How can we create programming to most effectively engage young girls in STEM?

119 2

Obstacles to Girls in STEM:

Many people know about the fact above, but many don’t understand what exactly causes the gender barrier in STEM fields. Here are four of the most pervasive obstacles to girls in STEM:



Here are some other important piece of knowledge to have before we tackle the issue:

  • Around third grade many girls lose confidence in math (link)
  • Teacher and parents often underestimate girls abilities in math, biases contributing to as much as half of the gender gap in math (link)
  • Parent talk more about spatial relationships with boy than girls (link)
  • Girls self-assess lower than boys from young ages
  • Teachers grade girls harsher than boys (link)

The Project:
Using game theory and the research above as my guide, I aim to find the most effective way for the Fem ‘n STEM club to allocate their time so that they have the greatest possible impact on the younger girls.

Fem ‘n STEM Club
The Fem ‘n STEM Club is a group of high school girls dedicated to engaging and empower younger girls in STEM through interactive labs, mentorship programs, and other engaging projects.

Using my research in this field and my four years of experience in this club, I have selected four (4) viable strategies for engaging girls in STEM. I will then compare their costs and benefits to find the optimal use of the club’s resources.

Now, to best compare the strategies I will assign them each a value for the children and a cost to the club. The value to the children will also be an approximation, calculated from the qualitative extent to which they work against the obstacles. The cost will be in terms of hours and labor, but will be a rough approximation.

The club has about 10 active participants and 4 leaders. Since we have a 1 hour club period about once a month from September to May, I would estimate that we can demand no more than 90 hours from them collectively. There is much more range in what can be expected from a leader, but for the sake of argument let’s say they each dedicate 3 hours a month to the club, making their total contribution 3*9*4 = 108. This means that in total the club has about 198 hours of labor to spend on activities.

Strategy #1BenefitsCosts

Bi-monthly group science activities led by high school girls

Goals that this strategy works toward effectively:
> Male-dominated culture
> Few role models
> “Math anxiety”

We have seen great success with this piece of programming, as it forms positive relationships with STEM and role models. Formative memories are made, science rooms become a place of laughter, friendship, and mystery. While these affinity groups are a powerful and important part of engaging girls in STEM, they cannot be the only part. I estimate the value of this strategy to be 90.

It takes three leaders approximately 2 hours to plan, 2 hours for the club to prep materials and make signs, and 1 hour to actually see the girls. I have increased the cost value slightly because these projects often also require physical materials. 

This strategy would involve doing one of these sessions twice a month for 6 months. Therefore, the total cost would be: 5 * 12 = 60

Strategy #2BenefitsCosts

One on one mentoring programs with continued communication

Goals that it works toward effectively:
> Few role models
> “Math anxiety”
> Gender stereotypes

This is a strategy that is currently rather underused by the club. In theory, we would pair each of the elementary school girls with a high school female mentor. This partnership would not only provide the young girl with a role model, but could help dissolve “math anxiety” and tailor support to each girl. I would value this strategy at 96.

Because this is a highly individual process, it requires immense dedication from our high school club members. Not only would they have to continually initiate meetings and communication, but they would also have to make sure that the meetings were effective. It is hard to approximate the cost of this, but assuming that we would have a mentor for each of the roughly 25 3rd grade girls at my school and that they would meet once a month for 7 months, I would estimate the cost to be around 25*7 = 175.

Strategy #3BenefitsCosts

Talking with parents

Goals that it works toward effectively:
> “Math anxiety”
> Gender stereotypes

We have never used this strategy before in Fem ‘n STEM, but it struck me in my research in a gap in our reach. Parents have an incredibly important role in shaping a child’s view of the world and their place in it, and according to the AAUW’s findings, many parents are unwittingly putting their daughters at a disadvantage in STEM, particularly math. We could either meet with parents, or simply send their children home with information, about “math anxiety”, the importance of framing things in terms of spatial relations, emphasizing a growth mindset at home, and more. I think that this program has a value of 40, because while it has the potential to trickle down to the students, it is highly dependant on the parents changing after a single piece of communication.

