Psychology without p values: Data analysis at the turn of the 19th Century.

The history of data-handling practices can remind psychologists that there are many ways to overcome the current uniformity of statistical practice.

Abstract

Although the fledgling psychology of 100 years ago was assertively empirical, there were no inferential statistics to guide psychologists' data analyses. However, 19th-century developments had left psychology with a rich array of techniques for analyzing and presenting data, some of which remain underutilized today. These include comparisons across replications, within-subject designs, reanalysis of data, analyses of factorial designs, and especially the use of tables and graphs. As the merits of hypothesis-testing statistics are debated at the turn of the 21st century, the history of data-handling practices can remind psychologists that there are many ways to overcome the current uniformity of statistical practice.