MIGRATION FLOWS IN INTRAURBAN SPACE: PLACE UTILITY CONSIDERATIONS

ABSTRACT

This paper examines the concept of place utility in an operational context. Its major objective is to construct and evaluate place utility functions which are based upon socioeconomic and migration characteristics of areal units within a city. A secondary objective is to identify major factors related to intraurban migration flows in an aggregate framework. The empirical basis for the study is a sample of actual migration flows for Cedar Rapids, Iowa, for a single year commencing in 1966 and terminating in 1967. Regression methods are used for estimating parameters of the place utility functions. One conclusion of the analysis is that migration friction between neighborhood types is considerably less than expected. Factors identified as most important in guiding intraurban migration streams are the aspirations of the household with respect to housing and the spatial characteristics of the vacancy market, the latter of which has been neglected in geographic research. With respect to the method of calibrating place utility functions, we were not satisfied with the regression approach. Although the implications of those functions in terms of migration processes are in concordance with existing theory and knowledge, it was felt that the amount of explained variation should be higher if such functions are to serve as input components for other analyses.

Notes

1 The support of several individuals and institutions has enabled completion of this report. In particular, the authors wish to acknowledge Mr. Robert Madson of the Cedar Rapids Planning Commission; Mr. Gerald Critz and Mr. Marvin Jewell of the Cedar Rapids office of Northwestern Bell Telephone Company; Dr. Harold McConnell of the Department of Geography, University of Iowa; the seventeen members of Dr. McConnell's field techniques class in the Spring semester of 1967; the University of Iowa Computer Center; and the University of Iowa Department of Geography, where this research was carried out. We also wish to express our appreciation to Dr. Eric G. Moore of Northwestern University, to Dr. Kenneth Fairbain of the University of Melbourne.

2 J. Wolpert, “Behavioral Aspects of the Decision to Migrate,”Papers of the Regional Science Association, Vol. 15 (1965), pp. 159 69.

3 In our other work to date focus has been upon the construction of normative models of migration which have an identifiable place utility component. One such model has been partially tested on the basis of migration flows in Cedar Rapids, Iowa for the year 1966 to 1967; see L. A. Brown and D. B. Longbrake, “On the Implementation of Place Utility and Related Concepts: The Intra-Urban Migration Case,” in K. R. Cox and R. G. Golledge (Eds.), Behavioral Problems in Geography: A Symposium (Evanston, Illinois: Northwestern University Studies in Geography, 1969);L. A. Brown, F. E. Horton, and R. I. Wittick, “Place Utility and the Normative Allocation of Intra-Urban Migrants,”Demography, Vol. 7 (June 1970).

4 Two main avenues of research characterize intraurban migration research in recent years: 1) Survey studies designed to establish the relevance of particular variables to movement behavior, usually within a sociological or sociopsychological context; and 2) the development of general models of residential site selection which apply to ecological units within the urban area and which utilize variables acting as surrogates for behavior at a micro-scale. As indicated, the research reported here is designed to fit within the framework of the latter approach rather than the former. Examples of survey studies designed to establish the relevance of particular variables to intraurban migration behavior, usually within a sociological or sociopsychological context include J. B. Lansing and G. Hendricks, Automobile Ownership and Residential Density (Ann Arbor, Michigan: Institute for Social Research, 1967); H. L. Ross, “Reasons for Moves to and from a Central City,”Social Forces, Vol. 40 (1962), pp. 261 63; and P. H. Rossi, Why Families Move: A Study in the Social Psychology of Urban Residential Mobility (New York: The Free Press, 1955). Examples concerned with the development of general models of residential site selection which apply to ecological units within the urban area and which utilize variables acting as surrogates for behavior at a micro-scale include F. S. Chapin, Jr., and S. F. Weiss, Factors Influencing Land Development (Chapel Hill: Institute for Research in Social Science, University of North Carolina, 1962); R. H. Ellis, “A Behavioral Residential Model,” (Evanston, Illinois: Technical Report, The Transportation Center, Northwestern University, 1966); T. G. Donnelly, F. S. Chapin, Jr., and S. F. Weiss, A Probabilistic Model for Residential Growth (Chapel Hill: Institute for Research in Social Science, University of North Carolina, 1964); J. D. Herbert and B. W. Stevens“A Model of the Distribution of Residential Activity in Urban Areas,”Journal of Regional Science, Vol. 2 (1960), pp. 21 36; J. F. Kain, “Journey to Work and Residential Location,”Papers of the Regional Science Association, Vol. 9 (1962), pp. 137 60; P. LaRouche, “The Simulation of Residential Land Use Growth in the Montreal Region” (New Haven: Unpublished Master of City Planning Thesis, Yale University); I. S. Lowry, “A Model of Metropolis,” (Santa Monica, California: Report Number RM-4035-RC, The Rand Corporation); and E. G. Moore, “Models of Migration and the Intra-Urban Case,”The Australian and New Zealand Journal of Sociology, Vol. 2 (1966), pp. 16 37.

