The evaluation of the impact of publicly supplied services has been an elusive goal of social scientists ever since the beginning of quantitative research. The reason behind encountered difficulties is straightforward: the direct costs of publicly supplied services can be readily measured but the benefits derived from those services cannot normally be directly ascertained. The latter problem is frequently true even where services are priced to the consumer as the administered price may not approximate the scarcity value of the service. Economists trained in neoclassical theory have approached the problem from a supply and demand framework. Supply and demand equations are estimated and then the supply function is shifted reflecting technological change caused by the publicly supplied services. Neoclassical theory utilizes this information to assess the social benefits derived from the services. The measurement of social benefits is approximate, to be sure, and the partial equilibrium environment of the studies leaves much to be desired; nevertheless, these studies have permitted economists to establish rough estimates of the economic or desirability of the publicly supplied services. David Leonard in his book, Reaching the Peasant Farmer, has abandoned the methodology developed by economists and attempted to develop new measures of the impact of publicly supplied services. The service he is attempting to evaluate is the agricultural extension service in the Western Province of Kenya. In this reviewer's opinion, Leonard's methodological approach is a mixed blessing. The study provides much more institutional detail and insight than is typically gleaned from economists' studies; yet the reader finally puts the book aside not knowing whether the costs of supplying extension services to the Western Province are equal to, less than, or greater than the social benefits attributable to those services. Moreover, it is not clear from the research whether the individual has benefited from the acceptance of essentially free services-the author has merely asserted that this is true. Consequently, the myriad of recommendations made by the author, based on organizational theory, may or may not make sense when cast in the perspective of the total economy. What lies behind this severe criticism of a painstaking work? Leonard eschews the economists'methodology because ...cost/benefit ratios are appropriate to study at levels of analysis and with types of questions with which we are not dealing (p. 25). Instead he argues that, A different set of problems would arise if one were to attempt to measure the productivity of the extension agent by interviewing farmers regarding how much impact the agents had had on their (p. 25). One would have no quarrel in conducting the analysis at this level, in fact this is a necessary first step to determine whether advice offered to farmers makes economic sense within the context of their farming systems. Leonard, however, asserts that the advice will generally be a beneficial thing for the farmer (p. 27) and develops the following alternative measures of extension agent output: (1) agricultural knowledge of the extension agent, (2) the extension agent's explanatory ability, (3) innovativeness, (4) average number of farm visits per day by extension agents--what Leonard calls Visit Effort, and (5) progressiveness skew which is an attempt to determine whether large farmers benefit more than small farmers from extension agents' visits. None of the above output indicators directly measures the impact on farmers, and the third output indicator relating to innovativeness might be perverse in actual farming situations. Leonard made no attempt to assess the profitability of modifications to standard recommendations issued by the