Loading the dataset
data.test4 <- read.csv("~/Final_Adult_Study_R_Docs/adult_study011615.csv")
# Load the psych package
library(psych)
data.test4$meanRES <- apply(data.test4[, c("RES1", "RES2", "RES3")], 1, mean, na.rm = TRUE)
library(reshape2); library(car); library(Amelia);library(mitools);library(nlme);library(predictmeans)
##
## Attaching package: 'car'
##
## The following object is masked from 'package:psych':
##
## logit
##
## Loading required package: Rcpp
## ##
## ## Amelia II: Multiple Imputation
## ## (Version 1.7.3, built: 2014-11-14)
## ## Copyright (C) 2005-2015 James Honaker, Gary King and Matthew Blackwell
## ## Refer to http://gking.harvard.edu/amelia/ for more information
## ##
## Loading required package: lme4
## Loading required package: Matrix
##
## Attaching package: 'lme4'
##
## The following object is masked from 'package:nlme':
##
## lmList
#Remove the meanRES and ID Group and wave from data.test4 and create a new #dataset with only those variables.
data <- data.test4[,c("ID", "GROUP", "wave", "meanRES")]
#Use dcast to cnage from long-format data to wide format data
data <- dcast(data, ID + GROUP ~ wave, mean, value.var = "meanRES")
# Changing all NaNs to NA
data[,3:5] <- apply(data[,3:5],2,function(x) recode(x, "NaN = NA") )
Unsing the mapply function we create a new data set with ID Group baseline meanRES and wave 2 and 3 of meanRES. So we have a Baseline, which is Time 1 (placed in column 3 one on top of the other) to compare to both Time 2 and 3 (placed in column 4 one on top of the other). In the next line of code we then create a separate column called “wave” which calls the first 89 (which compares Time 2 to Baseline) “wave 1” and then the second 89 we call “wave 2” which compares Time 3 to Baseline. In the third line of code we add names to the new columns of the dataset.
data2 <- as.data.frame(mapply(c,data[,1:4], data[,c(1:3,5)]))
data2$wave <- rep(1:2, each=89)
names(data2) <- c("ID", "GROUP", "BASELINE", "meanRES", "WAVE")
Intention to treat model (ITT) where we keep the cases who dropped out and did not complete the study (http://en.wikipedia.org/wiki/Intention-to-treat_analysis).
data2[which(data2$GROUP ==2), "GROUP"] <- 1
data2$GROUP <-as.factor(data2$GROUP)
data2$ID <-as.factor(data2$ID)
Imputing missing data
MI <- amelia(data2, 50, idvars = c("ID"), ords = "GROUP")
## -- Imputation 1 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 2 --
##
## 1 2 3 4 5 6 7 8 9 10 11 12 13
##
## -- Imputation 3 --
##
## 1 2 3 4 5 6
##
## -- Imputation 4 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 5 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 6 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 7 --
##
## 1 2 3 4 5 6 7
##
## -- Imputation 8 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 9 --
##
## 1 2 3 4 5 6 7 8 9 10 11
##
## -- Imputation 10 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 11 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 12 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 13 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 14 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 15 --
##
## 1 2 3 4 5 6 7 8 9 10 11 12 13
##
## -- Imputation 16 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 17 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 18 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 19 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 20 --
##
## 1 2 3 4 5 6
##
## -- Imputation 21 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 22 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 23 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 24 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 25 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 