In 1974, while still a graduate student, Stewart had started working on an investigation of the Radcliffe class of 1964 that was inspired by personality tests the women had taken as freshmen. Over the years, she grew increasingly interested in the process of "middle aging," but the women she interviewed in her studies kept talking about confidence, so Stewart shifted gears to focus specifically on that."I think middle age for everybody involves a sense of ownership of one’s self and clarity about who you are — the absence of which is part of what is difficult about being a young adult," Stewart told me during a phone interview. She looks back on her own days as a young academic to illustrate the point. "I watch my graduate students struggle with their first publications, and I remember feeling the way they feel," she says. "But I don’t feel that way anymore. It’s part of learning about yourself: ‘Oh, I’m not so anxious about this anymore; it doesn’t feel like my whole value as a human being rides on this.’"Stewart’s Radcliffe research, like the Mills and MIDUS studies, is longitudinal, meaning it tracks the same group of people over time. When Stewart interviewed some of the Radcliffe women at 50, many said they felt they’d come into their own, finally understanding their strengths and limitations. "One woman talked about being able to deploy herself effectively," Stewart recalls. "I think what she meant was that she knows what she can do and that she cannot waste time trying to do things she’s not good at. Ironically, being more comfortable accepting what you can’t do leads you to be more confident about what you can do." The women also talked — a lot — about not worrying about what other people think.Perspective, PleaseThe women’s comments are supported by data, and not just from the Radcliffe sample. Stewart and her colleagues have published several recent studies examining the lives of women from different generations and social classes. In one, they surveyed University of Michigan alumnae at the ages of, on average, 26, 46, and 66 to assess various traits, including confidence. The women in their 20s scored much lower in confidence than those in their 40s, who scored much lower than those in their 60s.This sense of increasing confidence often leads to what Stewart refers to as midcourse corrections — that is, major growth-oriented changes in their educational or work lives that women make in their early 40s as a result of regrets they articulate in their late 30s."It is not too late to change," says Lachman, who perceives midlife for women as both a time of reconsideration and a time of mastery. "You’ve got a lot of opportunities," she says. "There is still a lot of resilience and reserve capacity. People get wake-up calls." And it’s easier to answer them, Lachman says, because many middle-aged and older adults are also better at regulating their emotions, experiencing more positive moods and fewer negative ones: "With experience, you realize that one failure is not necessarily going to destroy your career. And you begin to put things into perspective." The finding is supported by research on brain activity at various life stages.The strongest scientific evidence yet for the confidence surge could emerge from the combined analyses of the MIDUS I and MIDUS II studies. Together, they could represent the most accurate picture of midlife to date, given the size of the project, its socioeconomic diversity, and its longitudinal nature (it compares the original participants, 75 percent of whom returned, with their decade-older selves). The findings have not yet been released to the public.But the principal investigator for MIDUS II, Carol Ryff, discussed her preliminary impressions. Ryff, 57 and the director of the Institute on Aging at the University of Wisconsin-Madison, where she is also a psychology professor, developed a model with which to measure well-being, incorporating elements such as autonomy, self- acceptance, and a sense of purpose in life.