5 Wolpert, op. cit., footnote 2, p. 162.

6 J. W. Simmons, “Changing Residence in the City: A Review of Intra-Urban Mobility,”Geographical Review, Vol. 58 (1968), pp. 622 51.

7 It should be noted that a household's expectations with respect to a residence site are by definition constrained by its ability to attain a particular type of residence site. Thus, household expectations and its threshold reference point are tempered by considerations such as the cost of a residence vis-a-vis household income and the availability of residences of a particular type.

8 Wolpert, op. cit., footnote 2, p. 162, introduced the term “threshold reference point.” For a complete discussion of behavioral options available to the household experiencing dissatisfaction or strain with respect to its experienced place utility, see L. A. Brown and E. G. Moore, “The Intra-Urban Migration Process: A Perspective,”Geografiska Annaler, Series B, Vol. 52 (1970). That paper presents a detailed conceptual model of the migration decision in a behavioral framework, and includes a wide range of ideas relevant to the study reported here.

9 For discussions of specific variables to describe environment (and aspirations) see Lansing and Hendricks, op. cit., footnote 4, pp. 102–44; Rossi, op. cit., footnote 4; Simmons, op. cit., footnote 6. In general, such variables can be grouped into five basic categories related to accessibility or proximity, physical characteristics of neighborhood, service and facilities, social environment, individual site and dwelling.

10 The concept of passive and active migrants was suggested by T. Hagerstrand, “Migration and Area,” in D. Hannerberg, T. Hagerstrand, and B. Odeving (Eds.), Migration In Sweden: A Symposium (Lund, Sweden: Gleerup, 1957), pp. 134–35. This concept distinguishes migrants who actively seek migration opportunities (without regard to the migration activities of their acquaintances) from those who tend to follow in the footsteps of acquaintances and relatives. We believe, however, that the concept is better viewed in terms of the effectiveness of information sources, whereas passive migrants depend upon personal information sources. A parallel distinction is made by Rogers in describing the adopter categories: Innovators, early majority, early adopters, late adopters, and laggards; E. M. Rogers, Diffusion of Innovations (New York: The Free Press, 1962).

11 A relevant theoretical discussion of psychological utility functions and their measurement can be found in R. D. Luce and P. Suppes, “Preference, Utility, and Subjective Probability,” in R. D. Luce, R. R. Bush, and E. Galanter (Eds.), Handbook of Mathematical Psychology (New York: John Wiley, 1965), Vol. 3, pp. 249–410.

12 Wolpert, op. cit., footnote 2, p. 161.

13 Sources for this were the 1960 United States Census block statistics, a 1964 Cedar Rapids Planning Commission special census, and the Cedar Rapids Traffic Survey conducted in 1964 by the State of Iowa Commission on Highways. These sources represent the available data which were most applicable to our sample of migrations. Despite the disparateness of these data sources, members of the Cedar Rapids Planning Commission, who worked closely with us, believed the sources were compatible and reflective of the urban structure in 1966–1967, the year of our migration sample. Data were tabulated by origin-destination zone, a unit established by the highway commission for traffic survey analysis. Similar units were used by the Cedar Rapids Planning Commission in their special census, and block statistics census information was also converted to these units. The sixty-four O-D zones chosen for analysis represent the built-up area of Cedar Rapids.

14 This appears to be a usual procedure in factorial ecology analyses; see, for example, B. J. L. Berry and R. A. Murdie. Socio-Economic Correlates of Housing Condition (Toronto: Urban Renewal Study, Metropolitan Toronto Planning Board, 1965).

15 Northwestern Bell Telephone Company directories were used for this purpose.

16 This employs the IBM scientific subroutine package routine FACTO, modified to output component scores by Theodore Miller of the Department of Geography, University of Iowa. For principal components analysis, see H. H. Harmon, Modern Factor Analysis (Chicago: The University of Chicago Press, 1967).

17 For linkage analysis, see L. L. McQuitty, “Elementary Linkage Analysis for Isolating Orthogonal and Oblique Types and Typal Relevancies,”Educational and Psychological Measurement, Vol. 17 (1957), pp. 207 29; for the Q-mode components analysis see R. R. Sokal and P. H. Sneath, Principles of Numerical Taxonomy (San Francisco: Freeman, 1963); for the Ward algorithm, see J. H. Ward, “Hierarchical Groupings to Optimize an Objective Function,”American Statistical Association Journal, Vol. 58 (1963), pp. 236 44. The hierarchical grouping.

