I'm not sure it is that simple. May be college/university education is a better predictor for Strava usage. But given that education in the US correlates with b

I'm not sure it is that simple. May be college/university education is a better predictor for Strava usage. But given that education in the US correlates with both race and income it is hard to say without controlling other variables.