26 --
##
## 1 2 3 4 5 6 7 8 9 10 11
##
## -- Imputation 27 --
##
## 1 2 3 4 5 6 7
##
## -- Imputation 28 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 29 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 30 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 31 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 32 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 33 --
##
## 1 2 3 4 5 6 7
##
## -- Imputation 34 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 35 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 36 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 37 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 38 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 39 --
##
## 1 2 3 4 5 6 7
##
## -- Imputation 40 --
##
## 1 2 3 4 5 6 7 8 9 10
##
## -- Imputation 41 --
##
## 1 2 3 4 5 6 7 8 9 10 11
##
## -- Imputation 42 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 43 --
##
## 1 2 3 4 5 6 7 8
##
## -- Imputation 44 --
##
## 1 2 3 4 5 6 7 8 9 10 11
##
## -- Imputation 45 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 46 --
##
## 1 2 3 4 5 6 7 8 9 10 11
##
## -- Imputation 47 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 48 --
##
## 1 2 3 4 5 6 7 8 9 10 11 12
##
## -- Imputation 49 --
##
## 1 2 3 4 5 6 7 8 9
##
## -- Imputation 50 --
##
## 1 2 3 4 5 6 7 8 9
Creating new dataset with missing data imputed
data(MI$imputations)
## Warning in data(MI$imputations): data set 'MI$imputations' not found
allimplogreg<-lapply(MI$imputations,function(X) {lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = X, method = "ML", na.action = "na.omit")})
betas<-MIextract(allimplogreg, fun=fixef)
vars<-MIextract(allimplogreg, fun=vcov)
summary(MIcombine(betas,vars))
## Multiple imputation results:
## MIcombine.default(betas, vars)
## results se (lower upper) missInfo
## (Intercept) 1.0271050 0.66072045 -0.2738149 2.3280249 43 %
## GROUP1 -0.2135718 0.59871930 -1.3884113 0.9612676 22 %
## WAVE 0.2632860 0.28685338 -0.3002616 0.8268336 31 %
## BASELINE 0.8248579 0.08035961 0.6658420 0.9838738 63 %
## GROUP1:WAVE 0.1837799 0.39331851 -0.5886972 0.9562571 29 %
Check results with Imputations using Zelig
library("Zelig")
## Loading required package: boot
##
## Attaching package: 'boot'
##
## The following object is masked from 'package:car':
##
## logit
##
## The following object is masked from 'package:psych':
##
## logit
##
## Loading required package: MASS
## Loading required package: sandwich
## ZELIG (Versions 4.2-1, built: 2013-09-12)
##
## +----------------------------------------------------------------+
## | Please refer to http://gking.harvard.edu/zelig for full |
## | documentation or help.zelig() for help with commands and |
## | models support by Zelig. |
## | |
## | Zelig project citations: |
## | Kosuke Imai, Gary King, and Olivia Lau. (2009). |
## | ``Zelig: Everyone's Statistical Software,'' |
## | http://gking.harvard.edu/zelig |
## | and |
## | Kosuke Imai, Gary King, and Olivia Lau. (2008). |
## | ``Toward A Common Framework for Statistical Analysis |
## | and Development,'' Journal of Computational and |
## | Graphical Statistics, Vol. 17, No. 4 (December) |
## | pp. 892-913. |
## | |
## | To cite individual Zelig models, please use the citation |
## | format printed with each model run and in the documentation. |
## +----------------------------------------------------------------+
##
##
##
## Attaching package: 'Zelig'
##
## The following objects are masked from 'package:psych':
##
## alpha, describe, sim
##
## The following object is masked from 'package:utils':
##
## cite
zelig.fit <- zelig(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = MI$imputations, model = "ls", cite = FALSE)
summary(zelig.fit)
