More Powerful Tests of Simple Interaction Contrasts in the Two-Way Factorial Design

Gregory R. Hancock, Daniel McNeish

Research output: Contribution to journalArticle

Abstract

For the two-way factorial design in analysis of variance, the current article explicates and compares three methods for controlling the Type I error rate for all possible simple interaction contrasts following a statistically significant interaction, including a proposed modification to the Bonferroni procedure that increases the power of statistical tests for deconstructing interaction effects when they are of primary substantive interest. Results indicate the general superiority of the modified Bonferroni procedure over Scheffé and Roy-type procedures, where the Bonferroni and Scheffé procedures have been modified to accommodate the logical implications of a false omnibus interaction null hypothesis. An applied example is provided and considerations for applied researchers are offered.

Original languageEnglish (US)
Pages (from-to)24-35
Number of pages12
JournalJournal of Experimental Education
Volume85
Issue number1
DOIs
StatePublished - Jan 2 2017
Externally publishedYes

Fingerprint

interaction
statistical test
analysis of variance
Analysis of Variance
Research Personnel

Keywords

  • Analysis of variance
  • factorial designs
  • interactions
  • pairwise interaction contrasts
  • simple interaction contrasts
  • Type I error

ASJC Scopus subject areas

  • Education
  • Developmental and Educational Psychology

Cite this

More Powerful Tests of Simple Interaction Contrasts in the Two-Way Factorial Design. / Hancock, Gregory R.; McNeish, Daniel.

In: Journal of Experimental Education, Vol. 85, No. 1, 02.01.2017, p. 24-35.

Research output: Contribution to journalArticle

@article{e5a8b5a3291b4492955318fea67bc29c,
title = "More Powerful Tests of Simple Interaction Contrasts in the Two-Way Factorial Design",
abstract = "For the two-way factorial design in analysis of variance, the current article explicates and compares three methods for controlling the Type I error rate for all possible simple interaction contrasts following a statistically significant interaction, including a proposed modification to the Bonferroni procedure that increases the power of statistical tests for deconstructing interaction effects when they are of primary substantive interest. Results indicate the general superiority of the modified Bonferroni procedure over Scheff{\'e} and Roy-type procedures, where the Bonferroni and Scheff{\'e} procedures have been modified to accommodate the logical implications of a false omnibus interaction null hypothesis. An applied example is provided and considerations for applied researchers are offered.",
keywords = "Analysis of variance, factorial designs, interactions, pairwise interaction contrasts, simple interaction contrasts, Type I error",
author = "Hancock, {Gregory R.} and Daniel McNeish",
year = "2017",
month = "1",
day = "2",
doi = "10.1080/00220973.2015.1065220",
language = "English (US)",
volume = "85",
pages = "24--35",
journal = "Journal of Experimental Education",
issn = "0022-0973",
publisher = "Routledge",
number = "1",

}

TY - JOUR

T1 - More Powerful Tests of Simple Interaction Contrasts in the Two-Way Factorial Design

AU - Hancock, Gregory R.

AU - McNeish, Daniel

PY - 2017/1/2

Y1 - 2017/1/2

N2 - For the two-way factorial design in analysis of variance, the current article explicates and compares three methods for controlling the Type I error rate for all possible simple interaction contrasts following a statistically significant interaction, including a proposed modification to the Bonferroni procedure that increases the power of statistical tests for deconstructing interaction effects when they are of primary substantive interest. Results indicate the general superiority of the modified Bonferroni procedure over Scheffé and Roy-type procedures, where the Bonferroni and Scheffé procedures have been modified to accommodate the logical implications of a false omnibus interaction null hypothesis. An applied example is provided and considerations for applied researchers are offered.

AB - For the two-way factorial design in analysis of variance, the current article explicates and compares three methods for controlling the Type I error rate for all possible simple interaction contrasts following a statistically significant interaction, including a proposed modification to the Bonferroni procedure that increases the power of statistical tests for deconstructing interaction effects when they are of primary substantive interest. Results indicate the general superiority of the modified Bonferroni procedure over Scheffé and Roy-type procedures, where the Bonferroni and Scheffé procedures have been modified to accommodate the logical implications of a false omnibus interaction null hypothesis. An applied example is provided and considerations for applied researchers are offered.

KW - Analysis of variance

KW - factorial designs

KW - interactions

KW - pairwise interaction contrasts

KW - simple interaction contrasts

KW - Type I error

UR - http://www.scopus.com/inward/record.url?scp=84949767924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949767924&partnerID=8YFLogxK

U2 - 10.1080/00220973.2015.1065220

DO - 10.1080/00220973.2015.1065220

M3 - Article

AN - SCOPUS:84949767924

VL - 85

SP - 24

EP - 35

JO - Journal of Experimental Education

JF - Journal of Experimental Education

SN - 0022-0973

IS - 1

ER -