Hostname: page-component-7b7567c8c-m6dxt Total loading time: 0 Render date: 2026-11-19T05:53:03.401Z Has data issue: false hasContentIssue false

Alternative methods for dealing by nonnormality and heteroscedasticity in paleontological data

Published online by Cambridge University Press:  20 May 2016

Steven J. Hageman*
Affiliation:
Department of Geology, University of Illinois, Urbana 61801

Abstract

Although numerical methods are highly useful in paleontological studies, potential problems arise with application of parametric statistical methods to paleontological data. Most common statistical tests assume data are normally distributed and that multiple populations have equal variances (homoscedasticity). Paleontological data frequently do not satisfy these assumptions, thereby affecting results of tests and potentially misleading scientific interpretations. Nonparametric tests should be used when assumptions of parametric tests are violated. Normal scores tests, which utilize expected normal deviates (rankits) substituted for original data, are the most powerful nonparametric tests. Despite their potential utility, normal scores tests have received little attention, primarily because of difficulties encountered with rankit conversion.

Recent advances in microcomputer technology provide viable methods for rankit conversion, thus making normal scores tests accessible for routine application. Normal scores tests provide a practical method of dealing with nonnormality and heteroscedasticity common in paleontological data.

Information

Type
Research Article
Copyright
Copyright © The Paleontological Society 

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Article purchase

Temporarily unavailable