The Confidence Gap: Why Girls Doubt Themselves and How to Close It

Picture a classroom. It’s math time. A boy raises his hand, answers incorrectly, and shrugs it off. A girl raises her hand, answers incorrectly, and quietly mutters, “I’m just not a math person.” This scene isn’t an anomaly; it’s a daily ritual in schools worldwide, a tiny performance of a much larger, more insidious drama: the confidence gap. It’s the quiet, persistent erosion of a girl’s belief in her own abilities, a gap that doesn’t just appear in adolescence but is often meticulously constructed, brick by invisible brick, from the earliest years of childhood. This chapter isn’t about telling girls to be more confident. It’s about excavating the foundation of that gap—the psychological blueprints and societal scaffolding that create it—and then handing over the architectural plans for a new, sturdier structure. We’re going to move from the “why” to the “how,” from diagnosing the problem to building the solution, using the most robust research in developmental psychology, neuroscience, and education to forge a practical, evidence-based path forward. The goal here is to dismantle the myth that confidence is an innate trait some are born with and others aren’t, and instead reveal it as a skill, a muscle, a set of learnable behaviors that can be cultivated, practiced, and strengthened.

Let’s start with the most fundamental, and perhaps surprising, truth: the confidence gap isn’t a bug in the system; it’s a feature. It emerges not from a lack of ability—girls often outperform boys in grades and standardized tests throughout elementary and middle school—but from a divergence in how success and failure are internalized. To understand this, we must first distinguish between two closely related but critically different constructs: self-esteem and self-efficacy. Self-esteem is your overall sense of self-worth, the global feeling that you are a valuable person. Self-efficacy, a term coined by psychologist Albert Bandura, is your belief in your capacity to execute the behaviors necessary to produce specific performance attainments. It’s task-specific. You can have high self-esteem (“I am a good person”) but low self-efficacy in a specific domain (“I cannot speak in public”). The confidence gap we’re investigating is primarily a chasm in domain-specific self-efficacy, particularly in areas stereotypically associated with masculinity: STEM subjects, leadership, public speaking, competitive negotiation, and physical risk-taking. A girl can love herself deeply yet be utterly convinced she is incapable of coding, negotiating a salary, or leading a team.

Why does this specific-efficacy gap widen for girls? The answer lies in a complex interplay of internal attribution styles and external feedback loops. Research consistently shows that girls, from a young age, are more likely than boys to attribute their failures to stable, internal, and global causes—a fixed mindset. A boy who fails a math test is more likely to say, “I didn’t study enough” (unstable, internal, specific). A girl is more likely to say, “I’m bad at math” (stable, internal, global). This isn’t a character flaw; it’s a learned cognitive schema, often reinforced by well-intentioned adults. Consider the common praise given to a young child who completes a puzzle. To a boy, we’re more likely to say, “You’re so smart!” To a girl, we’re more likely to say, “You’re so pretty! … And you worked hard on that puzzle.” The first is an ability-based, fixed praise. The second, while containing a process element (“worked hard”), is often sandwiched between appearance comments, subtly framing the intellectual effort as a secondary, less defining characteristic. Over time, this creates a psychological landscape where girls learn to monitor their performance for signs of “not being smart,” while boys learn to monitor their effort. The fixed mindset becomes a self-fulfilling prophecy: if you believe ability is fixed, you avoid challenges (lest you prove you’re not smart), ignore useful feedback (it’s critical, not constructive), and see effort as a sign of weakness (if you were truly smart, it would be easy). This is the first, crucial layer of the confidence gap: a mindset gap.

But mindset doesn’t develop in a vacuum. It is sculpted by a relentless stream of gendered feedback from parents, teachers, peers, and media. This feedback operates on two primary channels: direct communication and modeling/availability.

On the direct channel, the language we use matters immensely. A seminal study by sociologists Eleanor Maccoby and Carol Jacklin in the 1970s, and countless replications since, have catalogued subtle differences. Teachers, even those committed to equality, call on boys more frequently for answers, especially in math and science, and provide them with more substantive feedback—both positive and corrective. When a boy gives a wrong answer, a teacher is more likely to rephrase the question or provide a hint (“Think about what we did yesterday”). When a girl gives a wrong answer, a teacher is more likely to simply say “No” and move on, or provide the answer herself. This sends a powerful, unspoken message: your thinking process is less valuable; your wrong answer is a dead end. In language arts, the feedback reverses: girls receive more praise for neatness and compliance (“What a beautiful, quiet writer you are!”) while boys receive more feedback on content and ideas (“That’s a wild story! Let’s talk about that character”). This channels girls toward being “good” at following rules and boys toward being “creative” and “original,” a divergence that has profound implications for later leadership and innovation.

The second, more insidious channel is modeling and availability. This is the “you can’t be what you can’t see” principle on steroids. From the earliest picture books, girls see a world where the scientists, engineers, presidents, and pilots are almost exclusively men. The hero’s journey is a boy’s journey. When girls look around their classrooms, their corporate boardrooms, their newsfeeds, they see a severe underrepresentation of their gender in positions of power, intellectual authority, and physical mastery. This isn’t just about inspiration; it’s about cognitive availability. Psychologist Shelley Taylor’s work on social support shows that when we face a challenge, we unconsciously scan our mental database for similar others who have succeeded. If that database contains almost no women who have succeeded in a challenging STEM field, the brain’s quick, heuristic assessment is, “People like me don’t do this.” It’s not a conscious belief; it’s a pre-conscious filter that saps motivation before conscious effort even begins. This phenomenon, famously named stereotype threat by Claude Steele and Joshua Aronson, demonstrates that when girls are reminded of the “girls are bad at math” stereotype—even subtly, like by checking a box for their gender before a test—their performance actually declines. The threat consumes cognitive bandwidth in the form of anxiety and self-monitoring, leaving fewer mental resources for the task at hand. The gap, therefore, is not just in confidence *before* the task, but in performance *during* the task, because the environment itself is cognitively taxing.

This brings us to the developmental timeline. The confidence gap isn’t a sudden adolescent crisis; it’s a slow leak that begins in early childhood and widens dramatically during key transitions. Let’s trace the arc.

Ages 4-6: The “I Can Do Anything” Peak. Developmental research shows that at this age, boys and girls have nearly identical levels of self-efficacy across domains. A four-year-old girl will confidently declare she will be an astronaut, a dinosaur, and a princess all before lunch. She believes effort leads to mastery, and her worldview is filled with possibility. The gendered feedback is present but not yet dominant. The cracks begin here, however, in micro-moments: the subt