Blind Auditions

Is this how we should conduct interviews?

SFGate's recent article, Gender gap? Tech could take a cue from orchestras, highlights parallels between orchestras in the 1970s and the tech industry.  The similarities are eerie.  In the 1970s, women were widely considered unsuited to performance at the top levels of orchestra.  Women just didn't get music, or didn't want to devote their lives to it, or something....  It was rare to see a woman in an orchestra playing anything but a harp.  In fact, only 5% of the top symphonies' performers were female.  But now the gender gap has shrunk substantially - those orchestras are now 40% female.

How did they do it?  Blind auditions.  Candidates were behind a screen while demonstrating their skills, such that the judging panel had no idea who the candidate was or what they looked like.  Can the tech industry accomplish the same thing with gender-blind interviews?  Sounds simple enough.  Would that work for tech?  Probably not, for two reasons.

1) Interviews tend to be less of a technical exam and more of a real-time interaction between interviewer and interviewee to get a sense of the applicant's problem-solving and communication skills.  To truly mask the candidates gender, you'd have to prohibit the candidate from speaking, which would make communication much more difficult.  At least until we have surrogates or scramble suits that mask our true identity in the workplace while allowing us to communicate and interact naturally, we won't be able to judge a candidates aptitude without learning their gender.

2) It seems that candidates are getting excluded before the interview stage.  Based on my experience and that of many of my friends throughout the industry, we rarely see women interviewing, even adjusting for the gender imbalance in STEM education.  So we probably need to address whatever is blocking the pipeline before radically reforming the hiring process.

So what can we do?  I have a few thoughts:

Mask applicants' gender where we can.  There's no reason to see gender-identifying information at the initial resume screen, nor when looking at writing or code samples.  An organization committed to leveling the playing field should consider masking names and other information that would tend to reveal gender, race, ethnicity, or national origin in the early stages of review.

Convince women to apply. Studies show that women don't apply for jobs at the rate that men do.  Women tend to select out of the application process unless they meet nearly 100% of the job criteria, while men are willing to apply with only 60% of the criteria.  This fact needs to be repeated over and over again until women (myself included) actually come to terms with the fact that we don't need to reject ourselves before we apply.

Bridge the confidence gap in resume writing. Women's resumes tend to be substantially less strong than men's.  That's because we tend to claim less of the credit for our accomplishments and use less-active verbs to describe our roles.  We "help" or "work with" or "assist."  Men "develop" and "lead" and "direct."  This is even true when describing the exact same task.

Ensure interviewers are aware of their own cultural blind spots.  When we do get to the non-gender-blind interview, we need to be aware of the subtle biases we all have. Cultural blind spots inadvertently exclude excellent candidates because they don't have the same hobbies, interests, and backgrounds as the main team.  This excludes not only women, but men who don't fit the exact image of the team - too old, from a different socioeconomic background, of a different race or national origin than the more common in the field.  Companies should take care to train those who play a role in selected candidates to be aware of their own biases.

Those are my thoughts - but I'd love to read yours in the comments.

Science is Natural for Everyone

Someone's been reading this blog.

Well maybe not, but Girls Love Science. We Tell Them Not To. from The Daily Beast echoes the points I've been trying to make - that so many of the attempts to get girls interested in STEM start from a premise that science is unnatural.  The article quotes astrophysicist Katie Mack who says "Gender-based socialization, and messages [ ] that tell girls that science is an unnatural thing for them to do, are incredibly pervasive in our culture."

The article is a spectacular take-down of the idea that science has to be made "girly" to be appealing to girls.  It tells us to toss aside the notion that "[g]irls will love science, [as] the EU has suggested, because Bunsen burners look like lipstick and fiber-optic cables are sort of like powder brushes."

Finally, the article quotes another female astrophysicist, Meg Urry, who notes "Discrimination isn’t a thunderbolt, it isn’t an abrupt slap in the face. It’s the slow drumbeat of being underappreciated, feeling uncomfortable and encountering roadblocks along the path to success."

A+++++ would read again!

Don't Mess Up (Or Keep Trying Until You Succeed)

There are many awesome things about being a woman in a field dominated by men, but one of the less awesome things is the fear of having your failures held against not just yourself but all women.  Piled Higher and Deeper has a great take on this from the perspective of the woman and xkcd takes it on from the perspective of (some) men:

So obviously, the solution if you are a woman is never to make a mistake.  But let's be honest, you will.  Making mistakes is an important part of learning.  When you make that inevitable mistake or two or ten, it's tempting to give up and walk away.  Don't do it.  If you do, you don't have the opportunity to show everyone that you can get it right.  They'll only be able to remember you for the failure and not the triumph.

