Teenage turning point: does adolescence herald the twilight of girls' self-esteem?

Teenage Turning Point

Youngsters often experience a decline in self-esteem as they enter their adolescent years, a time marked by the abrupt move from the relatively cloistered confines of elementary school to the more complex social and academic demands of junior high. Social scientists have documented this trend -- often more pronounced among girls -- over the past 20 years through questionnaires and interviews aimed at gauging how adolescents feel about themselves.

But a new survey of U.S. elementary and secondary students bears the worst news yet about plummeting self-esteem among teenage girls. The controversial findings, released in January by the American Association of University Women (AAUW), have refocused researchers' attention on long-standing questions about the meaning of such studies and their implications, if any, for educational reform and for male and female psychological development.

The concept of self-esteem itself remains vague, contends psychiatrist Philip Robson in the June 1990 HARVARD MEDICAL SCHOOL MENTAL HEALTH LETTER. Some researchers assess a person's "global" self-esteem with questions about general feelings of worth, goodness, health, attractiveness and social competence. Others focus on people's evaluations of themselves in specific situations. Robson, of Oxford University in England, notes that an individual might score high on one type of test but not on another, presumably because the measures reflect different aspects of self-esteem.

Moreover, he argues, high test scores may sometimes indicate conceit, narcissism or rigidity rather than healthy feelings of self-worth.

Despite the complexities involved in determining how people tryly regard themselves, the AAUW survey suggests that adolescent girls experience genuine, substantial drops in self-esteem that far outpace those reported by boys. Girls also reported much less enthusiasm for math and science, less confidence in their academic abilities and fewer aspirations to professional careers.

The survey, conducted last fall by a private polling firm commissioned by AAUW, involved 2,400 girls and 600 boys from 36 public schools throughout the United States. Black and Hispanic students made up almost one-quarter of the sample. Participants, whose ages ranged from 9 to 16 (fourth through tenth grades), responded to written statements probing global self-esteem, such as " I like the way I look" and "I'm happy the way I am."

In a typical response pattern, 67 percent of the elementary school boys reported "always" feeling "happy the way I am," and 46 percent still felt that way by tenth grade. For girls, the figures dropped from 60 percent to 29 percent.

For both sexes, the sharpest declines in self-esteem occurred at the beginning of junior high.

Compared with the rest of the study sample, students with higher self-esteem liked math and science more, felt better about their schoolwork and grades, considered themselves more important and felt better about their family relationships, according to the survey.

Boys who reported doing poorly in math and science usually ascribed their performance to the topics' lack of usefulness, whereas girls who reported a lack of success in these areas often attributed the problem to personal failure.

Although the survey included too few boys to allow a racial breakdown for males, race did appear to play an important role in the strength of self-esteem among girls. White and Hispanic girls displayed shapr drops in all the measured areas of self-esteem -- appearance, confidence, family relationships, school, talents and personal importance -- as they grew older. In contrast, more than half the black girls reported high levels of self-confidence and personal importance in both elementary and high school, and most attributed this to strong family and community support, says psychologist Janie Victoria Ward of the University of Pennsylvania in Philadelphia, an adviser to the study. Their confidence in their academic abilities, however, dropped substantially as they passed through the school system, Ward says.

"Something is going on in the schools that threatens the self-esteem of girls in general," asserts psychologist Nancy Goldberger, another adviser to the survey. "A lot of girls come to doubt their own intelligence in school."

Goldberger, who teaches psychology at the Fielding Institute in Santa Barbara, Calif., calls for intensive, long-term studies to address how schools short-change female students.

An AAUW pamphlet published last August argues that school-age girls represent the proverbial square peg attempting to fit into the round hole of most educational programs.

Starting early in life, societal pressures urge girls and boys to think and behave in contrasting ways that create gender-specific learning styles, according to the AAUW pamphlet. Schools, however, generally tailor instructional techniques to the learning style of boys, leaving girls with a tattered education and doubts about their academic abilities, the pamphlet contends.

This argument rests heavily on research directed by Harvard University psychologist Carol Gilligan. In her much-praised and much-criticized book, In a Different Voice (1982, Harvard University Press), Gilligan asserted that girls and boys generally follow divergent paths of moral development. She based her contention on several studies of Harvard undergraduates, men and women at different points in the life cycle, and women considering abortion.

In Gilligan's view, females respond to an inner moral voice emphasizing human connections and care, and they attempt to solve moral dilemmas by responding to the needs and situations of those affected by the problem. Males, on the other hand, focus on abstract principles such as justice and follow a moral code centered on the impartial application of rules of right and wrong.

Gilligan's most recent research, described in Making Connections: The Relational Worlds of Adolescent Girls at Emma Willard School (1990, Harvard University Press), draws on findings collected over a three-year period among 34 students at a private girls' school in Troy, N.Y. Gilligan and her co-workers argue that many girls, at least in this predominantly white, privileged sample, show an aggressive confidence in their identities and ideas around age 11, only to find their self-assurance withering by age 15 or 16.

