Lisa Snel
2019-Jan-04 07:41 UTC
[R] How to perform Mixed Design ANOVA on MICE imputed dataset in R?
Hi all,
I have a question about performing a Mixed Design ANOVA in R after multiple
imputation using MICE. My data is as follows:
id <- c(1,2,3,4,5,6,7,8,9,10)
group <- c(0,1,1,0,0,1,0,0,0,1)
measure_1 <- c(60,80,90,54,60,61,77,67,88,90)
measure_2 <- c(55,88,88,55,70,62,78,66,65,92)
measure_3 <- c(58,88,85,56,68,62,89,62,70,99)
measure_4 <- c(64,80,78,92,65,64,87,65,67,96)
measure_5 <- c(64,85,80,65,74,69,90,65,70,99)
measure_6 <- c(70,83,80,55,73,64,91,65,91,89)
dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4,
measure_5, measure_6)
dat$group <- as.factor(dat$group)
So: we have 6 repeated measurements of diastolic blood pressure (measure 1 till
6). The grouping factor is gender, which is called group. This variable is coded
1 if male and 0 if female. Before multiple imputation, we have used the
following code in R:
library(reshape)
library(reshape2)
datLong <- melt(dat, id = c("id", "group"), measured =
c("measure_1", "measure_2", "measure_3",
"measure_4", "measure_5", "measure_6"))
datLong
colnames(datLong) <- c("ID", "Gender", "Time",
"Score")
datLong
table(datLong$Time)
datLong$ID <- as.factor(datLong$ID)
library(ez)
model_mixed <- ezANOVA(data = datLong,
dv = Value,
wid = ID,
within = Time,
between = Gender,
detailed = TRUE,
type = 3,
return_aov = TRUE)
model_mixed
This worked perfectly. However, our data is not complete. We have missing
values, that we impute using MICE:
id <- c(1,2,3,4,5,6,7,8,9,10)
group <- c(0,1,1,0,0,1,0,0,0,1)
measure_1 <- c(60,80,90,54,60,61,77,67,88,90)
measure_2 <- c(55,NA,88,55,70,62,78,66,65,92)
measure_3 <- c(58,88,85,56,68,62,89,62,70,99)
measure_4 <- c(64,80,78,92,NA,NA,87,65,67,96)
measure_5 <- c(64,85,80,65,74,69,90,65,70,99)
measure_6 <- c(70,NA,80,55,73,64,91,65,91,89)
dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4,
measure_5, measure_6)
dat$group <- as.factor(dat$group)
imp_anova <- mice(dat, maxit = 0)
meth <- imp_anova$method
pred <- imp_anova$predictorMatrix
imp_anova <- mice(dat, method = meth, predictorMatrix = pred, seed = 2018,
maxit = 10, m = 5)
(The imputation gives logged events, because of the made-up data and the simple
imputation code e.g id used as a predictor. For my real data, the imputation was
correct and valid)
Now I have the imputed dataset of class ?mids?. I have searched the internet,
but I cannot find how I can perform the mixed design ANOVA on this imputed set,
as I did before with the complete set using ezANOVA. Is there anyone who can and
wants to help me?
Best,
Lisa
[[alternative HTML version deleted]]
Ista Zahn
2019-Jan-04 14:11 UTC
[R] How to perform Mixed Design ANOVA on MICE imputed dataset in R?
Hi Lisa, The package web page at http://stefvanbuuren.github.io/mice/ has all the info you need to get started. Best, Ista On Fri, Jan 4, 2019 at 3:29 AM Lisa Snel <lisa_1995_s at hotmail.com> wrote:> > Hi all, > > I have a question about performing a Mixed Design ANOVA in R after multiple imputation using MICE. My data is as follows: > > id <- c(1,2,3,4,5,6,7,8,9,10) > group <- c(0,1,1,0,0,1,0,0,0,1) > measure_1 <- c(60,80,90,54,60,61,77,67,88,90) > measure_2 <- c(55,88,88,55,70,62,78,66,65,92) > measure_3 <- c(58,88,85,56,68,62,89,62,70,99) > measure_4 <- c(64,80,78,92,65,64,87,65,67,96) > measure_5 <- c(64,85,80,65,74,69,90,65,70,99) > measure_6 <- c(70,83,80,55,73,64,91,65,91,89) > dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4, measure_5, measure_6) > dat$group <- as.factor(dat$group) > > So: we have 6 repeated measurements of diastolic blood pressure (measure 1 till 6). The grouping factor is gender, which is called group. This variable is coded 1 if male and 0 if female. Before multiple imputation, we have used the following code in R: > > library(reshape) > library(reshape2) > datLong <- melt(dat, id = c("id", "group"), measured = c("measure_1", "measure_2", "measure_3", "measure_4", "measure_5", "measure_6")) > datLong > > colnames(datLong) <- c("ID", "Gender", "Time", "Score") > datLong > table(datLong$Time) > datLong$ID <- as.factor(datLong$ID) > > library(ez) > model_mixed <- ezANOVA(data = datLong, > dv = Value, > wid = ID, > within = Time, > between = Gender, > detailed = TRUE, > type = 3, > return_aov = TRUE) > model_mixed > > This worked perfectly. However, our data is not complete. We have missing values, that we impute using MICE: > > id <- c(1,2,3,4,5,6,7,8,9,10) > group <- c(0,1,1,0,0,1,0,0,0,1) > measure_1 <- c(60,80,90,54,60,61,77,67,88,90) > measure_2 <- c(55,NA,88,55,70,62,78,66,65,92) > measure_3 <- c(58,88,85,56,68,62,89,62,70,99) > measure_4 <- c(64,80,78,92,NA,NA,87,65,67,96) > measure_5 <- c(64,85,80,65,74,69,90,65,70,99) > measure_6 <- c(70,NA,80,55,73,64,91,65,91,89) > dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4, measure_5, measure_6) > dat$group <- as.factor(dat$group) > > imp_anova <- mice(dat, maxit = 0) > meth <- imp_anova$method > pred <- imp_anova$predictorMatrix > imp_anova <- mice(dat, method = meth, predictorMatrix = pred, seed = 2018, maxit = 10, m = 5) > > (The imputation gives logged events, because of the made-up data and the simple imputation code e.g id used as a predictor. For my real data, the imputation was correct and valid) > > Now I have the imputed dataset of class ?mids?. I have searched the internet, but I cannot find how I can perform the mixed design ANOVA on this imputed set, as I did before with the complete set using ezANOVA. Is there anyone who can and wants to help me? > > > Best, > > Lisa > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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.
