¹ See, for instance, the recent work by Williams, Oliver P. et al., Suburban Differences and Metropolitan Policies: A Philadelphia Story (Philadelphia: University of Pennsylvania Press, 1965).Google Scholar

² Oliver P. Williams in a recent paper has argued that metropolitan regions are collections of small groups of residents and the economic superstructures necessary to sustain them. Each group is characterized by the choice of a distinctive life style, and because members of the various groups wish to live in congenial environments they tend to be found in similar locations throughout the region. Precisely where they are located is a matter of economics and the remnants of past land uses in the region, but the fact of congeniality is a major cause of similarity in location choice. See “A Framework for Metropolitan Political Analysis,” prepared for the Conference on Comparative Research in Community Politics, held at Athens, Georgia, November 16–19, 1966.

³ See, for instance, Chapin, F. Stuart Jr., and Weiss, Shirley F. (eds.), Urban Growth Dynamics (New York: John Wiley & Sons, 1962)Google Scholar; Thompson, Wilbur R., A Preface to Urban Economics (Baltimore: Johns Hopkins University Press, 1965); the relevant literature is legion.Google Scholar

⁴ More complete studies, using multiple correlation and regression analyses, will appear in forthcoming publications of the City Council Research Project. But see also our earlier report: Eyestone, Robert and Eulau, Heinz, “City Councils and Policy Outcomes: Development Profiles,” in Wilson, James Q. (ed.), City Politics and Public Policy (New York: John Wiley & Sons, 1968).Google Scholar

⁵ Much classificatory activity, in the field of public policy analysis as elsewhere, is a game. Either the inventors of classifications and typologies do not make it clear just what analytical purpose the classification is to serve, or they may even imply that by having a classification they have explained something. We make this point to have it understood that we are not interested in justifying or defending the particular typology of policy development that we have constructed, but in examining its utility in the analysis at hand.

⁶ Since education and public welfare policies are not made at the city level in California, we cannot use expenditures in these areas as measures of policy outcomes.

⁷ The amenities measure is an attempt to tap Williams' and Adrian's concept of amenities. See Williams, Oliver P. and Adrian, Charles R., Four Cities: A Study in Comparative Policy Making (Philadelphia: University of Pennsylvania Press, 1963), pp. 198–225.CrossRefGoogle Scholar

⁸ Expenses by the planning commission include both expenses and outlays, therefore encompassing the range of items from paper supplies to salaries of full-time city planners to special outside studies commissioned by the city planning commission. California State law requires every city to have a planning commission, but this body may be, and frequently is, a standing committee of citizens appointed by the city council and incurring no expenses charged against the city. Therefore, the actual dollar amount spent by the planning commission would seem to be a good indicator of the extent of a city's commitment to the idea of planning as a way to control the environment. General government expenses are used as the percentage base rather than total government expenses in order to make planning definitionally independent of amenity expenditures.

⁹ It is important to keep in mind that while we are using categories reminiscent of such concepts as “traditional,” “transitional” and “modern” used in the literature of comparative politics, our observations cover only a small segment of that part of the historical developmental process usually called “modern.” It is all the more significant that, even within this small part, we can locate cities in clearly different stages of policy development. This suggests that a concept like “modern” disguises a great deal of the variance that more microscopic analysis can reveal. The point is that our stages “correspond” only analytically to similarly conceived stages used in the long-term analysis of national development.

¹⁰ For a more detailed discussion of how the development typology was constructed and cities assigned to a stage or phase of policy development, see Eyestone and Eulau in Wilson, op. cit.

¹¹ This calculation is made as follows: over eight years, each city's annual outcomes could change seven times. This would make for 82×7, or 574 opportunities for all cities. However, as we missed data for the first fiscal year in two cities, we must deduct two opportunities, giving us the 572 figure.

¹² For a discussion of system capabilities, see Almond, Gabriel A., “A Developmental Approach to Political Systems,” World Politics, 17 (January, 1965), 195–203.CrossRefGoogle Scholar

¹³ Assessed valuation includes private houses and property, commercial property and industrial property. From private property a city derives personal property revenues and a portion of state income tax revenues; from commercial property it receives property and sales tax revenues; and from industrial property it gets property tax revenues.

