But that doesn’t mean that there’s nothing to see here, so let’s move along.

A fairly consistent observation is that good programmers (for some definition of “good”) had an early start. They played with computers and programming long before they were exposed to computers in school.

There are good programmers who didn’t have an early start, but overall there is a very significant correlation between “good” and “early start.”

However, that being true does not mean that giving a child an early start will make them a good programmer. Lots of kids with access to computers from an early age become good at playing video games, or of complaining to their parents “not this again.”

Although there is a correlation between an early start and proficiency, there isn’t necessarily a strong causal relationship.

So, are we wasting our time giving computers to children? Should we intone “Correlation does not equal causation” in the bored voice of the social cynic? Should we “Move along, nothing to see here?”

No. Although correlation does not equal causation, there is actionable value in the correlation. Consider the possibility that some third factor, “X” is correlated with the potential for becoming a good programmer.

the x-factor

Some kids have lots of X, some an average amount, and some very little.

Many of the good programmers we observe have lots of X and had access to computers at an early age. Of course, we don’t observe the kids with low X and access to computers, they’re too busy doing other things to talk to us about programming.

But there is an interesting group of children who have high X but who don’t have access to computers. Giving them access removes a roadblock to them achieving their potential.

If we are correct about this “X factor,” the correlation between being good at programming and early access need not be causal. It could be that when we give away 100 computers, only five children become good programmers, the five out of a hundred with lots of “X.” But those five children might not have become good programmers without our support.

Thus, even though early access to a computer might not cause someone to be a good programmer, not having access prevents it. Providing computers is thus a question of removing roadblocks.

We have seen this dynamic elsewhere. In education, for example. Sending the nation’s children to school to become literate did not make everyone a poet or a novelist or an advertising copywriter. But it did remove the roadblocks preventing many otherwise disadvantaged children from a shot at careers requiring literacy.

Correlation does not equal causation. But it points the way towards roadblocks that we should remove.

if you’re interested in helping children have access to computers, Codestarter could use your help.