Until recently, this was all one session, but gradually the number of topics grew until it could not fit into one session no matter how much we rushed. We have not yet determined the optimal layout for the session split.
I attempt to go through the entirety of this session without ever mentioning any specific real stereotype or bias, in part to avoid triggering stereotype threat and in part to avoid tripping over some students’ personal biases. See also Teach, don’t Preach.
Stereotype roughly means
schemata we don’t like
Note: in social sciences, it is comparatively easy to determine a phenomenon exists, comparatively hard to defend any explanation as to why… but here is one postulate.
high-school calculusstyle study2:
explanation: selection bias.
To get into the study population, have to be performing at a given level.Stereotypically bad group performing at that level user STT situation; remove the threat and see their underling ability was higher
Warning: myth-busting is bad
I know you’ve heard people with small feet are bad at swimming, but it’s not true. Studies show that people with small feet are just as good at swimming as people with large feet.
To the degree possible we will avoid giving stereotypes airtime. There are many stereotypes about who is good and bad at CS; I assume you each know some of them (if not, hooray! let’s leave it that way). We’ll attempt to discuss things that work on all stereotypes without mentioning specifics.
I also assume, a priori, that you agree we should reduce the impact of stereotype threat and give all students an even playing field. If you are one of the people that instead want some group to be impeded, don’t let us see that while you are a TA or we’ll fire you.
Think-pair-share: what can a TA do to reduce stereotype threat or what might a reasonable TA do to increase it that we can avoid?
In the resulting discussion, or after it, cover the following:
Avoid using names and pronouns in examples
In English, names suggest ethnicity and pronouns suggest gender. There are ethnic and gender stereotypes about CS.
Person A and
Person B and don’t give them pronouns.
Don’t bring up groups. Never say even well-intentioned things like
group X’s average on the exam was higher than the average!
Avoid stereotype-triggering asides. Don’t refer to things that are common knowledge for people like you but not for other groups (like sports, TV shows, games, politics, history).
Hard to do
Giving context helps (i.e., release the grade distribution so students can place themselves relative to their peers)
But don’t give group-specific context
Reducing sense of
Don’t deny others’ identity (insulting)
Don’t ignore current events and issues – you might unwittingly raise some topic that has significant weight for some students
Computer Scientist, etc.
I am a computer scientistmindset
Note: some people will experience
imposture syndrome even if they are world-renowned luminaries in their field, always feeling like they do not belong and their true identity is
person who is faking it as an X. More detailed discussion of this phenomenon is out of the scope of this class, but know that not feeling you belong is common and not a sign of inadequacy or lack of belonging.
Success in CS is all all work talent |------------------------|
What dictates success in CS? (point to
Is it number of hours of work alone, and someone who has invested 101 hours is better than someone who has invested 100? (point to
Is it innate cognitive ability alone, and only people with big CSy brains can succeed? I know this is overly simplistic view, but I want each of you to come up and make a tick-mark on this scale.
(step out of sight while they do so, then return. My experience is the larger the class, the less variance there is and the more answers cluster around ¾ work; the smaller the class, the more variance and the more the averge slides towards ½ work. It’s not a study, it’s a pedagogical tool.)
(mostly to acknowledge that the activity was not a fair study, I ask)
How many of you lied because you knew other people were watching?
What’s the right answer? How would we know?
There are well-respected extremists who assert things like
less than 2% of people can ever be taught to design a good algorithm and
after years of study I have been unable to determine if such a thing as talent actually exists (note: I had references for these quotes when I started using them, but I have lost the citations since then; feel free to ignore this step if you can’t find good sourceable quotes)
The basic argument (nature vs nurture) has been going on for centuries
And the question is somewhat poorly put (e.g., better would be perpendicular axises and ask
draw the B+ curve)
Easier to correlate5 how a person answers a question like this and other attributes of the person.
Work in this space pioneered by Carol Dweck, expanded by others since; goes under the name of
all work a
Growth Mindset and
all talent a
Fixed Mindset – guess which one they like…
people like me lack talent
people like me have not tried; since I am trying, that’s not about me
See how much you’ve learned?
You are smart
You’ll be fine
But what if the truth is talent matters more?
There is some evidence it doesn’t, but…
Your job as educators is whatever percentage of truth is work; focus only on that
Example hidden pre-req in CS: spatial reasoning
Spatial reasoning = ability to manipulate shapes in head
(Note: there are other, more detailed approaches to this information; I use the following in interest of tying it into what we’ve already covered, even though it leave out some portion of the information)
Your brain has built stereotype-aligned schemata, even for stereotypes you do not believe in
the young one had more energy
the old one was wise
the tall one with a deep voice was trustworthy
Neck Tiesupposed to be
male? They avoid things that obviously conflicted, but problem remains…)
I have a strong implicit bias against Xis often heard as
I think X is bad
avoid giving stereotypes airtimesuggestion under the Interventions section above
average input of schemata
I know I’m implicitly biased against X, so even though I don’t like it lets give it another look.
This is a grab-bag section of various other best-practices that do not fit naturally into the above presentation.
This section, unlike most others, is not based on social science studies but rather on own my personal observations; caveat emptor.
— Luther Tychonievich
Humor is good: it reduces stress, builds a sense of belonging, etc.
But humor is dangerous.
|not get joke||
everyone else is laughing; I don’t fit in
I have found making fun of one’s self is pretty universal: even if they don’t find it funny, virtually everyone at least understands the goal.
When giving feedback on deficiencies in a student’s performance, always
Express there are high standards: we demand a lot, and this is hard.
Express confidence in the students, preferably worded personally: I know you will be able to improve and accomplish this.
Don’t lie; if they need more improvement than deadlines will admit, express they can improve now and over time gain full competency, even if the standards are so high that that competency is unlikely to arrive before the end of the course.
This topic has been studied for several years6 and has come to be called
There are many; see http://reducingstereotypethreat.org/bibliography.html for more than you can realistically read↩
See, e.g.: Good, C., Aronson, J., & Harder, J. (2008). Problems in the pipeline: Stereotype threat and women’s achievement in high-level math courses. Journal of Applied Developmental Psychology, 29(1), 17–28↩
See, e.g.: Stone, J., Lynch, C. I., Sjomeling, M., & Darley, J. M. (1999). Stereotype threat effects on black and white athletic performance. Journal of Personality and Social Psychology, 77(6), 1213–1227↩
See, e.g.: Aaronson, J., Lustina, M., Good, C., Keough, K., Steele, C., Brown, J. (1998). When White Men Can’t Do Math: Necessary and Sufficient Factors in Stereotype Threat. Journal of Experimental Social Psychology, 53(1), 29–46↩
See, e.g.: Good, C., Aronson, J., Inzlicht, M. (2003) Improving adolescents’ standardized test performance: An intervention to reduce the effects of stereotype threat. Journal of Applied Developmental Psychology, 24, 645–662↩
See, e.g.: Cohen, G. L.; Steele, C. M.; Ross, L. D. (1999)
The mentor’s dilema: Providing critical feedback across the racial divide. Personality and Social Psychology Bulletin, 25(10), pp. 1302–1318↩
See. e.g.: Yeager, D. S.; Purdie-Vaughns, V.; Garica, J.; Apfel, N.; Brzustoski, P. Master, A.; Hessert, W. T.; Williams, M. E. (2013)
Breaking the cycle of mistrust: Wise interventions to provide critical feedback across the racial divide. Journal of Experimental Psychology: General, 143, pp. 804–824↩