We re-examine the notion of consumer rationality in transition economies by modifying the conventional non-parametric tests of the axioms of revealed preference. To reflect the unique characteristics of transition economies, shortages are allowed in the state market, while not in the private market. Because of the way the modified tests are constructed, data for prices in both the state and private markets are needed, as are data on purchases from state markets. Through some convenient algebraic substitutions, the tests do not require data on actual quantities purchased in the private markets. This feature of the modified tests is especially important, given the near impossibility of gaining access to data on sales or purchases from private markets in transition economies. Using limited aggregate consumption data for four important food items in five cities of the former Soviet Union, we clearly demonstrate that conventional nonparametric tests can lead to an erroneous rejection of the rationality hypothesis of consumer behavior for economies in transition.