Replacing graph paper with interactive software in modeling and analysis of experimental data

Abstract

The techniques presented in introductory engineering textbooks for modeling and analysis of experimental data nowadays are essentially the same as the ones that were presented over 35 years ago. Considerable potential now exists for dramatic improvements in data correlation and analysis because of the introduction of personal computers along with user-friendly interactive software which performs linear and nonlinear regressions and yields standard statistical results. This article presents some basic statistical concepts which are required to understand the results obtained from a regression package and demonstrates, using an example of vapor pressure correlation the proper technique for modeling and analysis of experimental data. Our experience and student performance has indicated that a 1st-year course for engineering students can effectively introduce students to the correlation and the modeling of experimental data. This capability can be given to students during two lectures and a 1-hour computer laboratory period plus an appropriate assignment, provided that an interactive regression package Polymath or EZfit, for example or a spreadsheet program with multiple linear regression capabilities is available. © 1996 John Wiley & Sons, Inc.