Tech or Something Else? Why Not Both?

Talking to high school students, one of the most common things I hear is in the form "I really like [STEM field] but I have a passion for [arts/humanities field]."  As in, "I really love science, but I'm really into music" or "I love math, but I really want to study French."  College catalogs and job descriptions conspire to give people the impression that you have to pick just one field, and nothing could be further from the truth.

Technology touches nearly every aspect of society. Everything from advertising to zoo-keeping has benefited from technology.  The innovations that enable game-changing improvements often come from people who have a passion and experience in two or more fields. Combining substantive knowledge in one area with technical know-how creates a very powerful combination.

Have Confidence to Learn by Teaching

JennyxYoung

Lack of confidence is one of the biggest roadblocks to success, especially in STEM fields.  By now, it is well known that Impostor Syndrome, which disproportionately impacts professional women (but affects many men as well), causes the affected to doubt themselves and ultimately under-perform their potential.  Related confidence deficiencies cause women to downplay their successes in performance reviews, not apply for positions unless they are certain they meet 100% of the qualifications, and dwell on any minor shortcoming.  You don't have to get to the workforce for these confidence issues to crop up - they happen in school as well.

In Math is Hard, But So Is Everything Else, I wrote about how it's natural to feel stuck at times.  But how can we get unstuck?  One way is to try to teach someone else the concepts you barely understand.   Doing this requires healthy dose of confidence, but it doesn't require you to actually be an expert. 

Cultural Blind Spots

This excellent article puts into perspective how 'geek culture' can exclude otherwise very capable individuals who do not fit neatly in its bounds. This is one of the forms of subtle bias I commonly refer to - it's not that someone makes a conscious decision to hire only young white men for their tech startup, but it tends to happen anyway.  It's difficult in an interview to identify the kind of spark that makes someone a good engineer - you want to find people with the right mix of technical know-how, creativity, problem-solving skills, and dedication to make great things.  You can test for technical know-how, and references can help determine dedication, but creativity and problem-solving skills are a lot harder to identify.  So it's no surprise that companies come up with questions about superpowers and surviving a zombie apocalypse because they are fun scenarios that can demonstrate creativity and problem-solving skills.  However, these scenarios play off of fiction genres that tend to appeal much more to men than women.  These scenarios aren't relevant to the work in the company (unless the product is related to zombies or superpowers in some way), so tying hiring decisions to these kinds of questions unnecessarily biases hiring against women.

Many women who succeed in heavily male dominated tech fields have adopted the culture - whether out of genuine interest, feigned interest in an attempt to fit in, or cultivating an actual interest upon making an attempt to fit in.  But for many others who have the aptitude and interest in the science and technology of it all, but are unable or unwilling to adapt to the culture, the existence of this culture serves as a barrier to entry.

This is the kind of thing that all of us need to be conscious of - on the hiring side to reflect whether we are inadvertently missing great candidates and on the job-seeking side to make sure we don't get caught in someone else's cultural blind spot.

2^11

Sometimes the best way to learn is to not realize you're learning.  The game 2048 is like that.  The goal is very simple - you combine merge blocks of the same number into one with the combined value.  For example, if you combine two '4' blocks (2^2), you make one '8' block (2^3).  Two '8' blocks make a 16 block (2^4) and so on.  The goal is to create a 2048 block (2^11).  After each move, a block appears randomly somewhere on the board - usually a 2, but sometimes a 4.

This is a very fun and natural way to get comfortable with powers of 2, which is an important concept in computing.  I'm sure the creator of this game did not intend it to be an educational game, but it's just one of the little ways that you can pick up important concepts in unconventional ways.

No, I'm not writing a blog post solely to justify the hours I've wasted playing this game.  Why do you ask?

Targeted Programs and Backlash

The Don't Look Down post explores the message that targeted programs can send to the targets.  In this post, I take on the message that targeted programs send to the non-targeted.  As a reaction to these programs, men can develop resentment and prejudice against women.  Below I discuss this phenomenon, why that belief is a problem for all of us, and what programs I prefer because they don't foster this reaction.

Meeting Our Future Engineers

Last week I had the honor of meeting about a dozen students from my alma mater at a networking event.  They were visiting Washington, DC to get a sense of what engineering is like in the "real world" (if you can call DC the "real world") by spending a day shadowing hosts (including myself) at various government agencies, research labs, and companies.

