ISSN:
1432-0592
Source:
Springer Online Journal Archives 1860-2000
Topics:
Architecture, Civil Engineering, Surveying
,
Geography
,
Economics
Notes:
Abstract Many studies have explored the size distribution of income, but empirical analysis has usually involved the examination of statistics which, in effect, summarize characteristics of the distribution, or the relation of a statistic (usually average income) to a set of explanatory variables. The present study differs from previous empirical work by relating observations in specified income intervals to a set of explanatory variables. This multiple equation model appears to be an innovation in income distribution studies. Census data on family unit income distributions by county were taken as observations on a set of dependent variables. Each income class interval is associated with a corresponding dependent variable. The dependent variable consists of the proportionate number or frequency of income units measured in percentage terms. For each dependent variable, then, there will be as many observations as counties reported. Each of the dependent variables is related to a given set of independent variables using multivariate analysis. Observations on these variables are the same for all cases; that is, each dependent variable is related to the same set of independent variables. The multivariate approach employed here allows an investigation of the behavior of a coefficient in each income class in turn. A good deal of additional information (relative to work with income class averages and economic development) is thereby generated. The model was estimated for two time points on California data. The paper includes an interpretation of the results for several independent variables including education and location, and also some comparisons of the results over time.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF01283481
