feet").
In light of their behavioral differences, you'd probably imagine that the brains of little boys and girls are distinct in many important ways. Because of our society's intense interest in sex differences, researchers have done many thousands of studies of this topic, and journalists have been eager to publicize them. This literature is vast and variable, so as we evaluated it, where possible we relied on meta-analysis (see glossary) to evaluate the findings. From careful review of such papers, a few important patterns emerge.
When we evaluate reports of sex differences, it's important to pay attention to the size of the effects. Most gender differences are too small to matter in any practical way, and a minority of differences are important when comparing groups. But only a few tell us anything significant about individuals. For instance, girls—on average—are more likely to hear a relatively quiet sound. But it would be impossible to guess whether a particular child is male or female by knowing that child's hearing ability, because all possible scores are found in both boys and girls. And what's more, for nearly all sex differences, the differences among individual girls or among individual boys are much larger than the average differences between the sexes, with a few important exceptions.
What do we mean when we say that a gender difference is small or large? Let us be technical for a moment. Scientists often measure the size of a difference between two groups by calculating a statistic called d-prime (d') or effect size, defined as the difference between groups divided by the standard deviation, a measure of variability, of one or both groups. If there is no difference, the d' is zero. The d' gets bigger as the size of the difference in average scores between the groups increases, relative to the range of scores within each group.
This idea is easier to explain in pictures than in words. The figure below shows the differences between groups that correspond to typical d' values. The horizontal axis represents the possible scores, while the height of the curve represents the number of people in the population who get a particular score. From top to bottom, these differences would be considered small, medium, and very large.
Let's consider some specific examples. For gender differences in adult height in the U.S., the left curve in the bottom panel (d' = 1.9) would represent women, and the right curve would represent men. The horizontal axis would show heights from short (left) to tall (right), with the peak of the female curve at 5 feet 3.8 inches, the average height for women, and the peak of the male curve at 5 feet 9.4 inches, the average height for men. A man of average height is taller than 92 percent of women. In the research literature, a value of d' that is at least 0.8 is considered large, so this would be a very large difference.
At the other extreme, let's take as our example a small difference that we've already mentioned: hearing. Several authors have recently argued in favor of single-sex education based in part on the idea that girls have more sensitive hearing than boys and therefore respond best to teachers speaking quietly. For hearing sensitivity, the left curve in the top panel (d' = 0.2) would represent boys, and the right curve would represent girls. Because the difference between the two groups is small, as you can see, the two curves overlap substantially.
The individual differences in hearing within each sex are much larger than the differences between boys and girls. And given that many boys have sharper hearing than many girls, it doesn't make sense to argue for sex segregation on these grounds. If you think sensitive hearing affects the way people learn, you should separate them based on their hearing, not their gender—the two are not the same.
Only a few gender differences are big enough to predict individual behavior. The largest known behavioral difference at any age is toy preferences in three-year-olds. Parents who try to keep their sons from playing with toy guns often discover that any stick—or, in a pinch, even a doll—can be converted to a weapon in a boy's imagination. Given the choice between a boy-typical toy like a car and a girl-typical toy like a tea set, at age three children differ in their choice of boyish toys with a d' of 1.9, a difference corresponding to the bottom panel of the figure. This means that you can do quite well at guessing the sex of a young child based on his or her choice of toys, as 97 percent of boys are more likely to play with male-typical toys than an average girl. Because play helps children learn and practice a variety of skills, sex differences in how children spend their time can influence which abilities they carry through life (see Practical tip: Broadening your child's abilities, p. 68).
The emergence of toy preferences is an early stage in the development of gender identity, defined as your child's self-identification as male or female. Gender-influenced toy preferences are seen across cultures, beginning around one year of age. Even babies have some understanding of gender (see chapter 1), but only a few two-year-old children can accurately state whether they are boys or girls or reliably distinguish men from women in pictures. Most children—again across cultures—reach this milestone by two and a half, and almost all children get there by age three. Children who have reached this milestone are less likely to choose the "wrong" toy than children who have not.
