Prospect Theory asserts that people display risk attraction in high-probability losses. But our subjects tend to avoid fair risks for large ([euro]30 to [euro]90), high-probability (80%) real losses, vindicating Bernoulli's view that risk aversion is the dominant attitude.

We show how to minimize the probability of misclassifying individuals as being poor or not poor when data on some of their relevant attributes are missing, but an estimate of the population distribution of attributes is available.