Lisa Snel
2019-Jan-04 14:49 UTC
[R] How to perform Mixed Design ANOVA on MICE imputed dataset in R?
Dear Ista, Thank you for your response and the link you have sent me. However, I know the basic things of the MICE package (how impute, to pool and to do basic analyses), but the problem is that I cannot find anything about this specific analysis. Best, Lisa ________________________________ Van: Ista Zahn <istazahn at gmail.com> Verzonden: vrijdag 4 januari 2019 15:11 Aan: Lisa Snel CC: r-help at r-project.org Onderwerp: Re: [R] How to perform Mixed Design ANOVA on MICE imputed dataset in R? Hi Lisa, The package web page at http://stefvanbuuren.github.io/mice/ has all the info you need to get started. Best, Ista On Fri, Jan 4, 2019 at 3:29 AM Lisa Snel <lisa_1995_s at hotmail.com> wrote:> > Hi all, > > I have a question about performing a Mixed Design ANOVA in R after multiple imputation using MICE. My data is as follows: > > id <- c(1,2,3,4,5,6,7,8,9,10) > group <- c(0,1,1,0,0,1,0,0,0,1) > measure_1 <- c(60,80,90,54,60,61,77,67,88,90) > measure_2 <- c(55,88,88,55,70,62,78,66,65,92) > measure_3 <- c(58,88,85,56,68,62,89,62,70,99) > measure_4 <- c(64,80,78,92,65,64,87,65,67,96) > measure_5 <- c(64,85,80,65,74,69,90,65,70,99) > measure_6 <- c(70,83,80,55,73,64,91,65,91,89) > dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4, measure_5, measure_6) > dat$group <- as.factor(dat$group) > > So: we have 6 repeated measurements of diastolic blood pressure (measure 1 till 6). The grouping factor is gender, which is called group. This variable is coded 1 if male and 0 if female. Before multiple imputation, we have used the following code in R: > > library(reshape) > library(reshape2) > datLong <- melt(dat, id = c("id", "group"), measured = c("measure_1", "measure_2", "measure_3", "measure_4", "measure_5", "measure_6")) > datLong > > colnames(datLong) <- c("ID", "Gender", "Time", "Score") > datLong > table(datLong$Time) > datLong$ID <- as.factor(datLong$ID) > > library(ez) > model_mixed <- ezANOVA(data = datLong, > dv = Value, > wid = ID, > within = Time, > between = Gender, > detailed = TRUE, > type = 3, > return_aov = TRUE) > model_mixed > > This worked perfectly. However, our data is not complete. We have missing values, that we impute using MICE: > > id <- c(1,2,3,4,5,6,7,8,9,10) > group <- c(0,1,1,0,0,1,0,0,0,1) > measure_1 <- c(60,80,90,54,60,61,77,67,88,90) > measure_2 <- c(55,NA,88,55,70,62,78,66,65,92) > measure_3 <- c(58,88,85,56,68,62,89,62,70,99) > measure_4 <- c(64,80,78,92,NA,NA,87,65,67,96) > measure_5 <- c(64,85,80,65,74,69,90,65,70,99) > measure_6 <- c(70,NA,80,55,73,64,91,65,91,89) > dat <- data.frame(id, group, measure_1, measure_2, measure_3, measure_4, measure_5, measure_6) > dat$group <- as.factor(dat$group) > > imp_anova <- mice(dat, maxit = 0) > meth <- imp_anova$method > pred <- imp_anova$predictorMatrix > imp_anova <- mice(dat, method = meth, predictorMatrix = pred, seed = 2018, maxit = 10, m = 5) > > (The imputation gives logged events, because of the made-up data and the simple imputation code e.g id used as a predictor. For my real data, the imputation was correct and valid) > > Now I have the imputed dataset of class ?mids?. I have searched the internet, but I cannot find how I can perform the mixed design ANOVA on this imputed set, as I did before with the complete set using ezANOVA. Is there anyone who can and wants to help me? > > > Best, > > Lisa > > [[alternative HTML version deleted]] > > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > 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.[[alternative HTML version deleted]]