Toy preferences almost certainly have an innate basis (though they are also influenced by culture). One clue is that male monkeys prefer to play with trucks, while female monkeys prefer dolls. Another clue is that boy-typical toy preferences are more common in girls with a syndrome called congenital adrenal hyperplasia or CAH. Due to a genetic defect in adrenal hormone synthesis, CAH girls are exposed to an excess of testosterone and other androgens, masculinizing their brains and to some extent their bodies in utero. Because this hormonal malfunction can be treated starting at birth, CAH girls offer an opportunity to look at the effects of prenatal exposure to male hormones on later behavior.
As they get older, girls tend to become more flexible in their toy preferences. By age five, nearly half will pick a boy-typical toy when offered a choice. Boys, on the other hand, continue to refuse girl-typical toys, most likely because the social penalty for acting like a girl is very steep. Both peers and parents—especially fathers—actively discourage boys from playing with girl toys.
Some parents are concerned that allowing their son to play with girl toys will lead him to be gay in adulthood, but this worry confuses correlation with causation. Whether parents encourage or approve of their son's habits is irrelevant to his sexual orientation later in life. It is true that about half of the boys who prefer girl toys do grow up to be gay—and also true that many of them do not. (Tomboy girls, on the other hand, rarely turn out to be lesbians.) Playing with dolls doesn't cause boys to become gay, though. The most likely explanation is that playing with girl toys and adult homosexuality both result from earlier influences on some boys' brains, perhaps due to prenatal experiences or genetics.
Psychiatric treatment aimed at encouraging boyish behavior has no effect on adult sexual orientation, but the father who tries to discourage girlish behavior might be opening a rift with his son. By the time you can observe the behavior, the outcome is out of your control, so you might as well get comfortable with it.
You've probably noticed two other sex differences in young children's behavior. Boys are significantly more active and more physically aggressive than girls. These differences are medium sized, with a d' of 0.5, corresponding to the second panel of the figure on p. 64. That means an average boy is more active and more physically aggressive than 69 percent of girls, which does not predict individual behavior very well, but does make groups of boys obviously different from groups of girls. These differences are also probably influenced by the action of hormones on the brain, not just by culture. Juvenile male monkeys show more rough-and-tumble play than female monkeys, and this behavior can be modified by hormone treatment. Similarly, CAH girls are more aggressive and more active than other girls, again suggesting an early hormonal influence on this behavior.
Perhaps as a consequence of these behavioral differences, starting as early as two or three years of age, children prefer to play with other children of the same sex. These segregated play groups persist throughout elementary school. This pattern is seen across many societies, from agricultural villages to big cities. It does not seem to depend on whether or not the adults in the society have strongly segregated sex roles, though cultural factors can modify the pattern's expression somewhat. It even occurs in monkeys and apes. If there are only a few children available, boys and girls will play together, but when possible, they usually split the group by gender.
This behavior reinforces gender norms as children are learning about their gender identities. Social pressure from other children to conform to gender norms is particularly strong from age four to eight, perhaps because children's early concepts of sex roles (and many other rules of society) tend to be written in black and white, with a more flexible understanding emerging later in development. We know a neuroscientist couple whose son was best friends with a girl throughout the preschool years. When he was six, his male friends at summer camp let him know that playing with girls was unacceptable. Now their son will only see his female friend inside the house, with everyone involved sworn to secrecy so the other boys won't find out. Single-sex groups often act in such a way to communicate and enforce gender-specific behaviors.
Sex differences in behavior give girls a medium-sized advantage over boys in the classroom, where girls get better grades in high school and college. Girls' brains mature earlier than boys' brains, with the peak volumes of most brain structures occurring one to three years sooner in girls. Girls are moderately better at inhibitory control (d' of 0.4)—that is, sitting still and concentrating on their task—so the classroom culture is more friendly to girls. On average, girls are a bit more advanced in some areas of verbal development when starting school. Boys lag at fine motor coordination (d' of 0.6), giving them a moderate disadvantage in the ability to write letters, the largest sex difference in academic performance as school begins. These gaps persist through high school, with boys continuing to score lower on tests of both reading and writing.
But let's put these gender comparisons in perspective. All these differences have smaller effects than the difference between living in a middle-class neighborhood with good schools (judged by their average test scores) and living in a low-income neighborhood with poor schools. First graders from poor areas score lower than their middle-class counterparts (d' greater than 1.1) in both reading and mathematics performance, and those gaps too typically widen with age (see chapter 30).
