Are you a scientist? Are you known as “the science person” within your group of friends? Have you ever reflected on a professional milestone—your first publication, first large conference, or first lab class—and thought to yourself “wow, I’m a real chemist now”?
Justin Carmel, an assistant professor of chemistry at Florida International University, explains that an identity is “this feeling that you either belong or don’t belong to a certain group of people.” Science identity intersects with and is influenced by a person’s other identities, such as race, gender, or religion. Researchers have found that the sense of belonging in science matters more than grades and background when it comes to keeping students in science majors and careers.
Qualitative research into science identity requires subjective methods that probe personal experiences and interactions.
▸ Focus groups
▸ One-on-one interviews
▸ Journal prompts
▸ Observations in the lab or classroom
Creating a more equitable, inclusive, and leakproof STEM pipeline requires helping more students feel like scientists. The challenge is, how do science educators do that, when feeling like a scientist is such a personal thing?
Education researchers who study science identities break the concept down into several components. In 2007, education researchers Heidi Carlone and Angela Johnson defined the elements of science identity as competence (knowledge and understanding of science); performance (someone who can use the tools of a scientist, walk the walk, and talk the talk); and recognition (someone who is recognized as a science person by “meaningful others” such as family or other researchers).
Royel Johnson, a professor of education policy at Pennsylvania State University, surveyed a large cohort of Black students who enrolled in a STEM scholar program for racialized minorities and followed them for 2 years. He and his research team found that Black students with a higher sense of science identity were more likely to stay within STEM. But science identity alone wasn’t enough to keep them in STEM. Experiencing multiple instances of discrimination in the program counteracted that sense of science identity. “It’s hard to see yourself being successful within a STEM field when you’re exposed to discriminatory faculty members or peers who make you feel inferior,” says Johnson.
When probing science identity, chemical education researchers like Johnson have both qualitative and quantitative methods available to them. The advantage of quantitative methods such as surveys is the ability to see big patterns in a large group of STEM students, says Paulette Vincent-Ruz, a chemical education researcher at the University of Michigan. “We can see big relationships and create this theoretical link between things.” However, she cautions, researchers must think critically when designing surveys and interpreting quantitative results, because of the subjective nature of the questions. When a researcher asks students to rate the strength of their agreement with the statement “I am a science person” on a scale of one to four, the researcher has no control over how each person interprets that statement, Vincent-Ruz says.
“We need to understand that we cannot just take the numbers and automatically ascribe them to a group of people,” Vincent-Ruz says. “We need to be really critical about the assumptions we’re making when we’re defining things. When we’re defining ‘race’ and ‘science’ in a survey, we need to be really transparent about what we mean and are assuming by that.”
Although strong science identity is a predictor of academic success, other factors also help shape how students think about STEM careers. Vincent-Ruz studied science identity in high school students to better understand gender disparities in science. Female students were more likely to express an interest in biological or medical careers, and male students were more likely to express an interest in tech or engineering. This pattern held true regardless of how strongly the participants identified as scientists. “Even when you have a high [science] identity, people are going to make choices that match social expectations of them,” Vincent-Ruz explains. “Women are going to be pressured to present that identity in a way that is more feminine, or more gendered.”
By the time these students are in college pursuing STEM majors, their science identity starts to narrow, Vincent-Ruz hypothesizes. A high school student might think of themselves as “a scientist,” but that identity might become “a chemist” in college and “an analytical chemist” in graduate school.
Although researchers previously assumed most science identity formation happens in high school, not as much is known about what happens to science identity in college. “Students that come into college as STEM majors and they leave—is part of that because their science identity changes?” wonders Kathryn Hosbein, a chemical education researcher at East Carolina University.
Chemical education researchers like Carmel have looked at science identity in chemistry majors and how it influences academic success. In one study, during his postdoc at Michigan State University, Carmel and other researchers found three science identity trends in students over a semester of general chemistry: high and stable, moderate and slightly increasing, and moderate and declining. Students with “high and stable” science identities saw themselves as more competent at chemistry and on average had higher grades than students in the other categories. Students from racialized minorities were disproportionately clustered in the “moderate and slightly increasing” group.
Science identity surveys highlight important trends within chemistry but cannot explain what is causing inequitable academic outcomes for STEM students. Vincent-Ruz says that qualitative methods can offer a “deep dive” into individual science identities, providing important context for quantitative surveys.
