Stereotype Threat – why it matters

Recently I attending the National Council of Women in IT (NCWIT) Summit on Women and IT: practices and ideas to revolutionize computing (I wrote about my session presentation here – Tinkering and STEM – good for girls, good for all.)

The summit kicked off with a wonderful keynote by Joshua Aronson who is an Associate Professor of Psychology and Education at New York University (NYU). Aronson studies stereotypes, self-esteem, motivation, and attitudes. He showed some remarkable research results that showed that when people are reminded of their race or gender in a testing situation where there is a negative stereotype, they do worse on the test.

This is called Stereotype Threat – which he defined as being at risk of confirming, as self-characteristic, a negative stereotype about one’s group. The threat causes anxiety, and all kinds of measurable changes – from the brain to heart rate, and also greatly impacts test results.

Simply putting a box to mark gender, for example, at the front of a math test significantly changed test scores – for both men and women. Compared to a test where gender was not asked for, if gender was asked for at the beginning of a test, boy’s scores went up, girls’ scores went down. If gender was asked at the end, boys’ scores went down, girls’ scores went up.

Seriously, that was the only difference – there was no mention of the stereotype (boys are good at math, girls are not). The only change was the position of the request for gender identification.

The implication that such a casual, seemingly inconsequential reminder of a possible stereotype  had a HUGE impact says that there is much we do not know about testing.

It implies that assessing human knowledge is not that well understood. It also implies that much of what we think we are testing may be a false reflection. It may have a lot more to do with the context of the individual and the environment than a true assessment of learning.

And it’s not just about knowledge either. He shared a study where white college students increased their jump height over several attempts when tested by a white test monitor – but when a black test monitor gave the same tests, the stereotype “white men can’t jump” became real. The racial/gender makeup of the classroom, the test giver, or even the environmental clues can change everything.

This wasn’t just one research study, either. Aronson showed slide after slide of research that perceptions matter, and matter a lot. Some of this research is on his faculty bio page, lots more in the links below, and a good intro to his work is an ASCD Educational Leadership article – The Threat of Stereotype.

This article also has some great suggestions for reducing stereotype threat.

  • Talk about stereotype threat with students. This reduces anxiety that students may feel by acknowledging they are not alone in worrying about these things.
  • Teach students that intelligence and ability is not inborn and that they can work to do better. They are not limited by stereotypes that restrict what they can do. Talk openly about stereotypes and show that they aren’t true.
  • Build a cooperative classroom environment, not a competitive one. “…cooperative classroom structures in which students work interdependently typically produce immediate and dramatic gains in minority students’ grades, test scores, and engagement because such environments reduce competition, distrust, and stereotyping among students.” – The Threat of Stereotype

Further reading:


PS The best thing he said in the keynote was something like, “The number one predictor of academic success is a student’s answer to the question – does your teacher like you?” Would love to find a real quote!

6 Replies to “Stereotype Threat – why it matters”

  1. Wow would love to know more about that. Sounds like a great area for more research too.

  2. That is amazing. I knew that there were problems with testing in general, but I had no idea that such a simple change could make such a huge difference.

  3. Yeah that’s part of what makes teaching and instructional design so difficult. Every little detail makes a difference, from the appearance of some character in software or a book, to the word choice in some test question (see ).

    “The number one predictor of academic success is a student’s answer to the question – does your teacher like you?”

    That’s a bit sad, isn’t it. It probably can be generalized outside of education: the number one predictor of success is, does your boss like you?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.