Changing Rural Development Data Needs: Discussion

Abstract

affairs of rural people, Bonnen and Nelson have confronted important challenge. The strengths of the paper originate from the authors' insistence that a successful data base-including the definition of concepts, inquiry, measurement, and tabulation-can be derived only from a sound theoretical understanding of the relevant information system, and from their explicit recognition that social science data tend to become obsolete as a result of changes in technology and institutions and as a consequence of the learning (or adaptive) aspects of the information system itself. Others before them have emphasized that analytical hypothesis and the data for its empirical test must have the same definitional base (Morgenstern, p. 62-64). However, this still needs emphasizing to some economists and sociologists, who too often show relatively little interest in survey design and data collection, preferring, it seems, to devote their time to the sophisticated manipulation and analysis of inappropriate data. The authors argue rather convincingly that the deficiencies of existing rural data are due not to a lack of knowledge about data gathering, or a shortage of funds, but to a lack of an intellectually coherent, widely accepted normative vision of the future of rural This, they contend, is caused by our inability to resolve the social value conflicts inherent in rural development and by the very rapid, technologically induced, changes that are taking place in rural society. To help solve this problem, they offer a unique explanation of the evolution of rural development policy as attempt to remedy imbalances among major social values, together with a more traditional discussion of new technologies and changes in social institutions that have created urbanized rural society. The paper provides interesting conceptual overview of ideal rural information system. Yet, after discussing the increasing heterogeneity of rural society and the lack of agreement on development objectives, Bonnen and Nelson provide no specific suggestions to assist us in the difficult task of achieving a consensus on rural data needs. Thus, they do not move us much closer to understanding the specific dimensions of a practical rural data base. I think this shortcoming is related to two major weaknesses of the paper. First, the authors do not discuss the kinds of key decisions in which information derived from improved rural data would find a use. At first glance, it seems likely that information requirements vary greatly among potential users. Small town mayors, regional planners, federal program administrators, policy officials, and researchers are all interested in rural development data. But Bonnen and Nelson do not say whether there are any common threads linking their data needs together. Without a better understanding of key decisions, of the probability that certain events will occur once a decision is made, and of the payoffs one can expect, there is no way of being certain that improved or new data actually will influence the rural development process and generate a positive return to society. Early in the paper we are told that the nonfarm rural population can no longer be ignoredthat we may have devoted too many resources to collecting data on the commercial farm sector. Yet, it has been estimated that each dollar spent to increase the accuracy of USDA crop production estimates will return at least $100 worth of benefit to society (Hayami and Peterson, p. 128). Can Bonnen and Nelson claim a similar return for additional investment in rural development data? Can anyone demonstrate that a further expenditure of funds on social and demographic statistics, say more comprehensive data on rural housing condiPeter M. Emerson is a principal analyst, Congressional Budget Office, Washington, D.C.