About the SA Blog Network

Beautiful Minds

Beautiful Minds

Insights into intelligence, creativity, and the mind
Beautiful Minds Home

The Need for Belonging in Math and Science

The views expressed are those of the author and are not necessarily those of Scientific American.

Email   PrintPrint

From her earliest memories, Catherine Good was good at math. By second grade she was performing at the fourth grade level, sometimes even helping the teacher grade other students’ work. She was praised constantly for her “gift”, often overhearing her mother tell anyone who would listen that she was a “sponge” for anything mathematical.

By high school, Good’s identity as a “gifted mathematician” was so heavily tied up with her math abilities that she decided to pursue a math major in college. However, she felt as though she was making the choice more out of obligation than passion:

“selecting mathematics as my major was not as much driven by my intrinsic interest and love of mathematics as by my long history of being labeled, praised, and reinforced for my math skills.”

She felt even greater pressure to pursue math since she was a woman. There appeared to be a dearth of  women in mathematics, so she felt a great burden to increase the female representation and prove that women are capable of achieving in mathematics.

Achieve she did. Good did so well as an undergraduate, that she decided to pursue a Ph.D. in mathematics. Again, she wasn’t driven by the sheer joy, but by other forces:

“My counter-stereotypical achievement, coupled with my belief that those successes were rooted in an innate gift, not only fueled my academic pursuits, but also formed the basis for my academic identity.”

For awhile, Good performed as usual in her graduate program. But then something happened that would change the course of her career: her identity became threatened. As Good puts it, “the identity as a mathematician that I thought was so well-entrenched and established came crashing down, leaving me in a professional crisis.”

Despite her good grades, a flood of self-doubt crept in. She suddenly wondered: Was I simply no longer inspired by the level of rigor and originality necessary for graduate level mathematics? Was it the fact that for the first time in my academic life, I had to work, really work, at my studies?

For the first time, she also questioned whether she was ever “truly” gifted. The belief in the innate nature of math ability is particularly prominent in the mathematics community, which relies heavily on a “talent-driven approach to math.” The mathematics students that are encouraged and nurtured are the ones who appear to produce elegant solutions with ease, presumably due to an underlying natural gift.

Good wondered, Had this culture of talent led me to believe that I had reached the pinnacle of my abilities because I had to now work at my studies? Or was the counter-stereotypical identity as “gifted female mathematician” now responsible for my mathematical-undoing?

Whatever the cause(s), one thing was certain: she no longer felt a sense of belonging in mathematics. As a result, she left mathematics.



As fortune would have it, Good happened to attend a talk by Joshua Aronson, who was a  professor at the same University. Aronson was talking about his research on stereotype threat, a situation in which a person is at risk of confirming a negative stereotype about one’s group. At this talk, something clicked for Good: this was it. Not only did the research resonate with her personal experiences, but she got excited to pursue this line of research. She immediately signed up to do her Ph.D. in social psychology with Aronson, and then did her post-doctoral work with Carol Dweck at Columbia.

Good’s impressive body of research, along with the research of her colleagues, have painted a consistent and important picture: an individual’s sense of belonging matters. The need for belonging is a fundamental human motive. While a sense of connection and acceptance is important for everyone, signals of acceptance may be particularly impactful for socially stigmatized individuals, who are constantly asking themselves: Do I belong?

In an excellent paper from 2007, Gregory Walton and Geoffrey Cohen showed that in academic and professional settings, members of socially stigmatized groups were more uncertain of the quality of their social bonds and more sensitive to issues of social belonging. They called this “belonging uncertainty”, and they found it contributed to racial disparities in achievement.

Belonging uncertainty may also contribute to the underrepresentation of women in science, technology, engineering, and mathematics (STEM) fields. While the percentage of STEM degrees going to women has increased substantially over the past decade, there is still a sizable gender gap in STEM disciplines. In 2003, women earned only 24% of doctoral degrees in mathematics and 17% of doctoral degrees in engineering. STEM fields are overwhelmingly dominated by White men: in 2000, less than 10% of the mathematics faculty at universities were female.

Against this backdrop, Catherine Good, Aneeta Rattan, and Carol Dweck conducted a series of studies to help inform the answer to the question: why might females be less willing to pursue math-based disciplines?

