A regression method for analysing ordinal data from intervention trials

Schnell, Daniel; Magee, Elvin; Sheridan, John R.

doi:10.1002/sim.4780141104

Cited by 8 publications

(4 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In general, it is suggested to use nonparametric statistics to analyze changes in ordinal outcome variables. Examples of such nonparametric approaches are ridit analysis [33][34][35] and the so-called "rank-invariant non-parametric method" [36,37]. Another possibility is to use complicated statistical modeling [38,39].…”

Section: Analysis Of Ordinal Outcome Variablesmentioning

confidence: 99%

Evaluation of the results of a randomized controlled trial: how to define changes between baseline and follow-up

Twisk

Proper

2004

Journal of Clinical Epidemiology

198

129

View full text Add to dashboard Cite

Section: Analysis Of Ordinal Outcome Variablesmentioning

confidence: 99%

Evaluation of the results of a randomized controlled trial: how to define changes between baseline and follow-up

Twisk

Proper

2004

Journal of Clinical Epidemiology

198

129

View full text Add to dashboard Cite

“…Finally, our interest lies in comparing location between two treatments. If interest lies elsewhere, for example in comparing the relative frequencies or cumulative probabilities in the ordered categories between treatments, other techniques such as ridit analysis [31] or the proportional odds model [16] would be more appropriate.…”

Section: Discussionmentioning

confidence: 99%

Robustness and power of analysis of covariance applied to ordinal scaled data as arising in randomized controlled trials

Sullivan

D’Agostino

2003

Statistics in Medicine

View full text Add to dashboard Cite

In clinical trials comparing two treatments, ordinal scales of three, four or five points are often used to assess severity, both prior to and after treatment. Analysis of covariance is an attractive technique, however, the data clearly violate the normality assumption and in the presence of small samples, and large sample theory may not apply. The robustness and power of various versions of parametric analysis of covariance applied to small samples of ordinal scaled data are investigated through computer simulation. Subjects are randomized to one of two competing treatments and the pre-treatment, or baseline, assessment is used as the covariate. We compare two parametric analysis of covariance tests that vary according to the treatment of the homogeneity of regressions slopes and the two independent samples t-test on difference scores. Under the null hypothesis of no difference in adjusted treatment means, we estimated actual significance levels by comparing observed test statistics to appropriate critical values from the F- and t-distributions for nominal significance levels of 0.10, 0.05, 0.02 and 0.01. We estimated power by similar comparisons under various alternative hypotheses. The model which assumes homogeneous slopes and the t-test on difference scores were robust in the presence of three, four and five point ordinal scales. The hierarchical approach which first tests for homogeneity of regression slopes and then fits separate slopes if there is significant non-homogeneity produced significance levels that exceeded the nominal levels especially when the sample sizes were small. The model which assumes homogeneous regression slopes produced the highest power among competing tests for all of the configurations investigated. The t-test on difference scores also produced good power in the presence of small samples.

show abstract

“…Throughout this paper ridit effects (relative to an identified distribution, e.g. see Bross, 1958, Schnell et al, 1995 are discussed that offer a solution to both questions and show how the rank approach is related to the linear models.…”

Section: Introductionmentioning

confidence: 99%

A Global View on Parametric and Nonparametric Approaches to the Analysis of Ordered Categorical Data

Munzel

Langer

2004

Biometrical J

View full text Add to dashboard Cite

SummaryRank approaches are very common in the analysis of ordered categorical data but can only be interpreted on an experiment-wise level. Therefore, parametric tests from linear models, although based on metric structures, are used frequently to analyze this type of data. So the questions arise 1. what parametric tests measure in this context and 2. whether the rank approach could be modified to achieve a global level of interpretation. A possible solution to question 2. offers the so called ridit approach, which is based on known reference distributions. In this paper we discuss a global view that shows how rank analysis and ridit analyses are related and how parametric procedures fit into the same framework. The use of the uniform distribution as a reference in the ridit approach gives an explanation to question 1. The asymptotic multivariate normality of the effect estimators is shown and robust test statistics are discussed. Type I and type II error rates are examined in simulation studies and the approach is applied to a toxicological example.

show abstract

A regression method for analysing ordinal data from intervention trials

Cited by 8 publications

References 9 publications

Evaluation of the results of a randomized controlled trial: how to define changes between baseline and follow-up

Evaluation of the results of a randomized controlled trial: how to define changes between baseline and follow-up

Robustness and power of analysis of covariance applied to ordinal scaled data as arising in randomized controlled trials

A Global View on Parametric and Nonparametric Approaches to the Analysis of Ordered Categorical Data

Contact Info

Product

Resources

About