Statistically comparing user satisfaction instruments: an application of the bootstrap using a spreadsheet

Testing and comparing the effectiveness of measures of emerging technologies is a challenge to information systems researchers. One reason for this challenge is that, in the social sciences, the classical approach to constructing confidence intervals and testing for significance has relied on the assumption of distribution normality. Nonparametric techniques (which do not require such assumptions) have traditionally been difficult to implement. However, with the development and widespread use of the electronic spreadsheet, researchers are now able to easily and powerfully implement nonparametric techniques. This paper describes a spreadsheet application of one such technique: bootstrapping. Here, distributions of the R² statistic from two user satisfaction instruments were first simulated and then statistically compared