In particular, they tested the idea that a person’s sense of belonging–“one’s personal beliefs that one is an accepted member of an academic community whose presence and contributions are valued”– in math can predict their desire to pursue math. To measure sense of belonging, they administered their Sense of Belonging to Math scale to a group of undergraduates at Columbia, who were already high-achieving in math. The scale included the following five factors:

  • Membership (“I feel like I belong to the math community”)
  • Acceptance (“I feel accepted”)
  • Affect (“I feel comfortable”)
  • Desire to Fade (“I wish I could fade into the background and not be noticed”)
  • Trust (“I trust my instructors to be committed to helping me learn”).

They found that this scale predicted both men’s and women’s intention to pursue math in the future as well as other important math-related variables, such as math anxiety, math confidence, and perceived usefulness of math. These effects remained, even after taking into account prior achievement in math. What’s more, prior achievement in math didn’t predict a sense of belonging– a finding consistent with other research showing that just because a person has high ability doesn’t mean that that person will be intrinsically motivated to pursue the field, or feel a sense of belonging for the domain.

In another study, the researchers followed college students in their calculus course three times during the semester (beginning, midway, and just before their examinations). They found that the more women perceived signals in their environment that math ability is a fixed trait, and the more they perceived negative stereotypes about women in math, the more likely they were to show a drop in a sense of belonging. In turn, this lowered sense of belonging led to a lower desire to pursue math in the future as well as lower math grades.

In contrast, the more women perceived a malleable-ability mindset environment (e.g., received signals that math skills can be cultivated), the more likely they were to maintain their sense of belonging to math even if they perceived negative stereotypes in the environment. The same effects were not found among men, and these effects could not be explained by their sense of belonging at the start of the semester or by their prior ability (e..g, SAT Math scores). These results suggest that the message that math ability can be acquired can protect women from the damage of being in a threatening environment, allowing them to maintain a high sense of belonging in math and intention to pursue math in the future.

Although prior studies have shown that explicitly telling people that they can increase their intelligence can go a long way toward reducing their vulnerability to stereotype threat (e.g., Aronson, Fried, & Good, 2002; Good, Aronson, & Inzlicht, 2003), this study shows that supportive implicit messages perceived in the learning environment may be just as powerful in protecting students from the negative effects of stereotype threat.

The study also shows the long-term consequences of being in an environment that is full of negative stereotypes. As the semester wore on, women’s perceptions of negative stereotypes in their environment exerted greater influence over their sense of belonging, and their sense of belonging became increasingly important for their intent to pursue a career in math as well as their math performance.

This obviously has important implications for the STEM movement.  As the researchers note,

“Females’ lowered sense of belonging– perhaps in response to their perceptions of their learning environments– can make an academic community an uncomfortable, unwelcoming place to be, causing them to drop out of the domain. When the domain is something as fundamental as mathematics, domain avoidance essentially shuts the door to careers in science, engineering, and technology.”

They also point out that these same issues easily apply to members of any group who repeatedly face messages that their group is limited in ability. This includes Black and Latino Americans, as well as students with a specific learning disability.

The data are in: feeling as though you belong in a field, and that the learning environment is accepting, comfortable, and trustworthy, matters quite a lot– not only for people’s motivation to engage in a domain, but also how high they eventually soar.

© 2013 Scott Barry Kaufman, All Rights Reserved

image credit #1:; image credit #2:

Scott Barry Kaufman About the Author: Scott Barry Kaufman is Scientific Director of The Imagination Institute in the Positive Psychology Center at the University of Pennsylvania. Follow on Twitter @sbkaufman.

The views expressed are those of the author and are not necessarily those of Scientific American.

Rights & Permissions

Comments 6 Comments

Add Comment
  1. 1. Arun 1:14 pm 10/21/2013

    Great article. I’d be interested in seeing research done between one’s interest in a career and the cultivation of media/entertainment around them that also fosters that interest.

    For example, I’ve seen tons of doctors, lawyers, CEO’s, etc. in kids cartoon’s/anime that were all women. I can’t however recall ever seeing one that was a programmer, especially given how many women (in the West, I mean) actually used to do programming back in the day (all the Jane Hopper’s out there).

    Link to this
  2. 2. Arun 1:16 pm 10/21/2013

    Sorry, “Grace Hopper”. Not sure where Jane came from, haha…

    Link to this
  3. 3. Moulton 1:43 pm 10/21/2013

    While I’ve always been personally at home with the STEM culture, my impression is that the STEM culture is not at home in American culture. To the extent that I’ve received negative messages, the source of those messages originated from the political culture, not the academic culture.