Like other gender differences, gender-related differences in education may be modified by experience. Girls have recently caught up with boys in academic areas where they were lagging just a decade or two ago. In the U.S., there are no remaining gender differences in average performance on mathematical achievement tests through high school. In addition, women are now more likely than men to attend and complete college. In the U.S., there are 185 women for every 100 men with college degrees at age twenty-two. Because men take longer than women to graduate, this gap narrows considerably at later ages, but it does not close completely.
To help reduce this gap, we suggest that boys might benefit from extra training in language and study skills during the early school years. The most efficient way to improve overall performance would be to provide such help evenhandedly to all children who need it, a group that would include more boys than girls.
Curiously, in the face of this female progress, the famous male advantage on the mathematics section of the SAT has not narrowed at all (d' = 0.4, thirty-five points as of 2009). Despite their lower average SAT scores, however, women get better grades than men in college math classes. One study found that male freshmen get the same grades in college math classes as women whose math SAT scores are thirty-five points lower. Indeed, nearly all the standardized tests required for college admission underpredict the future grades of women. This poor prediction may be due to better study habits among women (giving them higher grades for the same aptitude), or it may be due to gender bias in the SAT (giving women lower scores for the same aptitude). Either way, the lower grades of college men relative to women can be attributed in part to the use of standardized tests for admission decisions.
Another well-known group of sex differences falls in the realm of emotional behavior. These differences are not as large as most people believe. Effect sizes range from small to medium. These differences do not predict individual behavior very well, but some of them are noticeable at the group level. Girls are more likely than boys to express fear and to cry, but in both sexes the physiological responses to distress are similar. Many differences are so small that they are drowned out by individual variability within each sex. One example is the idea that boys make moral decisions based on justice (d' = 0.19), while girls make moral decisions based on relationships (d' = 0.28). Similarly, girls are only slightly better than boys at identifying emotions in other people's faces (d' = 0.19 in childhood and adolescence). Boys are only slightly more likely than girls to take risks at all ages (d' = 0.13), with a larger effect between ages ten and thirteen, and between ages eighteen and twenty-one (d' around 0.25). The gender gap in risk taking seems to be closing over time, as it was smaller in later studies (1980s and 1990s) than in earlier studies (1960s and 1970s).
Kids return to mixed-sex socializing as teenagers. The hormones of puberty usher in a medium-sized sex difference (d' = 0.53): 70 percent of teenage boys masturbate more often than the average teenage girl, a pattern that continues into adulthood. The size of this difference has declined from a d' of 0.96 over the past two decades, suggesting a strong cultural influence. Sex differences in self-esteem peak in adolescence as well, with teenage girls showing lower self-esteem than teenage boys (d' of 0.33, another small difference).
One area where girls could clearly use extra support is body image. Especially as teenagers, girls experience much more dissatisfaction than boys with their bodies, which is a risk factor for eating disorders and depression. The size of this difference increased from a small d' of 0.27 in the 1970s to a moderate d' of 0.58 in the 1990s, perhaps because of the progressively thinner standards of female beauty across recent decades.
Even if you're concerned about your daughter's weight, criticizing her body is likely to be counterproductive. In one longitudinal study (in which the same people are followed over a period of time), teenage girls and boys who reported being teased about their weight by family members were much more likely than average to have developed an eating disorder or to have become overweight five years later. Similarly, in another longitudinal study, repeated dieting in fourteen-year-old girls (many of whom were not overweight when the study began) increased their risk of becoming overweight a year later by almost a factor of five.
One interesting thing about all of these sex differences is that the size of the difference does not predict how malleable it is. Though initial preferences can be modified by environmental influences, they often do launch boys and girls onto paths that can lead them to have different experiences for much of childhood. To sum it up, your child probably has some initial inclination toward fixing cars, taking care of babies, or whatever, but there are many opportunities to broaden your child's horizons by introducing him or her to new interests.
From Welcome to Your Child's Brain: How the Mind Grows from Conception to College by Sandra Aamodt and Sam Wang. Copyright 2011 by Sandra Aamodt and Sam Wang. Excerpted by permission of Bloomsbury.