[R] mixed models lmer function help!!
bgunter.4567 at gmail.com
Mon Jun 19 22:08:40 CEST 2017
1. A mess, because you failed to read and follow the posting guide:
This is a **plain text** mailing list, which means that html can get
mangled, as you have demonstrated.
2. And wrong list: the r-sig-mixed-models list is where this would be
"The trouble with having an open mind is that people keep coming along
and sticking things into it."
-- Opus (aka Berkeley Breathed in his "Bloom County" comic strip )
On Mon, Jun 19, 2017 at 9:29 AM, Shelby Leonard via R-help
<r-help at r-project.org> wrote:
> Hi,I have tumor growth curve data for a bunch of different mice in various groups. I want to compare the growth curves of the different groups to see if timing of drug delivery changed tumor growth.I am trying to run a mixed models with repeated measures over time with each mouse as a random effect with linear and quadratic terms for time.This took me a long time to figure out and I just wanted to make sure I did it correctly and I am interpreting it correctly. This is the code I ran
> Rtumor<-lmer(volume~Group+Time+(1|Subject), data=Rtumor)summary(Rtumor)Rtumor.null=lmer(volume~Time+(1|Subject), data=Rtumor, REML=FALSE)Rtumor.full=lmer(volume~Group+Time+(1|Subject), data=Rtumor, REML=FALSE)anova(Rtumor.null,Rtumor.full)
> Here is my output Rtumor<-lmer(volume~Group+Time+(1|Subject), data=Rtumor)> summary(Rtumor)Linear mixed model fit by REML ['lmerMod']Formula: volume ~ Group + Time + (1 | Subject) Data: Rtumor
> REML criterion at convergence: 1541.2
> Scaled residuals: Min 1Q Median 3Q Max -1.8006 -0.6348 -0.0658 0.3903 4.7551
> Random effects: Groups Name Variance Std.Dev. Subject (Intercept) 3.197e-09 5.654e-05 Residual 3.348e+05 5.786e+02Number of obs: 101, groups: Subject, 11
> Fixed effects: Estimate Std. Error t value(Intercept) -495.520 303.619 -1.632Group 24.350 115.615 0.211Time 79.653 7.886 10.101
> Correlation of Fixed Effects: (Intr) Group Group -0.933 Time -0.300 -0.007
>> Rtumor.null=lmer(volume~Time+(1|Subject), data=Rtumor, REML=FALSE)> Rtumor.full=lmer(volume~Group+Time+(1|Subject), data=Rtumor, REML=FALSE)> anova(Rtumor.null,Rtumor.full)Data: RtumorModels:Rtumor.null: volume ~ Time + (1 | Subject)Rtumor.full: volume ~ Group + Time + (1 | Subject) Df AIC BIC logLik deviance Chisq Chi Df Pr(>Chisq)Rtumor.null 4 1576.5 1586.9 -784.24 1568.5 Rtumor.full 5 1578.4 1591.5 -784.22 1568.4 0.0457 1 0.8307There were 50 or more warnings (use warnings() to see the first 50)
> My questions are1) Did I do this correctly?2) Do I still need to run it again with quadratic terms for time?, If so, how do I do this?3) If I am understanding these results correctly, they say There is no difference between these groups on volume growth curves
> [[alternative HTML version deleted]]
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help