##
## Model: ls
## Number of multiply imputed data sets: 50
##
## Combined results:
##
## Call:
## lm(formula = formula, weights = weights, model = F, data = data)
##
## Coefficients:
## Value Std. Error t-stat p-value
## (Intercept) 1.0241802 0.66779766 1.5336684 1.262473e-01
## GROUP1 -0.2136624 0.62291450 -0.3430044 7.316545e-01
## WAVE 0.2632860 0.29912069 0.8801998 3.790986e-01
## BASELINE 0.8253065 0.07946188 10.3861943 1.653024e-18
## GROUP1:WAVE 0.1837680 0.41062360 0.4475339 6.546282e-01
##
## For combined results from datasets i to j, use summary(x, subset = i:j).
## For separate results, use print(summary(x), subset = i:j).
Check assumptions with Random Computations
data1=MI$imputations[[1]]
library("Zelig")
zelig.fitdata1 <- zelig(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data1, model = "ls", cite = FALSE)
summary(zelig.fitdata1)
##
## Call:
## lm(formula = formula, weights = weights, model = F, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.3325 -0.7106 0.0322 0.8306 3.0941
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.34557 0.54118 2.486 0.0139 *
## GROUP1 -0.08889 0.59425 -0.150 0.8813
## WAVE 0.53166 0.27014 1.968 0.0507 .
## BASELINE 0.74068 0.05097 14.531 <2e-16 ***
## GROUP1:WAVE -0.08614 0.37576 -0.229 0.8190
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.253 on 173 degrees of freedom
## Multiple R-squared: 0.5579, Adjusted R-squared: 0.5476
## F-statistic: 54.57 on 4 and 173 DF, p-value: < 2.2e-16
Describe the meanRES variable by the GROUP variable
describeBy(data1[,3:4], group = data1$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 86 6.52 2.01 7.00 6.59 1.98 2.67 10.00 7.33 -0.38
## meanRES 2 86 6.97 1.99 7.33 7.06 0.99 0.67 12.13 11.46 -0.56
## kurtosis se
## BASELINE -0.94 0.22
## meanRES 0.73 0.21
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 92 6.81 1.70 7.00 6.90 1.48 2.67 10.00 7.33 -0.48
## meanRES 2 92 6.97 1.74 7.05 7.09 1.59 2.33 10.36 8.03 -0.49
## kurtosis se
## BASELINE -0.41 0.18
## meanRES -0.40 0.18
Create a plot that visualizes meanRES variable by the GROUP variable
library(ggplot2)
##
## Attaching package: 'ggplot2'
##
## The following object is masked from 'package:psych':
##
## %+%
library(influence.ME)
##
## Attaching package: 'influence.ME'
##
## The following object is masked from 'package:stats':
##
## influence
Take a look at the residuals
residual <- lm(meanRES ~ BASELINE, data=data1)$residual
Plot the residuals to see that they are random
plot(density(residual))# A density plot
qqnorm(residual) # A quantile normal plot to checking normality
qqline(residual)
Checking the different between intervention and control groups residuals. This allows us to control for individual unsystematic differences.
data2$residual <- NA
sel1 <- which(!is.na(data1$meanRES))
sel2 <- which(!is.na(data1$BASELINE))
data1$residual[intersect(sel1,sel2)] <- residual
qplot(GROUP, meanRES, data=data1, geom="boxplot")
Plot of the difference between intervention and control groups.
qplot(GROUP, residual, data=data1, geom="boxplot")
Two way repeated measures ======================================================== Graphing the Two-Way Interaction. Both meanRES and the Residuals
# Load the nlme package
library(nlme)
with(data1, boxplot(meanRES ~ WAVE + GROUP))
with(data1, boxplot(residual ~ WAVE + GROUP))
Comparing Basline to Wave 2 and 3 by Group.
fullModeldata1 <- lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data1, method = "ML", na.action = "na.omit")
CookD(fullModeldata1)
plot(fullModeldata1, which="cook")
Explanation of significance:
We asses the significance of our models by comparing them from the baseline model using the anova() function.
(Intercept): Where everything is 0
GROUP1: Is there a difference between group. If it is significant than there is a difference and the treatment had an effect.
WAVE: Asseses whether the effects gets bigger beteen time 2 and 3 (does not have to be significant)
BASELINE: Should not be significant. If it is then it shows that there is a difference between groups before the treatment.