Because we have never done this before, it is particularly challenging for me to try and estimate its cost. Let’s assume that the only difference in talking synchronously with the parents and simply sending them information is the 3 hours it would take to plan the event (emails to the administration, the parents, and club leaders) and the 1 hour of the event itself. That means that no matter the format, it would still take about 5 hours to prep (doing research, putting together pamphlets/slideshows, etc.). Splitting the difference of the two formats, the cost of this strategy would be 7.

Strategy #4BenefitsCosts

A role model/mentor of the week for the girls to learn about

Goals that it works toward effectively:
> Gender stereotypes
> Male-dominated culture
> Fewer Role Models

This idea came to me when reading about a study that found that an increased number of female role models in stem could cut the gender barrier in innovation in half. This idea is relatively simple, we would simply create some medium of information (likely a video or pamphlet) to distribute to the 3rd-grade girls through their technology teacher. The role models could be anyone from Mae Jemison to a classmate of mine that just presented her research in a national competition. I would value this strategy at 60

Because, much like the talking with parents strategy, this strategy does not involve a lot of scheduling or interdepartmental communication, it is significantly cheaper than the first two on the list. Let’s say that we have 30 active weeks of the club and that it takes 1 hour to create each of the pamphlets, which makes the total cost of this strategy 30. For the sake of this project, we will ignore the fact that these pamphlets would be reusable from year to year. 

As a club, we need to decide how to allocate our time in a specific order. Therefore, my analysis of this game will be a movement tree. In the end, I will see which of the outcomes provides the greatest value while being below our maxim cost of 198.

If you look at the diagram above, you will see that the optimal solution with a cost below 198 is:
> Don’t do strategies #3
> Do strategies #1, #2 and #4

You may note that that leaves us with many additional hours of labor, which suggests to me that we should strive to find more low-cost, medium-benefit strategies to supplement this course of action.

I’d love to hear your feedback on my project, the club, next steps, or anything else! Please fill out the form below:


Lavy, Victor, and Edith Sand. “On the Origins of Gender Human Capital Gaps: Short and Long Term Consequences of Teachers’ Stereotypical Biases.” NBER Working Papers Series, 2015,

Pruden, Shannon M., and Susan C. Levine. “Parents’ Spatial Language Mediates a Sex Difference in Preschoolers’ Spatial-Language Use.” Psychological Science, vol. 28, no. 11, Nov. 2017, pp. 1583–1596, doi:10.1177/0956797617711968

Robinson-Cimpian, J. P. et al. “Teachers’ perceptions of students’ mathematics proficiency may exacerbate early gender gaps in achievement.” Developmental psychology 50 4 (2014): 1262-81.

Sarah T. Lubienski, et al. “Girls’ and Boys’ Mathematics Achievement, Affect, and Experiences: Findings from ECLS-K.” Journal for Research in Mathematics Education, vol. 44, no. 4, 2013, pp. 634–645. JSTOR, Accessed 18 Apr. 2021.

“The STEM Gap: Women and Girls in Science, Technology, Engineering and Math – AAUW: Empowering Women Since 1881.” AAUW, 5 Oct. 2020, 



  1. Hi Julia! I love your project and topic. It is something that I feel is extremely important and needs to be addressed immediately, so thank you for doing so. The club sounds awesome, too. All of the strategies you mentioned are great ideas. I think strategy #1, bi-monthy group science activities led by high school girls, is particularly good. You pretty much get the best of both worlds using that strategy, in my opinion. You get the mentorship from high school girls as well as experience with science activities. Best of luck to you and the Fem n’ Stem club!

  2. Julia, I really enjoyed reading your research, analysis, and potential strategies to use as part of Fem ‘N’ STEM! I was amazed by the way you could calculate costs and benefits numerically to arrive at ideas about which strategies would be most effective to try implementing. I hope you’ll leave this research with younger leaders in the club when you graduate so that not only they but also the middle schoolers you work with can benefit from it. I can’t wait for my daughter to get to work with role models like you in a few years!

Leave a Reply