<!-- /* 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.