4.2Variables and Data

The main variables of the model previously described are output (Y), traditional and IT capital (KNIT and KIT respectively) and labor (L). Different levels of study (national, industries and states) make the construction of the dataset complex.

In this study, I consider pooled cross-section data for 52 industries, for 50 contiguous U.S. states and the District of Columbia, for the 21 years between 1977 and 1997 inclusive. The industries account for all of the nonagricultural nongovernmental production in the U.S. economy. Next, I describe the various levels of study considered in this dissertation before discussing the construction and data source for each variable.