1967
DOI: 10.1037/h0020421
|View full text |Cite
|
Sign up to set email alerts
|

Nonparametric Post Hoc Comparisons for Trend.

Abstract: Post hoc procedures using orthogonal polynomials are illustrated for 3 nonparametric tests that are frequently employed by behavioral scientists in the analysis of empirical data. These examples involve null hypotheses which have been rejected by the Kruskal-Wallis 1-way analysis-of-variance test for rank data, by the Friedman 2-way analysis-of-variance test for rank data, and by the Cochran extension of the McNemar test of change for dichotomous data. Finally, the proposed post hoc procedures are compared to … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
81
0
1

Year Published

1972
1972
2017
2017

Publication Types

Select...
5
3

Relationship

0
8

Authors

Journals

citations
Cited by 142 publications
(82 citation statements)
references
References 13 publications
0
81
0
1
Order By: Relevance
“…The multivariate model null hypothesis (25) is less restrictive than the univariate model null hypothesis (24) because F (R) uses a fixed covariance structure (i.e., K(K+1)/12) in the denominator (Marascuilo & McSweeney, 1967), thus implying compound symmetry of the covariance matrix. Thus, the null hypothesis in (24) implies sphericity because it translates to the assumption that the errors are IID[0, σ ε 2 ] for all j and k from the univariate model null hypothesis in (16).…”
Section: Assumptions and Hypotheses For Interaction Tests Performed Omentioning
confidence: 99%
See 2 more Smart Citations
“…The multivariate model null hypothesis (25) is less restrictive than the univariate model null hypothesis (24) because F (R) uses a fixed covariance structure (i.e., K(K+1)/12) in the denominator (Marascuilo & McSweeney, 1967), thus implying compound symmetry of the covariance matrix. Thus, the null hypothesis in (24) implies sphericity because it translates to the assumption that the errors are IID[0, σ ε 2 ] for all j and k from the univariate model null hypothesis in (16).…”
Section: Assumptions and Hypotheses For Interaction Tests Performed Omentioning
confidence: 99%
“…For comparing the J independent groups, a set of pairwise or group combination contrasts would most likely be of interest for defining a. For comparing the K repeated measures either pairwise, polynomial, or trend contrasts would most typically define b (Lix & Keselman, 1996;Marascuilo & McSweeney, 1967). In some cases, it may be desirable to normalize the trend coefficients, b, so that the metric of the repeated measures variable will not ψ change, thus making confidence intervals more interpretable.…”
Section: Multiple Comparison Procedures For Aligned Rank Proceduresmentioning
confidence: 99%
See 1 more Smart Citation
“…Subsequently, multiple pairwise comparisons after Marascuilo and McSweeney (1967) allowed us to check for significant differences between the relative attractiveness of differently loaded food pellets.…”
Section: Statisticsmentioning
confidence: 99%
“…To test the hypothesis 2 (about individual variation in antipredator responses among the microenvironments to the same stimulus), we compared the proportions of 0 and 1 among the three microenvironments for the first (approach) and second (touch) stimuli independently. Here we also used Cochran test, followed by the multiple comparison procedure of Marascuilo and McSweeney (1967). To test the hypothesis 3 (about the individual variation in jumped distance among the microenvironments to the same stimulus), we compared the jumped distances by individuals in the three microenvironments by using an ANCOVA, in which body size were entered as covariate.…”
Section: Intraspecific Analysesmentioning
confidence: 99%