18 This analysis was carried out with the STEPRE program of the Computer Center of the University of Iowa.

19 Rossi, op. cit., footnote 4, p. 6.

20 Rossi, op. cit., footnote 4, p. 175.

21 Rossi, op. cit., footnote 4, p. 44.

22 Rossi, op. cit., footnote 4, p. 69.

23 Rossi, op. cit., footnote 4, p. 6.

24 J. M. Beshers and E. O. Laumann“Social Distance: A Network Approach,”American Sociological Review, Vol. 32 (1967), pp. 225 36; for social distance concepts and other approaches to measurement of social distance, see E. O. Laumann, Prestige and Association in an Urban Community (New York: Bobbs-Merrill, 1966); for the concept of mean first passage time and Markov chains, see J. G. Kemeny and J. L. Snell, Finite Markov Chains (Princeton: D. Van Nostrand, 1960).

25 A more general discussion of the use of this approach for geographic research can be found in L. A. Brown and F. E. Horton, “Functional Distance: An Operational Approach,”Geographic Analysis, Vol. 2 June 1970).

26 Simmons, op. cit., footnote 6, p. 633, presents a table of migration flows between five types of social class areas taken from the work of S. Goldstein and K. B. Mayer, “Metropolitanization and Population Change in Rhode Island” (Publication Number 3, Planning Division, Rhode Island Development Council, 1961). It would be interesting to compute Markov chain properties for that matrix and compare their findings with ours, taking into account differences in the classification schemes. For techniques for comparing Markov chain transition matrices, see T. W. Anderson, “Probability Models for Analyzing Time Changes in Attitudes,” in P. F. Lazarsfeld (Ed.), Mathematical Thinking in the Social Sciences (New York: The Free Press, 1954), pp. 17–66.

27 Rossi, op. cit., footnote 4, p. 179.

28 Simmons, op. cit., footnote 6, pp. 637–40, in his review of intraurban migration research, provides several valuable references to studies concerned with the housing market. With regard to land development in particular, the interested reader may consult the following: Chapin and Weiss, op. cit., footnote 4; E. J. Kaiser, “Location Decision Factors in a Producer Model of Residential Development,”Land Economics, Vol. 44 (1968), pp. 351 62; E. J. Kaiser, A Producer Model for Residential Growth (Chapel Hill: Institute for Research in Social Science, University of North Carolina, 1968); and S. F. Weiss, J. E. Smith, E. J. Kaiser, and K. B. Kenney, Residential Developer Decisions (Chapel Hill: Institute for Research in Social Science, University of North Carolina, 1966).

29 For a general view of such research see Lansing and Hendricks, op. cit., footnote 4; Rossi, op. cit., footnote 4; or Simmons, op. cit., footnote 6.

30 For social area analysis, see W. Bell, “Social Areas: Typology of Urban Neighborhoods,” in M. B. Sussman, (Ed.), Community Structure and Analysis (New York: Thomas Crowell, 1959) pp. 61–92; for social area analysis in a spatial context, see R. A. Murdie, Factorial Ecology of Metropolitan Toronto, 1951–1961: An Essay on the Social Geography of the City (Chicago: Department of Geography Research Paper No. 116, University of Chicago, 1968); and B. J. L. Berry and P. H. Rees, “The Factorial Ecology of Calcutta,”American Journal of Sociology, Vol. 74 (1969), pp. 445 91.

31 The use of ecological correlation is discussed in H. M. Blalock, Causal Inferences in Non-Experimental Research (Chapel Hill: University of North Carolina Press, 1964); L. A. Goodman, “Ecological Regression and Behavior of Individuals,”American Sociological Review, Vol. 18 (1953), pp. 663 64; and W. S. Robinson, “Ecological Correlations and the Behavior of Individuals,”American Sociological Review, Vol. 15 (1950), pp. 351 57. Early migration research formulated within an ecological frame-work include T. E. Sullenger, “A Study in Intra-Urban Mobility,”Sociology and Social Research, Vol. 17 (1932), pp. 16 24, and H. W. Zorbaugh, The Gold Coast and the Slum (Chicago: University of Chicago Press, 1929).

32 An interesting study, complementary to ours, is E. W. Butler, F. S. Chapin, Jr., G. C. Hemmens, E. J. Kaiser, M. A. Stegman, and S. F. Weiss, Moving Behavior and Residential Choice: A National Survey (Chapel Hill: Institute for Research in Social Science, University of North Carolina, 1968).

33 See, for example, Kain, op. cit., footnote 4, and Lowry, op. cit., footnote 4.

34 This point is corroborated by Simmons, op. cit., footnote 6, drawing upon the work of Rossi, op. cit., footnote 4, and of H. S. Shryock, “Urban, Rural, and Other Types of Residence,” in H. S. Shryock (Ed.), Population Mobility Within The United States (Chicago: University of Chicago Press, 1964).

35 Tables 8 and 9 present preliminary findings from an interview based study in Cedar Rapids, Iowa, which was undertaken after the research reported here. Classification of interviewees into socioeconomic groupings employed the criteria characterizing O-D zone types reported in this paper.

36 E. G. Moore, “Residential Mobility in an Urban Context” (Unpublished Doctoral Dissertation, University of Queensland, Australia, 1966), and Rossi, op. cit., footnote 4, p. 6.