Which is more tragic: Timmy, a young boy so food-deprived that he resembles a skeleton and can hardly move, or the fact that 34 million children suffer from malnutrition. While the latter clearly describes a larger problem, people are more likely to give assistance when the former is described to them. This is because of the identifiable victim effect; we take more action to help smaller, more acutely described groups of people. The reason for this is that our morality is far from rational. People give aid based on sympathy and guilt, feelings that our aroused more by individual descriptions than statistics. This bias is problematic in the context of charitable research distribution. Often, the most efficient ways of helping people will be passed on in favor of causes that are more effective at arousing sympathy.
In 1987, a small child named Jessica was trapped in a well. Americans donated 700,000 dollars to help save Jessica. While the donations were beneficial, and should not be condemned, $700,000 could clearly be spent far more effectively. However, while assuaging a famine in a poor country would be more effective, it doesn’t create as much sympathy as Jessica.