¹⁴ This is not the only question we asked in this connection. For instance, we also asked a great many closed “agree-disagree” questions some of which we used in the earlier analysis, in Wilson, op. cit.

¹⁵ See Lasswell, Harold D. and Kaplan, Abraham, Power and Society: A Framework for Political Inquiry (New Haven: Yale University Press, 1950), p. 25.Google Scholar

¹⁶ This is not the place to discuss the methodological problems and procedures involved in “stepping up” the data from the level of the individual (micro-analysis) to the level of the group (macro-analysis). Suffice it to say that our empirical results justify the viability of the procedures, although we would be the first to admit that many technical problems remain to be solved.

¹⁷ Paul F. Lazarsfeld has written in many places about the variety of “group properties” that need to be distinguished in analysis lest errors of inference be made. See, for instance, Lazarsfeld, Paul F., “Evidence and Inference in Social Research,” in Lerner, Daniel (ed.), Evidence and Inference (New York: The Free Press, 1959), pp. 117–125 Google Scholar; or Lazarsfeld, Paul F. and Menzel, Herbert, “On the Relation between Individual and Collective Properties,” in Etzioni, Amitai, Complex Organizations (New York: Holt, Rinehart and Winston, 1961), pp. 422–440.Google Scholar We are not dealing with the global, structural or relational properties of councils in this analysis.

¹⁸ That is, the absolute number of individual problems named was divided by all problem responses made in a council. The resulting scores, that could range from zero to one, were rank-ordered and divided into the quartile ranges used in the analysis.

¹⁹ That is, the number of problems named was multiplied by the number of respondents and divided by all responses squared. The resulting score was subtracted from one to rank-order the Councils from high to low. The formula then is: 1-NP×NR/r² , where NP = number of problems, NR = number of respondents, and r = number of total responses.

²⁰ The measure of agreement on a single problem is simply the proportion of councilmen among all respondents who mentioned the most frequent problem. For the measure of problem area agreement, the number of responses in the area receiving the most responses was divided by the number of responses in all areas. Five “problem areas” were provided for classification of individual problems: Services and Utilities, Amenities, Promotion and Development, Social and Remedial Problems, and Governmental and Intergovernmental Problems.

²¹ Our measure of salience, as mentioned in the text, was whether a problem or problem area was mentioned by at least three respondents. We shall not try to interpret the proportions obtained for the services and utilities as well as promotion and development areas across the developmental continuum because the results may be an artifact of council size. As five councils in the transitional stage, three in the maturing phase and seven in the advanced stage had more than five members (usually seven), and as no retarded or emergent council had more than five members, clearly any one problem had more of a chance to be named by at least three respondents in the more developed cities. But as, for instance, nine of the advanced councils had only five members, yet all advanced councils are accounted for in naming at least one problem, the council size factor does not seem to have too much of a distorting effect. But we note it as interesting that the more developed a city's policy, the more councils tend to mention problems related to utilities and services and to promotion and development.

²² The dominant set of reasons was simply defined as that set which included the most responses among all sets, regardless of absolute number.

²³ We could argue our case more liberally on statistical grounds and possibly test it if we had more and numerically more diverse legislative bodies available for analysis: the larger a legislative body, the more likely it is that averaged individual preferences will approximate, if not correspond to, the preference of the collectivity.

²⁴ The improvement diversity measure was constructed in the same way as the problem diversity measure. See fn. 18, above.

²⁵ The improvement visibility measure was constructed in the same manner as the problem visibility measure. See fn. 19, above.

²⁶ The improvement agreement measures are the same as those used in connection with problem agreement. See fn. 20, above.

²⁷ We would like to point out here that we had very similar results in the earlier study in which we used a closed agree-disagree scale measuring attitudes concerning the scope of government activity and in which we used individual councilmen as our units of analysis: see Eyestone and Eulau, in Wilson, (ed.), op. cit.

²⁸ An improvement area was assumed to be salient in council preferences if at least three respondents articulated problems in the area.

²⁹ Because an “industrial” future was envisaged in only a handful of councils, we combined this category with the “balanced” category which implies that the council envisages a balance in residential, commercial and industrial development.