[R] mixed effects or fixed effects?
Hi, I am running a learning experiment in which both training subjects and controls complete a pretest and posttest. All analyses are being conducted in R. We are looking to compare two training methodologies, and so have run this experiment twice, once with each methodology. Methodology is a between-subjects factor. Trying to run this analysis with every factor included (ie, subject as a random factor, session nested within group nested within experiment) seems to me (after having tried) to be clumsy and probably uninterpretable. My favoured model for the analysis is a linear mixed-effects model, and to combine the data meaningfully, I have collated all the pretest data for controls and trained subjects from each experiment, and assumed this data to represent a population sample for naive subjects for each experiment. I have also ditched the posttest data for the controls, and assumed the posttest training data to represent a population sample for trained subjects for each experiment. I have confirmed the validity of these assumptions by ascertaining that a) controls and trained listeners did not differ significantly at pretest for either experiment; and b) control listeners did not learn significantly between pretest and posttest (and therefore their posttest data are not relevant). This was done using a linear mixed-effects model for each experiment, with subject as a random factor and session (pretest vs posttest) nested within Group (trained vs control). Therefore, the model I want to use to analyse the data would ideally be a linear mixed-effects model, with subject as a random factor, and session (pre vs post) nested within experiment. Note that my removal of the Group (Trained vs Control) factor simplifies the model somewhat, and makes it more interpretable in terms of evaluating the relative effects of each experiment. What I would like to know is- a) would people agree that this is a meaningful way to combine my data? I believe the logic is sound, but am slightly concerned that I am ignoring a whole block of posttest data for the controls (even though this does not account for a significant amount of the variance); and b) given that each of my trained subjects appear twice- one in the pretest and once in the posttest, and the controls only appear once- in the pretest sample, is there any problem with making subject a random factor? Conceptually, I see no problem with this, but I would like to be sure before I finish writing up. Many thanks for your time Dan __ R-help@stat.math.ethz.ch 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.
[R] R interpretation
Hi, I am new to R (and not really a stats expert) and am having trouble interpreting its output. I am running a human learning experiment, with 6 scores per subject in both the pretest and the posttest. I believe I have fitted the correct model for my data- a mixed-effects design, with subject as a random factor and session (pre vs post) nested within group (trained vs control). I am confused about the output. The summary command gives me this table: D.lme- lme(score~GROUP/session, random=~1|subject, data=ILD4L ) summary(D.lme) Linear mixed-effects model fit by REML Data: ILD4L Subset: EXP == F AIC BIC logLik -63.69801 -45.09881 37.84900 Random effects: Formula: ~1 | subject (Intercept) Residual StdDev: 0.1032511 0.1727145 Fixed effects: score ~ GROUP/session Value Std.Error DF t-value p-value (Intercept) 0.10252778 0.05104328 152 2.008644 0.0463 GROUPT 0.09545347 0.06752391 12 1.413625 0.1829 GROUPC:sessionpost -0.00441389 0.04070919 152 -0.108425 0.9138 GROUPT:sessionpost -0.23586042 0.03525520 152 -6.690090 0. Correlation: (Intr) GROUPT GROUPC GROUPT -0.756 GROUPC:sessionpost -0.399 0.301 GROUPT:sessionpost 0.000 -0.261 0.000 Standardized Within-Group Residuals: Min Q1 Med Q3 Max -2.66977386 -0.52935645 -0.08616759 0.57215015 3.26532101 Number of Observations: 168 Number of Groups: 14 I believe the fixed-effects section of this output to be telling me that my model intercept (which I assume to be the control group pretest?) is significantly different from 0, and that GROUPT (i.e. the trained group) does not differ significantly from the intercept- therefore no pretest difference between groups? The next line is, I believe showing that the GROUPC x sessionpost interaction (i.e., control posttest scores?) is not significantly different from the intercept (i.e. control pretest scores). Finally, I am interpreting the final line as indicating that the GROUPT x sessionpost interaction (ie, trained posttest scores?) is significantly different from the trained pretest scores (GROUPT). A treatment contrast that I would like to apply would be for Control-post vs Trained-post, to see if the groups differ after training, but I'm not sure how to do this- and I feel I am probably overcomplicating the matter. also, I am confused about how to report this output in my publication. For instance, what should I be reporting for df? Those found on the output of the anova table? Would it be possible to look through this for me and indicate how to interpret the R output, and also how I should be reporting this? Apologies for asking such basic questions, but I would like to start using R more regularly and to make sure I am doing so correctly. Many thanks, Dan __ R-help@stat.math.ethz.ch 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.