GROUP1:WAVE: If this is significant then it means that the effect was either fleeting or it happened after the treatment i.e. between time 2 and 3.
summary(fullModeldata1)
## Linear mixed-effects model fit by maximum likelihood
## Data: data1
## AIC BIC logLik
## 590.9248 613.1973 -288.4624
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.5404128 1.110359
##
## Fixed effects: meanRES ~ GROUP * WAVE + BASELINE
## Value Std.Error DF t-value p-value
## (Intercept) 1.3555209 0.5346814 87 2.535194 0.0130
## GROUP1 -0.0885296 0.5468859 87 -0.161879 0.8718
## WAVE 0.5316604 0.2429020 86 2.188786 0.0313
## BASELINE 0.7391517 0.0556006 86 13.293955 0.0000
## GROUP1:WAVE -0.0860835 0.3378739 86 -0.254780 0.7995
## Correlation:
## (Intr) GROUP1 WAVE BASELI
## GROUP1 -0.512
## WAVE -0.681 0.666
## BASELINE -0.678 -0.024 0.000
## GROUP1:WAVE 0.494 -0.927 -0.719 -0.006
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.14617133 -0.52981052 0.01388978 0.56475712 2.63406133
##
## Number of Observations: 178
## Number of Groups: 89
Another random imputation
data10=MI$imputations[[10]]
library("Zelig")
zelig.fitdata10 <- zelig(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data10, model = "ls", cite = FALSE)
summary(zelig.fitdata10)
##
## Call:
## lm(formula = formula, weights = weights, model = F, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.6235 -0.6328 0.1470 0.6860 2.8041
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.22719 0.49072 2.501 0.0133 *
## GROUP1 -0.11339 0.53661 -0.211 0.8329
## WAVE 0.52643 0.24392 2.158 0.0323 *
## BASELINE 0.76094 0.04654 16.350 <2e-16 ***
## GROUP1:WAVE 0.06512 0.33929 0.192 0.8480
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.131 on 173 degrees of freedom
## Multiple R-squared: 0.6173, Adjusted R-squared: 0.6084
## F-statistic: 69.76 on 4 and 173 DF, p-value: < 2.2e-16
Describe the meanRES variable by the GROUP variable
describeBy(data10[,3:4], group = data10$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 86 6.52 2.01 7.00 6.59 1.98 2.67 10 7.33 -0.38
## meanRES 2 86 6.98 1.85 7.36 7.11 1.22 0.67 10 9.33 -0.83
## kurtosis se
## BASELINE -0.94 0.22
## meanRES 0.63 0.20
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 92 6.77 1.65 7.00 6.86 1.48 2.67 10.00 7.33 -0.53
## meanRES 2 92 7.15 1.77 7.63 7.27 1.71 2.33 11.22 8.89 -0.54
## kurtosis se
## BASELINE -0.33 0.17
## meanRES -0.24 0.18
Create a plot that visualizes meanRES variable by the GROUP variable
library(ggplot2)
library(influence.ME)
Take a look at the residuals
residual <- lm(meanRES ~ BASELINE, data=data10)$residual
Plot the residuals to see that they are random
plot(density(residual))# A density plot
qqnorm(residual) # A quantile normal plot to checking normality
qqline(residual)
Checking the different between intervention and control groups residuals. This allows us to control for individual unsystematic differences.
data2$residual <- NA
sel1 <- which(!is.na(data10$meanRES))
sel2 <- which(!is.na(data10$BASELINE))
data10$residual[intersect(sel1,sel2)] <- residual
qplot(GROUP, meanRES, data=data10, geom="boxplot")
Plot of the difference between intervention and control groups.
qplot(GROUP, residual, data=data10, geom="boxplot")
Two way repeated measures ======================================================== Graphing the Two-Way Interaction. Both meanRES and the Residuals
# Load the nlme package
library(nlme)
with(data10, boxplot(meanRES ~ WAVE + GROUP))
with(data10, boxplot(residual ~ WAVE + GROUP))
Comparing Basline to Wave 2 and 3 by Group.
fullModeldata10 <- lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data10, method = "ML", na.action = "na.omit")
CookD(fullModeldata10)
plot(fullModeldata10, which="cook")
Explanation of significance:
We asses the significance of our models by comparing them from the baseline model using the anova() function.
(Intercept): Where everything is 0
GROUP1: Is there a difference between group. If it is significant than there is a difference and the treatment had an effect.
