The conceptual and practical problems in using data on income distribution are not widely understood in the profession. As with the World Tables on economic growth, the ready availability of a secondary data-set such as that assembled by Deininger and Squire(1996), does not guarantee that users are aware of its subtleties.