This paper, contrary to the title, focuses on county and municipal governments. According to the authors, it “. . . explores a provocative question; Are computerized systems at the local government level bringing about new services and more equitable distribution of service benefits and costs, or are they more likely to be instruments of political elites, reinforcing the traditional distribution of values and services?” After analyzing the responses of over 500 city and county computing managers to a mail questionnaire, the authors conclude that computers are more exploitive than service-oriented.
The question may be provocative and the conclusion might be accurate, but the analysis leaves much to be desired. The authors use a four-fold category system based on the dimensions of service (direct support and administrative support) and control (high and low). An inventory of county and municipal computer applications becomes a National Portfolio when the applications are distributed by category. This National Portfolio is inappropriately used as a base of comparison. An unbiased selection of computer applications by a county or municipal government will yield a distribution similar to that of the National Portfolio only if each application is equally useful. Since that assumption is clearly inappropriate, all comparisons between the National Portfolio and the county/municipal distribution of applications are meaningless. Such comparisons form the bulk of the analytic data.
The authors are excited to find the number of police functions that have been computerized. They make no mention of the federal and state dollars that have been available to implement such applications. It may be interesting that federal and state governments make dollars available for this governmental function in preference to others, but it is not surprising that county and municipal computer applications are strongly influenced by the presence of such dollars.
None of the data demonstrates that political elites have anything to do with the types of computer applications implemented. That would require considerable more data of a totally different kind.
The question is interesting, but the analysis is seriously flawed, and the conclusions are unwarranted.