A structural equation approach to measuring technological change: An application to Southeastern U.S. Agriculture

The study demonstrates a method of obtaining consistent estimates of technological change. This was accomplished by extending Theil's differential input demand system into a Multiple Indicator, Multiple Cause (MIMIC) structural equation latent variable model which explicitly accounts for the fact that technological progress is not directly observed and therefore is invariably measured with error. Results obtained for the agricultural sector of the Southeastern states suggest that technological change has been an important determinant of agricultural production and is bias towards the use of capital and against the use of land and labor.This research was partially supported by the McKethan-Matherly Eminent Scholar Chair, graduate school of Business Administration, University of Florida.