Upward and downward bias when measuring inequality of opportunity

We show that, when measuring inequality of opportunity with survey data, scholars incur two types of biases. A well-known downward-bias, due to partial observability of circumstances that affect individual outcome, and an upward bias, which depends on the econometric method used and the quality of the available data. We suggest a simple criterion to balance between the two sources of bias based on cross validation. An empirical application, based on 26 European countries, shows the usefulness of our method.