WAVE: Asseses whether the effects gets bigger beteen time 2 and 3 (does not have to be significant)
BASELINE: Should not be significant. If it is then it shows that there is a difference between groups before the treatment.
GROUP1:WAVE: If this is significant then it means that the effect was either fleeting or it happened after the treatment i.e. between time 2 and 3.
summary(fullModeldata10)
## Linear mixed-effects model fit by maximum likelihood
## Data: data10
## AIC BIC logLik
## 556.7672 579.0397 -271.3836
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.3739858 1.050431
##
## Fixed effects: meanRES ~ GROUP * WAVE + BASELINE
## Value Std.Error DF t-value p-value
## (Intercept) 1.2274227 0.4875767 87 2.517394 0.0137
## GROUP1 -0.1133764 0.5119261 87 -0.221470 0.8252
## WAVE 0.5264335 0.2297921 86 2.290912 0.0244
## BASELINE 0.7609037 0.0490780 86 15.503958 0.0000
## GROUP1:WAVE 0.0651190 0.3196338 86 0.203730 0.8390
## Correlation:
## (Intr) GROUP1 WAVE BASELI
## GROUP1 -0.525
## WAVE -0.707 0.673
## BASELINE -0.656 -0.026 0.000
## GROUP1:WAVE 0.506 -0.937 -0.719 0.003
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.8570766 -0.4974058 0.1049904 0.5687370 2.6181529
##
## Number of Observations: 178
## Number of Groups: 89
Another random imputation
data15=MI$imputations[[15]]
library("Zelig")
zelig.fitdata15 <- zelig(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data15, model = "ls", cite = FALSE)
summary(zelig.fitdata15)
##
## Call:
## lm(formula = formula, weights = weights, model = F, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.6359 -0.7149 0.1456 0.6815 3.8616
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.8262 0.5687 3.211 0.00158 **
## GROUP1 -0.4035 0.6222 -0.648 0.51753
## WAVE 0.1522 0.2828 0.538 0.59122
## BASELINE 0.7344 0.0539 13.626 < 2e-16 ***
## GROUP1:WAVE 0.3839 0.3934 0.976 0.33053
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.311 on 173 degrees of freedom
## Multiple R-squared: 0.5276, Adjusted R-squared: 0.5167
## F-statistic: 48.31 on 4 and 173 DF, p-value: < 2.2e-16
Describe the meanRES variable by the GROUP variable
describeBy(data15[,3:4], group = data15$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 86 6.52 2.01 7.00 6.59 1.98 2.67 10.00 7.33 -0.38
## meanRES 2 86 6.84 1.91 7.04 6.93 1.43 0.67 11.41 10.75 -0.56
## kurtosis se
## BASELINE -0.94 0.22
## meanRES 0.45 0.21
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 92 6.77 1.66 7.00 6.86 1.48 2.67 10.0 7.33 -0.52
## meanRES 2 92 7.20 1.86 7.46 7.31 1.96 1.44 10.4 8.97 -0.61
## kurtosis se
## BASELINE -0.36 0.17
## meanRES 0.07 0.19
Create a plot that visualizes meanRES variable by the GROUP variable
library(ggplot2)
library(influence.ME)
Take a look at the residuals
residual <- lm(meanRES ~ BASELINE, data=data15)$residual
Plot the residuals to see that they are random
plot(density(residual))# A density plot
qqnorm(residual) # A quantile normal plot to checking normality
qqline(residual)
Checking the different between intervention and control groups residuals. This allows us to control for individual unsystematic differences.
data2$residual <- NA
sel1 <- which(!is.na(data15$meanRES))
sel2 <- which(!is.na(data15$BASELINE))
data15$residual[intersect(sel1,sel2)] <- residual
qplot(GROUP, meanRES, data=data15, geom="boxplot")
Plot of the difference between intervention and control groups.
qplot(GROUP, residual, data=data15, geom="boxplot")
Two way repeated measures ======================================================== Graphing the Two-Way Interaction. Both meanRES and the Residuals
# Load the nlme package
library(nlme)
with(data15, boxplot(meanRES ~ WAVE + GROUP))
with(data15, boxplot(residual ~ WAVE + GROUP))
Comparing Basline to Wave 2 and 3 by Group.
fullModeldata15 <- lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data15, method = "ML", na.action = "na.omit")
CookD(fullModeldata15)
plot(fullModeldata15, which="cook")
Explanation of significance:
We asses the significance of our models by comparing them from the baseline model using the anova() function.
