Failing in STEM is not the end of the world

Last Friday, I participated in TEDxGeorgiaStateU as a speaker, along with fellow SIF Krisna Patel who talked about the incredible potential of the 3D Atlanta Project. You can read Krisna’s blog here. Our talks will not be out on YouTube for several months now so I thought I would post the text of my talk here. (As a side note, I messed up during the presentations and skipped a few points in my talk so this version is more complete than the video version will be.)

Here is my talk, which was tentatively titled “Failing in STEM.”

When I was an undergrad, I was the only woman in my physics program. I was afraid to ask questions because I thought my lack of knowledge would reflect poorly on all women. What I didn’t realize then is that the other students in my class had gaps in their knowledge too. It was normal not to know everything yet. Why would we take classes if not to learn what we didn’t already know?

Now that I am an instructor, I see this mindset in my own students. Many of them are anxious about answering my questions or asking their own. Many students are reluctant to be vulnerable by admitting that they don’t know everything. If students are afraid to make mistakes and take risks, what does that say about science education and the culture of science? What does it say about us if junior scientists think that mistakes are unacceptable in science?

In STEM, which stands for Science, Technology, Engineering, and Mathematics, it is important to ask these questions in the context of underrepresented groups. People from underrepresented groups often believe that stereotypes about their group’s low achievement in science and math are real – that there isn’t really anything they can do to overcome these stereotypes. This is known as stereotype threat, a well-documented phenomenon that is made worse by making mistakes that should otherwise be a normal part of the learning process. People from groups that are underrepresented in science are especially susceptible to internalizing failures and mistakes, seeing them as something that they caused. Just like when I was the only woman in my physics class and afraid to ask questions, people who are underrepresented in science fear that their mistakes will reflect poorly on their entire group rather than just on them alone.

Something that we in science do not hear enough is that failure is a necessary part of discovery. From students who are just starting out to the most accomplished professional scientist, the culture of science does not allow people to acknowledge their failures. The ironic thing is that the scientific method depends on failure. The scientific method relies on the fact that the process of inquiry itself is fallible – that we must constantly refine and improve our ideas about how the world works.

In science education, we are trying to understand how to teach students that making mistakes is okay.

A study published earlier this year looked at three groups of high school science students from low-income areas of the Bronx and Harlem. One group read about the wonderful, unbelievable accomplishments of three world-famous scientists Albert Einstein, Marie Curie, and Michael Faraday. A second group read about the scientists’ personal struggles, like the discrimination that Curie faced as a female scientist or the persecution that Einstein faced as a Jew in Nazi Germany. The third group read about the scientists’ intellectual struggles, such as Curie’s repeatedly failed experiments at one point in her career.

The study showed that students who read about the personal or intellectual struggles of the famous scientists performed significantly better in their science classes. Not only that, but the students who were initially the lowest achievers were the ones who made the biggest improvements. The students who had only heard about the amazing accomplishments, did not see an improvement in their science grades. In fact, their grades suffered.

Highlighting the failures and struggles of great scientists makes students realize that scientists are human beings. It helps them realize that great scientists are not some special type of ascendant genius to whom they will never measure up.

How students engage in science is tied to whether they see themselves as “doers of science.” People who see themselves as “doers of science” are said to have a science identity, meaning that doing science is an important part of their sense of self. A strong science identity is one of the most important predictors of performance in science class and of having a future career as a scientist. If students think that they are destined to be bad at science and math due to internalized stereotypes, then they do not form a science identity and their performance in science suffers as a result.

So many times I have heard students say “I’m not a science person.” or “I’m just bad at math.” Many of you may have even said these things about yourselves. But many studies have repeatedly suggested that there is no such thing as a “math person” or a “science person.”

Research shows that simply teaching students about malleable intelligence, or the notion that intelligence is not a fixed quantity and is changeable with hard work, vastly improves performance in math and science classes across all levels, especially at the lower initial levels of achievement. Consider that for a moment: simply letting students know that they have the capacity to change their performance makes them perform better in math and science classes. In most science classrooms, however, we are not sending the message that it is okay to fail, make mistakes, or admit when you don’t know something. We teachers need to show students that they can trust themselves to make mistakes.

Changing science education is a start but it is not enough. We also need to shift the culture of science. We need to talk about the reality of being a scientist instead of the idealized vision of what a scientist is and does. We need to stop putting genius on a pedestal. This means talking about the drudgery of doing everyday science and the setbacks of scientists.

Last year, I failed an important exam in my graduate program, which left me feeling completely ashamed. I told my friend Nicole the news and she talked me down, telling me that what I was going through was normal, that failure is not uncommon. I told her I believed her. At that time, I was lying – I couldn’t internalize what she was saying.

Nicole recognized that my experience was anything but unique among scientists. So she started a conversation on Twitter under the hashtag #FailingInSTEM. Nicole tweeted that we need to let our young people know that regular, fallible people do science and that they make mistakes everyday. #FailingInSTEM went viral and stories started pouring in from scientists at all levels about their past scientific failures.

A few scientists talked about failing big exams, which immediately made me feel a little bit better. One scientist said that she spent 30 minutes at a telescope coming up with blank pictures and trying to figure out what was wrong with the parts. It turns out that the dome of the telescope was closed.

Another tweet talked about how a group of scientists were analyzing some space rocks when they found some organic material, or signs of life, Really exciting stuff! But it turned out to be from the bag the rocks were being held in.


Many of those who shared their stories of failure are incredibly successful scientists. Their failures did not ruin their careers; in fact, their failures made them better at doing science.

Destroying lab equipment, failing tests, or completely ruining your experiments – all of that is okay in the long run. No one has all the answers when they first start out. I am simultaneously trying to instill this message into my students and into myself – to build resiliency and openness into our culture. If I hadn’t failed my big exam, I never would have shifted my focus in my graduate program to something that interests me more. If scientists didn’t have failures or make mistakes, they would never know what works and what doesn’t. We would never make progress.

We need to rebrand failure as something productive because it is so important to discovery. Let’s keep the conversation about failure going in our classrooms and in our research labs and online so we can improve the culture of science for everyone.

Leave a Reply