At the networking reception, I had the opportunity to speak with engineering students in several different programs.  Most of these students are sophomores and are just starting to figure out what they want to do with their lives.  I spoke with one young woman studying mechanical engineering who is interested in designing roller coasters.  One young man studying chemical engineering is trying to decide between going into manufacturing of pharmaceuticals or entering the growing field of natural gas extraction in his home state of Pennsylvania.  One young engineer is on her way to a career defending networks from cyber-attacks, another is looking to use his background to help shape policy.

Talking to these students, I'd have no idea there was a shortage of women engineers.  The group was evenly split between men and women, and all of the students impressed me.  The statistics show we have a lot of work to do still, but it was nice to be in a room of engineers and not be heavily outnumbered.

Pi is for Everyone

Source: Wikipedia

The formulas, jargon, and code in this post may not be the best way to prove that STEM is for everyone, but it's Pi Day, and I think it's good to keep in mind that things we take for granted now were once considered extremely difficult.  According to CNN, it's cool to like pi.

So today, in honor of Pi Day, I decided to refresh my memory and sharpen my programming skills by implementing Viète's Forumla. Viète's formula, developed in 1593, expands on the Archimedes method of approximating pi developed around 250 B.C.  This method relies on computing the circumference of regular polygons with many sides. As you increase the sides of the polygon, it gets closer and closer to being a circle, and thus, the ratio of the perimeter to the "diameter" gets closer to the value of pi.

Math Is Hard - But So Is Everything Else

The first time I swung a golf club, I missed the little white ball entirely (and the second time I made pretty solid contact with the big green ball - ouch!).  Once I started hitting the golf ball reliably, it didn't go very far. Once I started to hit it a bit farther, it didn't go where I wanted.  And once it started going more or less where I wanted, I couldn't figure out how to shape my shots.  And so on.  At each step, I was tempted to give up - unsure if I would ever be good enough to play with friends and colleagues (and frankly, I still am).  But the truth is that I always could get better with some more practice.

Like golf, math takes practice.  "I'm not good at math" is just code for "I don't want to practice anymore." And just like you can get muscle fatigue and reach a point where practicing is counterproductive, you can get brain fatigue as well.  But if you give yourself a chance to recover and come back to it later, maybe try a different approach or work on a different skill, you will be able to get past it.  

I think this video sends a pretty compelling "stick with it" message and I'd encourage anyone who thinks they don't have a brain for math to watch it:

Do Women Hold an Irrational Fear of Lower Grades?

Yesterday, a friend pointed out this new piece at the Washington Post, which exposes an issue that women tend to shy away from subjects where they are likely to struggle - instead sticking to areas where they expect to receive higher grades.  According to studies cited in that article, "women might also value high grades more than men do and sort themselves into fields where grading curves are more lenient."  Female students apparently do in fact receive higher average grades than male students, but female students also drop out of STEM majors at a higher rate than male students.  


This article immediately reminded me of studies that show that praising young children for effort rather than intelligence encourages them to work through problems when they encounter them.  In contrast, praising intelligence or achievement conditions children to do things they know they will succeed at.  For some reason, parents and teachers both tend to praise effort more in boys and achievement more in girls.  The result is that girls are feel encouraged to do things they know they will be good at, while boys are more apt to try something new and risk failure.  I suspect that this conditioning may in part explain the phenomenon discussed in the WaPo piece.


To some extent, students who seek easier classes may be doing the right thing.  In life we are judged on our output and not our effort.  So applying one's skills where they are more suited is not completely unreasonable.  But I see it as a problem if women are self-selecting out of tech fields at a higher rate just for fear of messing up a perfect GPA. GPA's don't matter as much as you think they do.

Making STEM Interesting: Programming Edition

As careers go, being an engineer is pretty great. Engineers tend to work fewer hours than other comparably paid professions, and engineers report a higher job satisfaction than most other professions as well.  But the problem is that while being an engineer isn't that hard, becoming an engineer can be.  Engineering is considered one of the hardest majors at college, and it's even harder without a strong foundation before that.  It's not easy to convince a teenager that it's worth all the extra work now for a payoff in the distant future.

So the payoff needs to be immediate.  And it can be.  In this post, I'll focus on programming.  What really motivates me to write computer programs is the ability to do something useful and/or fun.  In particular, I like to be able to see immediate visual results as proof that I am doing something.

Some examples of relatively easy projects with high visual impact after the jump: