Apologies, not sure why that was so garbled first time! - posting again more
simply
I want to use lmer for analysis of response time (& error data) from a
reaction time experiment,
and have a question about specifying the structure of random effects in the
model.
I am using a repeated measures design where 30 subjects judge the laterality
of a hand shown on
a computer screen, completing 18 trials in all combinations of the following
experimental factors
(Angle, 8 levels (hand shown in 8 orientations), Laterality, 2 levels (both
right & left hands shown),
Condition, 3 levels (participant holds own hands in posture a,b or c). With
repeated measures ANOVA
the Error structure is specified as follows:
aov(percentErrors~angle*condition*laterality+Error(subject/(angle*condition*laterality)),data=errorData)
aov(meanRT~angle*condition*laterality+Error(subject/(angle*condition*laterality)),data=RTdata)
where response variables, percentErrors and meanRT, are the percentage of
errors (out of 18 trials) and the mean
reaction time over 18 trials . I need to move to lmer as variance is not
constant across Angle and error data are
bounded (0,1) so I should use the binomial family my first pass (looking
at all main effects & possible interactions) is:
lmer(cbind(numErrors,numCorrect)~angle*condition*laterality+(angle*condition*laterality|subject),family=binomial),data=errorData)
where numErrors+numCorrect = 18 for each subject & Angle-Laterality-Condition
combination
lmer(meanRT~angle*condition*laterality+(angle*condition*laterality|subject),data=RTdata)
I am unsure if this is correct? Help is welcome, thanks - Nuala
----- Original Message -----
From: nuala brady <nuala.br...@ucd.ie>
Date: Wednesday, July 28, 2010 2:09 pm
Subject: [R] Help with specifiying random effects in lmer - psychology
experiment
To: r-help@r-project.org
> <!-- /* Font Definitions */ @font-
> face {font-family:"Cambria Math"; panose-1:2 4 5 3 5
> 4 6 3 2 4; mso-font-charset:0; mso-generic-font-
> family:roman; mso-font-pitch:variable; mso-font-
> signature:-1610611985 1107304683 0 0 159 0;} @font-
> face {font-family:Calibri; panose-1:2 15 5 2 2 2 4 3
> 2 4; mso-font-charset:0; mso-generic-font-
> family:swiss; mso-font-pitch:variable; mso-font-
> signature:-1610611985 1073750139 0 0 159 0;} /* Style
> Definitions */ p.MsoNormal, li.MsoNormal,
> div.MsoNormal {mso-style-unhide:no; mso-style-
> qformat:yes; mso-style-parent:""; margin-
> top:0cm; margin-right:0cm; margin-
> bottom:10.0pt; margin-left:0cm; line-
> height:115%; mso-pagination:widow-orphan; font-
> size:11.0pt; font-family:"Calibri","sans-serif"; mso-
> fareast-font-family:Calibri; mso-bidi-font-family:"Times
> New Roman"; mso-fareast-language:EN-US;}
> .MsoChpDefault {mso-style-type:export-only; mso-
> default-props:yes; font-size:10.0pt; mso-ansi-font-
> size:10.0pt; mso-bidi-font-size:10.0pt; mso-ascii-
> font-family:Calibri; mso-fareast-font-
> family:Calibri; mso-hansi-font-family:Calibri;} @page
> Section1 {size:595.3pt 841.9pt; margin:72.0pt 72.0pt
> 72.0pt 72.0pt; mso-header-margin:35.4pt; mso-footer-
> margin:35.4pt; mso-paper-source:0;}
> div.Section1 {page:Section1;} --> Hi all, Im
> a psychologist moving from anova to lmer for analysis of
> response time (and error data) from a reaction time experiment,
> and have a question about specifying the structure of random
> effects in the model. Many of the examples I am reading about
> involve split-plot designs where effects are nested within each
> other .... This is not the case here: I am using a typical
> repeated measures design where each subject does
> everything: 30 subjects judge the laterality of a hand
> shown on a computer screen, completing 18 trials in all
> combinations of the following experimental factors (Angle, 8
> levels (hand shown in 8 different orientations), Laterality, 2
> levels (both right and left hands shown), Condition, 3 levels
> (participant holds own hands in posture a,b or c). With
> repeated measures ANOVA the Error structure is specified as
> follows
> aov(percentErrors~angle*condition*laterality+Error(subject/(angle*condition*laterality)),data=errorData)
>
> aov(meanRT~angle*condition*laterality+Error(subject/(angle*condition*laterality)),data=RTdata)
> where response variables, percentErrors and meanRT, are the percentage of
> errors (out of 18 trials) and the mean reaction time over 18 trials I
> need to move to lmer as variance is not constant across Angle and error data
> are bounded (0,1) so I should use the binomial family my first pass
> (looking at all main effects & possible interactions) is:
> lmer(cbind(numErrors,numCorrect)~angle*condition*laterality+(angle*condition*laterality|subject),family=binomial),data=errorData)
> where numErrors+numCorrect = 18 for each subject &
> Angle-Laterality-Condition combination
> lmer(meanRT~angle*condition*laterality+(angle*condition*laterality|subject),data=RTdata)
> I am unsure if this is correct? Help is welcome, thanks - Nuala
>
>
> Nuala Brady
> School of Psychology
> University College Dublin
> Belfield, D4
> IRELAND
>
> +353 (0)1 716 8247
> nuala.br...@ucd.ie
>
>
>
> [[alternative HTML version deleted]]
> > ______________________________________________
> R-help@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide http://www.R-
> project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
Nuala Brady
School of Psychology
University College Dublin
Belfield, D4
IRELAND
+353 (0)1 716 8247
nuala.br...@ucd.ie
[[alternative HTML version deleted]]
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