(Intercept): Where everything is 0
GROUP1: Is there a difference between group. If it is significant than there is a difference and the treatment had an effect.
WAVE: Asseses whether the effects gets bigger beteen time 2 and 3 (does not have to be significant)
BASELINE: Should not be significant. If it is then it shows that there is a difference between groups before the treatment.
GROUP1:WAVE: If this is significant then it means that the effect was either fleeting or it happened after the treatment i.e. between time 2 and 3.
summary(fullModeldata15)
## Linear mixed-effects model fit by maximum likelihood
## Data: data15
## AIC BIC logLik
## 610.5805 632.853 -298.2902
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.1675616 1.282038
##
## Fixed effects: meanRES ~ GROUP * WAVE + BASELINE
## Value Std.Error DF t-value p-value
## (Intercept) 1.8260974 0.5682031 87 3.213811 0.0018
## GROUP1 -0.4034717 0.6179857 87 -0.652882 0.5156
## WAVE 0.1521878 0.2804585 86 0.542639 0.5888
## BASELINE 0.7344467 0.0543483 86 13.513710 0.0000
## GROUP1:WAVE 0.3839045 0.3901089 86 0.984096 0.3278
## Correlation:
## (Intr) GROUP1 WAVE BASELI
## GROUP1 -0.550
## WAVE -0.740 0.681
## BASELINE -0.624 -0.020 0.000
## GROUP1:WAVE 0.534 -0.947 -0.719 -0.002
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.5421682 -0.5524806 0.1054307 0.5113592 2.9833331
##
## Number of Observations: 178
## Number of Groups: 89
Another random imputation
data25=MI$imputations[[25]]
library("Zelig")
zelig.fitdata25 <- zelig(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data25, model = "ls", cite = FALSE)
summary(zelig.fitdata25)
##
## Call:
## lm(formula = formula, weights = weights, model = F, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -4.2226 -0.7073 0.0577 0.7612 3.2153
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.72056 0.50929 3.378 0.000901 ***
## GROUP1 -0.37840 0.55798 -0.678 0.498579
## WAVE 0.05181 0.25368 0.204 0.838399
## BASELINE 0.77125 0.04814 16.021 < 2e-16 ***
## GROUP1:WAVE 0.18907 0.35286 0.536 0.592773
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.176 on 173 degrees of freedom
## Multiple R-squared: 0.5988, Adjusted R-squared: 0.5895
## F-statistic: 64.55 on 4 and 173 DF, p-value: < 2.2e-16
Describe the meanRES variable by the GROUP variable
describeBy(data25[,3:4], group = data25$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 86 6.52 2.01 7.00 6.59 1.98 2.67 10.00 7.33 -0.38
## meanRES 2 86 6.83 1.82 7.01 6.93 1.47 0.67 10.27 9.60 -0.69
## kurtosis se
## BASELINE -0.94 0.22
## meanRES 0.59 0.20
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 92 6.76 1.67 7 6.84 1.48 2.67 10 7.33 -0.45
## meanRES 2 92 6.92 1.86 7 7.02 1.98 2.33 10 7.67 -0.41
## kurtosis se
## BASELINE -0.37 0.17
## meanRES -0.77 0.19
Create a plot that visualizes meanRES variable by the GROUP variable
library(ggplot2)
library(influence.ME)
Take a look at the residuals
residual <- lm(meanRES ~ BASELINE, data=data25)$residual
Plot the residuals to see that they are random
plot(density(residual))# A density plot
qqnorm(residual) # A quantile normal plot to checking normality
qqline(residual)
Checking the different between intervention and control groups residuals. This allows us to control for individual unsystematic differences.