    Link to this
  4. 4. Melanie Tannenbaum 2:34 pm 10/21/2013

    What a great, comprehensive overview & explanation of a HUGE and important field of research. Thanks for writing this, Scott!

    Link to this
  5. 5. dadster 6:38 am 10/22/2013

    good analysis. But everything said is so uiversal to almost all human transactions, involvements and commitments and not just to the domain of mathematics .Army recruits dont fight for their country but fight for the name of their own limited army unit to which he belongs. An aristocrat moves with confidence in society , her or his confidence derived directly from her / his sense of belonging to the aristocracy. Others with probably much more to fel confident about themselves through their own higher achievements or performnace dont move with the same confidence and elan as an aristocrat in aristocratic circles . Same can be siad of radical groups too . For their group they volutarily and happily lay down their lives powered by their intense sense of beloging . These same radicals wont lay down their lives for their country or for a cause that their group is not interested in. Dawkins “Selfish Gene” amplifies this principle which is the fundamental motive for behaving in particularly loyal ways in a group.
    Now here the relevent point is why most women dont feel that women belong to the group of maths ? and, why most women feel that the comfortable domain of women is love , romance and caring for others ? To my mind , there is one over-riding reason . XX chromosome makes women more amenable to emotions than to reasoning instinctively since they have to bear children , rear infants for which their bodies are softer than those of men , they have to be more loving and caring to facilitate the rearing of vulnerable and fragile helpless infant . hence women feel they “belong ” to the domain of emotions . Now , maths is the domain of pure unadulterated “reason” and “logic ” both of which have little or no space in the domain of emotions. I had often felt that the domains of reason and emotion are orthogonal to each other like the conventional x-axis and y-axis , the orthogonal projections of each being zero units on the other. Hence it is no wonder that women feel strangers to the domain of unadulterated rigid rigourous reason , devoid of all emotions which the domain of maths is and, why men feel more at home with maths , because of the charcteristics of XY chromosomes that men are made up of are not so much subject to emotions as women are. Men find themselves often out of tune and uncomfortable in the domain of emotions . Now, the question is , can we make maths a bit emotional too ? Can we analyse emotions through the medium of maths which is highly quantitative by nature and, not qualitative intrinsically ? In order to bring any phenomena into the realm of maths that phenomena has to be quantified and be rendered measurable . Measurement involves designing a suitable “unit of measurement” for the phenomena that has to be measured. Can we design such a unit of measurement for measuring emotions , for quantifying the qualities of emotions ? Once emotions too are roped in the domain of maths . it is sure that women might develop a sense of beloging to the domain of maths. Those women who have no problem in being in domains devoid of emotions are indeed comfortable with maths but the majority of women are not like that. Women are fighting shoulder to shoulder with men in the armed forces but only some women, not the majority.Normally women would rather walk the ramp in a fashion parade than be in the forward trenches fighting enemy. Its more or less the same with being in the forward trenches of the domain of pure reason and logic and philosophy . Whereas in arts , in music, in dance and drama women shine and have a collective sense of belonging as they are then in the forward trenches of emotions . Make maths emotional to give women a sense of belonging in the domain of maths .

    Link to this
  6. 6. bucketofsquid 6:12 pm 10/23/2013

    @dadster – You are a walking stereotype. There is no need to make math more emotional. The need is to make men more secure in their identity so they stop acting like petty dictators. Every man I’ve ever met has been very emotional. Most of them just lie about it. And the whole fighting in the trenches thing is silly. No one with a functioning brain wants to be in combat. I know of very few combat veterans that want to go back into combat. Most do because of the unit loyalty or personal loyalty to the individuals they serve with. Very few will prefer combat to being in a parade or fashion show (I haven’t found any actually but that is hardly more than anecdotal).

    Women used to only wear dresses or skirts until it became common for them to wear pants. Things change when people make them change. More men stay at home with the kids than ever before. More women work than ever before. The trend is to have American STEM workers be more women than in the past. This is offset by the large number of STEM workers from other countries where Chauvinism is alive and well.

    Link to this

Add a Comment
You must sign in or register as a member to submit a comment.

More from Scientific American


Get All-Access Digital + Print >


Email this Article