During this period of increasing separation from parents, marked by a search for an independent identity and future career possibilities, girls feel torn between responding to others and caring for themselves, the Harvard researchers maintain. In addition, they say, adolescent girls encounter more pressure from parents and teachers to keep quiet and not make a fuss than do adolescent boys or younger girls.

The gender gap seen in academic achievement during early adolescence arises largely because a social and educational emphasis on career development and personal advancement clashes with girls' distinctive sense of connection to others, Gilligan's team asserts. The researchers maintain that girls often learn best and gain increased self-confidence through collaboration with other students and faculty, not through competition among individuals as practiced in most schools.

Boys in contrast, often perform best on competitive tasks or in games with a strict set of prescribed rules, the investigators contend.

Some adolescence researchers argue that Gilligan paints too stark a contrast between the moral development of boys and girls. Others say Gilligan's ideas have an intuitive appeal, but her small studies lack a sound empirical foundation on which to build educational reforms. These researchers see Gilligan's work as a preliminary corrective for previous studies, based largely on male participants, that suggested the ability to reason from abstract principles represented the pinnacle of moral development.

Similarly, social scientists differ over the extent to which self-esteem dips during adolescence and the meaning of the AAUW survey data. In fact, some investigators question whether a significant gender gap in self-esteem exists at all.

Most surveys of teenagers' self-esteem, including the AAUW project, focus on students and neglect school dropouts. This approach may lead to overestimates of self-esteem among boys, argues sociologist Naomi Gerstel of the University of Massachussetts in Amherst. More boys than girls drop out of school, and male dropouts may regard themselves in an especially poor light, Gerstel points out.

Furthermore, she says, since no one has examined the moral "voice" of boys in the intensive way Gilligan studied her group of girls, Gilligan's theory has yet to meet a scientifically rigorous test. Gilligan's ideas prove "problematic" when educators attempt to use them to formulate specific educational reforms, Gerstel writes in the Jan. 4 SCIENCE.

The self-esteem reports gathered in the AAUW survey fail to provide evidence for any particular need to change school instruction, contends psychologist Joseph Adelson of the University of Michigan in Ann Arbor. "It's been known for some time that girls report greater self-esteem declines in adolescence, but the reasons for those declines are unclear," he says. "It's inappropriate to take the correlations in this survey to politicized conclusions about educational reform."

In his view, gender differences in mathematics achievement remain particularly mysterious and probably stem from a number of as-yet-unspecified social or family influences (SN: 12/6/86, p.357). Preliminary studies directed by Carol S. Dweck, a psychologist at Columbia University in New York City, suggest that bright girls show a stronger tendency than bright boys to attribute their difficulty or confusion with a new concept -- such such as mathematics -- to a lack of intelligence. Thus, when bright girls confront mathematics, initial confusion may trigger a feeling of helplessness, Dweck writes in At The Threshold (1990, S. Shirley Feldman and Glen R. Elliot, editors, Harvard University Press).

Many girls with considerable potential in mathematics may deal with this sense of helplessness by throwing their energies into already mastered verbal skills, Dweck suggests. Rather than indict their intelligence, both boys and girls who shrink from challenging new subjects may need to learn how to channel initial failures into a redoubled effort to master the material, she says.

Gender differences in reported well-being -- an aspect of personal experience closely related to self-esteem -- also prove tricky to study, Adelson observes. A statistical comparison of 93 independent studies, directed by psychologist Wendy Wood of Texas A&M University in College Station, serves as a case in point. In examining these studies, which focused on well-being and life satisfaction among adult men and women, Wood and her colleagues found that women reported both greater happiness and more dissatisfaction and depression than men. Wood contends that societal influences groom women for an acute emotional responsiveness, especially with regard to intimate relationships, and that this helps explain why women report more intense emotional highs and lows than men.

"No clear advantage can be identified in the adaptiveness and desirability of [men's and women's] styles of emotional life," she and her colleagues write in the March 1989 PSYCHOLOGICAL BULLETIN.

Researchers have yet to conduct a similar statistical comparison of the literature on adolescent self-esteem and well-being. But according to Adelson, a persistent problem plagues the interpretation of all such studies. If females generally show more sensitivity to and awareness of emotions than males, they may more easily offer self-reports about disturbing feelings, creating a misimpression that large sex differences exist in self-esteem, he suggests.

Although this potential "response bias" muddies the research waters, psychologist Daniel Offer of Northwestern University in Evanston, Ill., cites several possible explanations for the tendency among early-adolescent girls to report more self-dissatisfaction than boys.

One theory holds that since girls experience the biological changes of puberty up to 18 months before boys, they may suffer earlier and more pronounced self-esteem problems related to sexual maturity. Several studies have found that early-maturing girls report the most dissatisfaction with their physical appearance, a particularly sensitive indicator of self-esteem among females. Social pressures to begin dating and to disengage emotionally from parents may create additional problems for early-maturing girls, Offer says.

