Search PubMed⌕ Search
Home/Archive/PubMed

PubMed · 11414585

Drawbacks to integer scoring for ordered categorical data.

A Ivanova·V W Berger
View original source ↗Export metadata ↓

Abstract

Linear rank tests are widely used when testing for independence against stochastic order in a 2 x J contingency table with two treatments and J ordered outcome levels. For this purpose, numerical scores are assigned, possibly by default, to the J outcome levels. When the choice of scores is not apparent, integer (equally spaced) scores are often assigned. We show that this practice generally leads to unnecessarily conservative tests. The use of slightly perturbed scores will result in a less conservative and uniformly more powerful test.

Explore related subjects

BiometryHumansModels, StatisticalProbabilityReproducibility of ResultsStochastic ProcessesSurvival AnalysisJournal Article

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

A Ivanova, V W Berger. 2001. Drawbacks to integer scoring for ordered categorical data.. https://doi.org/10.1111/j.0006-341x.2001.00567.x

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2002

Intensive short courses in biostatistics for fellows and physicians.

At both of our universities we teach (with colleagues) introductory courses in statistics for fellows and physicians. We do not expect that those taking these courses will be able to do their own statistical work, but rather the intention is for them to 'learn the language' and to facilitate future collaboration. Basic principles of study design are introduced in the courses, as well as some of the most common statistical procedures. We will discuss the factors (what works and what does not) that may contribute to a successful course, a comparison to other courses, and our self-evaluation strategy. Finally, we will cover the financial arrangements that we have made when teaching these courses.

Biometry↗
PubMed2002

Likelihood methods for measuring statistical evidence.

Focused on interpreting data as statistical evidence, the evidential paradigm uses likelihood ratios to measure the strength of statistical evidence. Under this paradigm, re-examination of accumulating evidence is encouraged because (i) the likelihood ratio, unlike a p-value, is unaffected by the number of examinations and (ii) the probability of observing strong misleading evidence is naturally low, even for study designs that re-examine the data with each new observation. Further, the controllable probabilities of observing misleading and weak evidence provide assurance that the study design is reliable without affecting the strength of statistical evidence in the data. This paper illustrates the ideas and methods associated with using likelihood ratios to measure statistical evidence. It contains a comprehensive introduction to the evidential paradigm, including an overview of how to quantify the probability of observing misleading evidence for various study designs. The University Group Diabetes Program (UGDP), a classic and still controversial multi-centred clinical trial, is used as an illustrative example. Some of the original UGDP results, and subsequent re-analyses, are presented for comparison purposes.

Biometry↗