Abstract
This package implements the resampling and bootstrapping approach to statistical inference described recently in this journal by Cirincione and Gurrieri. 1 Resam pling uses the brute force of computer power to establish confidence intervals for various statistical procedures, based on the researcher's own data set rather than on generic statistical tables based on formal mathematical theory. With this package, one can substitute computer- intensive Monte Carlo data simulation techniques for a wide range of parametric and nonparametric statistical tests. Resampling Stats is utilitarian in design, requires writing program code in its English-like built-in language, and lacks some of the niceties of full-fledged Windows programs from industry leaders. Nonetheless, Resampling Stats involves only a modest learning curve and the rewards to the social scientist are great. Arguably, the resampling approach to statistical inference, which Resampling Stats imple ments, is superior to conventional methods from both a research and an instructional viewpoint.
Get full access to this article
View all access options for this article.