Other research suggests that, unlike their male counterparts, adolescent girls often maintain close emotional ties to their mothers that interfere with the development of a sense of independence and self-confidence, Offer says. In addition, parents may interrupt and ignore girls more than boys as puberty progresses, according to observational studies of families, directed by psychologist John P. Hill of Virginia Commonwealth University in Richmond.

Despite these findings, the director of the most ambitious longitudinal study of adolescent self-esteem to date says her findings provide little support for the substantial gender gap outlined in the AAUW survey, which took a single-point-in-time "snapshot" of self-esteem.

During the 1970s, sociologist Roberta G. Simmons of the University of Pittsburgh and her co-workers charted the trajectory of self-esteem from grades 6 through 10 among more than 1,000 youngsters attending public schools in Milwaukee and Baltimore. Simmons discusses the research in Moving Into Adolescence (1987, Aldine de Gruyter).

Overall, adolescents reported a gradual increase in self-esteem as they got older, she says, but many girls entering junior high and high school did experience drops in feelings of confidence and self-satisfaction.

Simmons agrees with Gilligan that adolescent girls increasingly strive for intimacy with others. Large, impersonal junior high schools throw up a barrier to intimacy that initially undermines girls' self-esteem, Simmons asserts. As girls find a circle of friends and a social niche, their self-esteem gradually rebounds, only to drop again when they enter the even larger world of high school.

"We don't know if that last self-esteem drop [in high school] was temporary or permanent," Simmons points out.

As in the AAUW survey, Simmons' team found that black girls, as well as black boys, consistently reported positive and confident self-images.

But given the increased acceptance of women in a wide variety of occupations since the 1970s, Simmons expresses surprise at how much the self-esteem of girls lagged behind that of boys in the AAUW survey.

A new study of 128 youngsters progressing through junior high, described in the February JOURNAL OF YOUTH AND ADOLESCENCE, also contrasts with the AAUW findings. The two-year, longitudinal investigation reveals comparable levels of self-esteem among boys and girls, notes study director Barton J. Hirsch, a psychologist at Northwestern University. Hirsch and his colleagues used a global self-esteem measure much like the one in the AAUW survey.

The researchers gathered self-reports from boys and girls as the students neared the end of sixth grade, then repeated the process with the same youngsters at two points during seventh grade and at the end of eighth grade. Students lived in a midwestern city and came from poor or middle-class families. Black children made up about one-quarter of the sample.

In both sexes, about one in three youngsters reported strong self-esteem throughout junior high school, the researchers report. These individuals also did well in school, maintained rewarding friendships and frequently participated in social activities.

Another third of the sample displayed small increases in self-esteem, but their overall psychological adjustment and academic performance were no better than those of the group with consistently high self-esteem.

Chronically low self-esteem and school achievement dogged 13 percent of the students, who probably suffered from a long history of these problems, Hirsch says.

But the most unsettling findings came from the remaining 21 percent of the youngsters. This group -- composed of roughly equal numbers of boys and girls -- started out with high self-esteem, good grades and numerous friends, but their scores on these measures plunged dramatically during junior high, eventually reaching the level of the students with chronically low self-esteem.

The data offer no easy explanations for the steep declines seen among one in five study participants, Hirsch says. An examination of family life might uncover traumatic events that influenced the youngsters' confidence and motivation, but this remains speculative, he says.

One of the most comprehensive longitudinal studies of the relation between child development and family life (SN: 8/19/89, p.117) suggests that particular parenting styles produce the most psychologically healthy teenagers. The findings indicate that parents who set clear standards for conduct and allow freedom within limits raise youngsters with the most academic, emotional and social competence.

Directed by psychologist Diana Baumrind of the University of California, Berkeley, the ongoing study has followed children from 124 families, most of them white and middle-class. At three points in the youngsters' lives -- ages 3, 10 and 15 -- investigators assessed parental styles and the children's behavior at home and school.

Baumrind assumes that self-esteem emerges from competence in various social and academic tasks, not vice versa. For that reason, she and her colleagues track achievement scores and trained observers' ratings of social and emotional adjustment, not children's self-reports of how they feel about themselves.

In fact, Baumrind remains unconvinced that girls experience lower self-esteem than boys upon entering adolescence. Her study finds that girls in elementary grades show a more caring and communal attitude toward others, while boys more often strive for dominance and control in social encounters. But by early adolescence, she maintains, such differences largely disappear.

The gender-gap debate, however, shows no signs of disappearing. In a research field characterized by more questions than answers, most investigators agree on one point. "Most kids come through the years from 10 to 20 without major problems and with an increasing sense of self-esteem," Simmons observes.

Yet that trend, too, remains unexplained. "Perhaps the steady increase in self-esteem noted in late adolescence results more from progressive indoctrination into the values of society than from increasing self-acceptance," says Robson. "We simply do not have the empirical data necessary to resolve this question."