Affiliation:
1. Virginia Polytechnic Institute and State University Blacksburg, Virginia
Abstract
A procedure is described for calculating empirical models based on the combined data collected in a series of sequential experiments. A hypothetical example involving ten independent variables is used to illustrate these procedures. Issues related to the comparability of data from several experiments, the use of both continuous and discrete variables in the model, the specification of additional data needed to calculate a second-order polynomial, and the procedure for determining the optimal combination of variables are discussed. Necessary rules for combining data from several studies are suggested. Mathematical criteria of efficiency, variance of coefficients, and orthogonality are used to evaluate this procedure.
Cited by
2 articles.
订阅此论文施引文献
订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. Data Bridging: Statistical Methods for Integrating Results across Studies;Proceedings of the Human Factors Society Annual Meeting;1991-09
2. Quantitative Guidelines for Telephone Information Systems;Proceedings of the Human Factors Society Annual Meeting;1991-09