An individual’s science identity is amorphous and subjective, which makes it difficult to measure, Hosbein says. Carmel and Hosbein use survey tools adapted from the work of physics education researcher Zahra Hazari at Florida International University. In interviews, Hosbein asks students open-ended questions about whether their parents see them as chemistry people, how the students know this, and if they enjoy learning chemistry.
In addition to exploring science identity as a stand-alone identity, many science education researchers examine how science identity intersects with gender and race. Terrell Morton, a professor of education at the University of Missouri, has interviewed Black women in STEM majors, including chemistry, to find out how their scientific, racial, and gender identities impacted their experiences in science. “I look to see the extent to which science or STEM is seen or understood by the participant as being part of or the opposite of who they are as a person and why,” Morton explains. Many of these women described a need to overcome perceived stereotypes of Black women if they were to succeed in STEM. They developed identities as strong, independent Black women to help them navigate difficult and hostile science environments.
For Black women who do persist in STEM, Morton explains, many are motivated by the desire to prove people wrong about themselves, to be a role model for the younger generation of Black scientists, or to promote change within their communities. “They see STEM or science as the vehicle [through] which they can promote change,” says Morton.
However, some of the coping mechanisms Morton’s study participants use to succeed in science can become detrimental to their overall well-being. If students decide that they have to work twice as hard as other people to survive a hostile environment, it may lead to burnout, he says.
Sarah Rodriguez, a professor of higher education and learning technologies at Texas A&M University–Commerce, studies issues of equity and retention for Latino students in higher education. Rodriguez’s interviews with Latina STEM students provided insights into their motivations and persistence within science that surveys could not elucidate. “A lot of times we don’t think science and religious identities would go hand in hand,” she notes. However, some Latina students she interviewed viewed religion as a source of strength when going through science programs, and a greater understanding of science deepened their spiritual appreciation of the world around them. Other participants spoke about the importance of family and community supporting them through their studies.
Rodriguez sees recognition and support of these intersecting identities as crucial, not only for keeping marginalized groups within STEM but also for improving the quality of science as a whole. “Hegemonic thought, hegemonic experience, having everyone have the same experience and background, is not good science, period,” she says.
Helping students in marginalized groups see themselves as scientists requires change at the institutional and individual level. Vincent-Ruz says mentors should educate themselves about the possible experiences somebody who does not look like them may go through in academia, and not assume their mentoring relationship must focus solely on science. “We’re acting like science is something that only happens when the individual is working by themselves. Science is a social phenomenon,” Vincent-Ruz says. When a professor or other mentor is working with a student from a marginalized group, they have to realize common experiences they both face, such as a heated discussion with a colleague, may lead to “two different realities,” Morton says.
Carmel’s research also points to role models being important for individuals’ success in STEM. “That role model is helpful because it shows students that there’s this attainable goal—I would like to be like that—and it gives them a target.” The role model does not necessarily have to be from the same racial or gender group as the student, Carmel says, but it’s important their values and personalities align.
Small but sincere interactions can make a huge difference to someone’s sense of belonging. Rodriguez recalls a story one of her students told. Several years ago, this student presented her undergraduate research at a conference. After her talk, a professor she didn’t know approached her and said: “I really appreciated your work . . . I look forward to reading more.” The student told Rodriguez, “that was the moment I knew that [science] was for me.” This student, a Latina who carried this memory for years, went on to earn a science PhD, Rodriguez says. Just by briefly talking to her as if she were a peer, the professor at the conference provided external validation of the student’s science identity.
Helping students from racialized minorities see themselves as scientists also requires redefining traditional notions of science, Vincent-Ruz says. What many people think of as science is derived from white, European norms. In Morton’s research, this dominant view of science stops his Black participants from truly belonging in the sciences the way white scientists can: “STEM is not part of who they are but rather something that they do,” he explains. To Morton, widening the definition of science means getting away from rigid textbook definitions of scientific inquiry and instead doing a better job of equating scientific research to lived experiences such as cooking or herbal medicine, where there isn’t always a right answer or a strict protocol.
There’s still a lot science educators don’t know about how science identity forms and how it influences success in STEM. Johnson, who has advised universities and colleges on higher education equity issues, points out that meaningful promotion and protection of science identities must occur at the institutional level. “It’s not enough to just get students into the [STEM] programs, we also have to talk about the larger STEM culture and the climate students are navigating within their college,” Johnson says. “We need to really reform the culture in ways that engender a sense of belonging for all students.”
Claire L. Jarvis is a freelance science writer based in Atlanta.