data2$residual <- NA
sel1 <- which(!is.na(data25$meanRES))
sel2 <- which(!is.na(data25$BASELINE))
data25$residual[intersect(sel1,sel2)] <- residual
qplot(GROUP, meanRES, data=data25, geom="boxplot")
Plot of the difference between intervention and control groups.
qplot(GROUP, residual, data=data25, geom="boxplot")
Two way repeated measures ======================================================== Graphing the Two-Way Interaction. Both meanRES and the Residuals
# Load the nlme package
library(nlme)
with(data25, boxplot(meanRES ~ WAVE + GROUP))
with(data25, boxplot(residual ~ WAVE + GROUP))
Comparing Basline to Wave 2 and 3 by Group.
fullModeldata25 <- lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data25, method = "ML", na.action = "na.omit")
CookD(fullModeldata25)
plot(fullModeldata25, which="cook")
Explanation of significance:
We asses the significance of our models by comparing them from the baseline model using the anova() function.
(Intercept): Where everything is 0
GROUP1: Is there a difference between group. If it is significant than there is a difference and the treatment had an effect.
WAVE: Asseses whether the effects gets bigger beteen time 2 and 3 (does not have to be significant)
BASELINE: Should not be significant. If it is then it shows that there is a difference between groups before the treatment.
GROUP1:WAVE: If this is significant then it means that the effect was either fleeting or it happened after the treatment i.e. between time 2 and 3.
summary(fullModeldata25)
## Linear mixed-effects model fit by maximum likelihood
## Data: data25
## AIC BIC logLik
## 571.0713 593.3438 -278.5356
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.3555108 1.103773
##
## Fixed effects: meanRES ~ GROUP * WAVE + BASELINE
## Value Std.Error DF t-value p-value
## (Intercept) 1.7260087 0.5062630 87 3.409312 0.0010
## GROUP1 -0.3782460 0.5366099 87 -0.704881 0.4828
## WAVE 0.0518131 0.2414614 86 0.214581 0.8306
## BASELINE 0.7704099 0.0502963 86 15.317413 0.0000
## GROUP1:WAVE 0.1890996 0.3358694 86 0.563015 0.5749
## Correlation:
## (Intr) GROUP1 WAVE BASELI
## GROUP1 -0.536
## WAVE -0.715 0.675
## BASELINE -0.648 -0.017 0.000
## GROUP1:WAVE 0.518 -0.939 -0.719 -0.006
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.42041126 -0.57931853 0.04822548 0.59758321 2.83428234
##
## Number of Observations: 178
## Number of Groups: 89
Check assumptions on model without any imputations
Describe the meanRES variable by the GROUP variable
describeBy(data2[,3:4], group = data2$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 86 6.52 2.01 7.00 6.59 1.98 2.67 10 7.33 -0.38
## meanRES 2 59 6.95 1.90 7.33 7.11 0.99 0.67 10 9.33 -0.97
## kurtosis se
## BASELINE -0.94 0.22
## meanRES 0.95 0.25
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew
## BASELINE 1 88 6.73 1.68 7 6.81 1.48 2.67 10 7.33 -0.48
## meanRES 2 54 7.31 1.73 8 7.48 1.48 2.33 10 7.67 -0.86
## kurtosis se
## BASELINE -0.42 0.18
## meanRES 0.16 0.24
Create a plot that visualizes meanRES variable by the GROUP variable
library(ggplot2)
library(influence.ME)
Take a look at the residuals
residual <- lm(meanRES ~ BASELINE, data=data2)$residual
Plot the residuals to see that they are random
plot(density(residual))# A density plot
qqnorm(residual) # A quantile normal plot to checking normality
qqline(residual)
Checking the different between intervention and control groups residuals. This allows us to control for individual unsystematic differences.
