ABSTRACT
The growth and decline of organizational forms is affected not only by changes in the numbers of organizations, but also by changes in their size. Any findings about what causes organizations to change in size are steps towards a more complete organizational ecology. Whetten notes that a critical failing of past research into organizational size has been a failure to model size dynamically. The goal of this analysis is to determine what causes hotel chains to grow larger, or decrease in size. Autocorrelation, or serial correlation of disturbances across periods, is a common problem in time-series analysis. Typically, autocorrelation causes least squares to be inefficient, although unbiased and consistent. Heteroscedasticity often occurs because errors vary with some measure of the size of the observation. Large observations may be associated with greater variance in the error. Among the population-level control variables, Aggregate Chain Units and the number of Independent Hotels were not significant.
