2004
DOI: 10.1001/archpsyc.61.3.310
|View full text |Cite
|
Sign up to set email alerts
|

Move Over ANOVA

Abstract: Repeated-measures ANOVAs continue to be used widely for the analysis of repeated-measures data, despite risks to interpretation. Mixed-effects models use all available data, can properly account for correlation between repeated measurements on the same subject, have greater flexibility to model time effects, and can handle missing data more appropriately. Their flexibility makes them the preferred choice for the analysis of repeated-measures data.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
449
0
5

Year Published

2009
2009
2018
2018

Publication Types

Select...
9
1

Relationship

1
9

Authors

Journals

citations
Cited by 1,275 publications
(454 citation statements)
references
References 49 publications
0
449
0
5
Order By: Relevance
“…We applied a mixed-effects model of the analysis of variance (ANOVA) that allowed a flexible dependency structure for the model and did not exclude the participant when a missing value was encountered (Gueorguieva & Krystal, 2004). Separate mixedmodel ANOVAs were calculated for P3a and P3b responses.…”
Section: Discussionmentioning
confidence: 99%
“…We applied a mixed-effects model of the analysis of variance (ANOVA) that allowed a flexible dependency structure for the model and did not exclude the participant when a missing value was encountered (Gueorguieva & Krystal, 2004). Separate mixedmodel ANOVAs were calculated for P3a and P3b responses.…”
Section: Discussionmentioning
confidence: 99%
“…Generally, GLMMs provide great flexibility in choosing covariates and dealing with complex research designs, and often outperform classic techniques in terms of statistical power (Gueorguieva and Krystal 2004).…”
Section: Explanations On the Statistical Methodsmentioning
confidence: 99%
“…Again, choosing a primary outcome measure, a standard practice in clinical trials, is commonly employed based on a plan to use simple analysis of variance models for examining change due to intervention, rather than selecting data analysis strategies that match the complexity of the measurement problem (Gueorguieva and Krystal 2004;Jacobson and Truax 1991). The basic problem is that ''it is difficult to argue that adequate examination of such complex constructs can be captured with a single indicator '' (De Los Reyes and Kazdin 2006, p. 556).…”
Section: A Range Of Possible Changes In Outcomementioning
confidence: 99%