City planning with biased data

Criado Perez¹ shows several examples of poor city planning due to the lack of good data. The core of the problem is that by averaging, for example, travel patterns, one tends to prioritize more homogeneous patterns. If you look at a heatmap in a city, there's a high chance there'll be a concentration of commutes to the center and one may be tempted to conclude public transport must focus on those trips. However, Criado Perez explains that most of the trips women make are related to their care jobs (child, elderly, supermarket) and tend to be tangential instead of radial.

The core problem is, therefore, not having disaggregated data based on gender. Especially in a situation where a group shows much more homogeneous behavior than the rest, it is very easy to overestimate its weight overall.

Tags: #gender-bias #gender-bias-in-cities