Winning at the great game: How to improve student results

Winning at the great game: How to improve student results

The consistently excellent Farnam Street podcast recently featured a marathon episode with the investor and educator Adam Robinson. It was so long it had to be split into 2 parts, but it’s worth the investment nonetheless (although I’ve never heard someone use the term ‘air quotes’ as much as he does)

Towards the end of episode 1 was a story about education that really grabbed my attention. 

You can listen to the full episode here – skip to 1hr 37mins.

Here’s my abridged version, with a few additional thoughts:

The value of drive & communication

Robinson was involved in the creation of the Stuyvesant test, for access to the renowned New York school. The test was hard – and more competitive than Harvard’s.

The school’s population inevitably leaned towards students from affluent backgrounds who could afford private tutoring.

Robinson offered to create a new program to aim to remedy this and level the playing field. 

But after reviewing the results of the program, he was shocked and disappointed to see that when going from comparable starting scores, White and Asian students improved far more than Black and Hispanic students.

After creating a huge questionnaire for the following year’s students covering topics from family setup to psychology and self-esteem, the results came back 1 year later.

The number one factor for improvement?

A student with a father born in another country, and their mother born in the US.

These students improved twice as much as the reverse (father born in US, mother born in another country), who in turn improved twice as much as students with both parents born in another country. These students improved a little more than those with both parents both in the US.

And then there was the kicker: those students with fathers born in another country and mothers born in the US (I.e. The number 1 factor for improvement) improved the same, regardless of race.


Robinson puts it down to two things: the father born in another country instilled an immigrant drive into his children; and the mother’s native language skills enabled her kids to learn, communicate and interact.

It’s also worth noting there’s nothing about vocational skills here; it wasn’t the children of doctors or lawyers or construction workers that did better or worse. These skills are soft skills, human skills – perseverance, drive, empathy, communication.

You probably won’t improve by being confident

Another insight from the questionnaire was that improvement was inversely related to perceived chances of success. 

Many students started from a score of 20 at the beginning of the year, and they needed 65+ to get into the top 3 schools (85 for Stuyvesant). This meant a huge improvement was needed.

As you may expect, the students who improve the most are unlikely to be the ones who completely write themselves off or have no interest, but they’re also not the most confident ones either. 

The students most likely to improve are those who say “I don’t think I’m going to pass the test”. 

They think they have a chance, yet they don’t really think they’ll make it. But their teachers do – they believe in them.

The students who were very confident didn’t show much sign of improvement – no matter what level they started at. More broadly, it turned out academic performance and self-esteem were inversely connected.

This links to our bias of overconfidence and perhaps a little survivorship bias too. The kid who is convinced they’ll get into the NBA probably has no idea how hard it is to actually make it.

This goes against the received wisdom of thinking positive and following our dreams, and opens up the question of how we may best motivate people to improve.

3 ways to improve

Robinson cites three main aspects for students to focus on if they want to improve on any topic. As with many things in life, they’re simple but not easy.

  • Work at something: this means focus, good study materials, a motivation to study whatever the topic is, and know what the end game is
  • Get feedback: The feedback loop is crucial. It gradually builds confidence, prevents students going off track, and allows them to see alternative paths. Most existing education has feedback loops built in, but arguably they’re way too long. Perhaps there’s something in the startup maxim of ‘Build, Measure, Learn’ that applies just as much to education as it does to app development.
  • Clearly see you improved: New education offerings aren’t gamified just to keep you coming back; they’re to help you see the progress you’ve made. And it’s not just software that can do this – the key is in a student being able to see what progress they made from where they were before. This can be verbal, visual, auditory, experiential – whatever is effective.

There’s of course a lot more nuance to all of this – after all, this story is based largely on one questionnaire, in one city.

Yet it’s extremely interesting nonetheless – especially if we start setting our sights on improvement, rather than reaching for one particular metric or standard that’s been set by someone else.

No matter whether for you, your own children, or the students you’re working with, it’s worth considering the following:

  • Recognize that learning is hard
  • Don’t overinflate self-esteem
  • Embrace struggling and persistence

Counter-intuitive perhaps, and certainly misaligned with much of the messaging we’re bombarded with today, but maybe we can all benefit from giving ourselves a little bit of that immigrant drive and maternal language ability, and stay humble and embrace the struggle.

There's something wrong. Great! Check your inbox and click the link to confirm your subscription.