data2$residual <- NA
sel1 <- which(!is.na(data2$meanRES))
sel2 <- which(!is.na(data2$BASELINE))
data2$residual[intersect(sel1,sel2)] <- residual
qplot(GROUP, meanRES, data=data2, geom="boxplot")
## Warning: Removed 65 rows containing non-finite values (stat_boxplot).
Plot of the difference between intervention and control groups.
qplot(GROUP, residual, data=data2, geom="boxplot")
## Warning: Removed 69 rows containing non-finite values (stat_boxplot).
Two way repeated measures ======================================================== Graphing the Two-Way Interaction. Both meanRES and the Residuals
# Load the nlme package
library(nlme)
with(data2, boxplot(meanRES ~ WAVE + GROUP))
with(data2, boxplot(residual ~ WAVE + GROUP))
Comparing Basline to Wave 2 and 3 by Group.
fullModel <- lme(meanRES ~ GROUP * WAVE + BASELINE, random = ~1 | ID, data = data2, method = "ML", na.action = "na.omit")
CookD(fullModel)
plot(fullModel, which="cook")
Explanation of significance:
We asses the significance of our models by comparing them from the baseline model using the anova() function.
(Intercept): Where everything is 0
GROUP1: Is there a difference between group. If it is significant than there is a difference and the treatment had an effect.
WAVE: Asseses whether the effects gets bigger beteen time 2 and 3 (does not have to be significant)
BASELINE: Should not be significant. If it is then it shows that there is a difference between groups before the treatment.
GROUP1:WAVE: If this is significant then it means that the effect was either fleeting or it happened after the treatment i.e. between time 2 and 3.
summary(fullModel)
## Linear mixed-effects model fit by maximum likelihood
## Data: data2
## AIC BIC logLik
## 355.5365 374.3759 -170.7682
##
## Random effects:
## Formula: ~1 | ID
## (Intercept) Residual
## StdDev: 0.5619731 1.027192
##
## Fixed effects: meanRES ~ GROUP * WAVE + BASELINE
## Value Std.Error DF t-value p-value
## (Intercept) 1.0216798 0.6513498 66 1.568558 0.1215
## GROUP1 -0.2956335 0.6435147 66 -0.459404 0.6475
## WAVE 0.1958078 0.2882305 38 0.679345 0.5010
## BASELINE 0.8389021 0.0733463 66 11.437555 0.0000
## GROUP1:WAVE 0.2415480 0.4232303 38 0.570725 0.5715
## Correlation:
## (Intr) GROUP1 WAVE BASELI
## GROUP1 -0.406
## WAVE -0.602 0.625
## BASELINE -0.744 -0.061 -0.018
## GROUP1:WAVE 0.402 -0.922 -0.681 0.023
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -3.3505420 -0.3916260 0.0791085 0.4685979 2.7998940
##
## Number of Observations: 109
## Number of Groups: 69
Table with P-value
| | Value| Std.Error| DF| t-value| p-value|
|:------------|-----------:|----------:|---:|-----------:|----------:|
|(Intercept) | 1.0216798| 0.6513498| 66| 1.5685578| 0.1215333|
|GROUP1 | -0.2956335| 0.6435147| 66| -0.4594044| 0.6474534|
|WAVE | 0.1958078| 0.2882305| 38| 0.6793447| 0.5010368|
|BASELINE | 0.8389021| 0.0733463| 66| 11.4375551| 0.0000000|
|GROUP1:WAVE | 0.2415480| 0.4232303| 38| 0.5707248| 0.5715460|
Table with confidence intervals
| est. | lower | upper | |
|---|---|---|---|
| (Intercept) | 1.0216798 | -0.2486048 | 2.2919644 |
| GROUP1 | -0.2956335 | -1.5506378 | 0.9593708 |
| WAVE | 0.1958078 | -0.3741443 | 0.7657600 |
| BASELINE | 0.8389021 | 0.6958596 | 0.9819445 |
| GROUP1:WAVE | 0.2415480 | -0.5953552 | 1.0784513 |