The variability of generation rates for trips to various subclasses of manufacturing land is tested to determine whether a significant increase in the precision with which such trips can be estimated is achieved by treating each subclass as a separate category in contrast to dealing with manufacturing land as a single class. Separate regression equations are developed relating trips to manufacturing land at the two-digit standard industrial classification /sic/ level as a function of /1/ employment /estimated by first work trips/, /2/ floor space, and /3/ truck trips for 21 cats traffic districts in Chicago. On the basis of these equations, covariance analysis is used to test the utility of subdividing manufacturing land into subclasses for estimating trip attractions. It is concluded for the set of data used in the analysis that' /1/ when employment is used as the basis for estimating trip attractions to manufacturing land, no significant improvement in precision can be derived by using a finer /two-digit sic/ classification system, and /2/ when floor area or truck trips are used as the independent variable, significant improvement in the precision of estimates can be obtained by going to the two-digit level. /author/.
Abstract