# Loading the dataset that has been reset into a long version
load("/Users/levibrackman/data.test.RData")
For lme to work GROUP and ID need to be seen as factors
data.test$GROUP <- as.factor(data.test$GROUP)
data.test$ID <- as.factor(data.test$ID)
# Load the psych package
library(psych)
Describe the PERMAHAPPY variable by the GROUP variable
describeBy(data.test$PERMA17, group = data.test$GROUP)
## group: 0
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 80 6.85 1.83 7 7.06 1.48 1 10 9 -1.02 0.85 0.2
## --------------------------------------------------------
## group: 1
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 70 7.14 1.84 7.5 7.38 0.74 2 10 8 -1.07 0.76 0.22
## --------------------------------------------------------
## group: 2
## vars n mean sd median trimmed mad min max range skew kurtosis se
## 1 1 5 6.6 2.88 7 6.6 2.97 2 9 7 -0.58 -1.51 1.29
Create a plot that visualizes PERMAHAPPY variable by the GROUP variable
library(ggplot2)
##
## Attaching package: 'ggplot2'
##
## The following object is masked from 'package:psych':
##
## %+%
qplot(GROUP, PERMA17, data = data.test, geom = "boxplot")
# Load the nlme package
library(nlme)
with(data.test, boxplot(PERMA17 ~ wave + GROUP))
Graphing the Two-Way Interaction.
# Load the nlme package
library(nlme)
I am not sure if I am doing this right
baseline <- lme(PERMA17 ~ 1, random = ~1 | ID/GROUP/wave, data = data.test,
method = "ML")
PERMAHAPPYModel <- lme(PERMA17 ~ GROUP, random = ~1 | ID/GROUP/wave, data = data.test,
method = "ML")
PERMAHAPPY2Model <- lme(PERMA17 ~ GROUP + wave, random = ~1 | ID/GROUP/wave,
data = data.test, method = "ML")
fullModel <- lme(PERMA17 ~ GROUP * wave, random = ~1 | ID/GROUP/wave, data = data.test,
method = "ML")
We again the significance of our modePERMAHAPPY by comparing them from the baseline model using the anova() function.
anova(baseline, PERMAHAPPYModel, PERMAHAPPY2Model, fullModel)
## Model df AIC BIC logLik Test L.Ratio p-value
## baseline 1 5 604.8 620.0 -297.4
## PERMAHAPPYModel 2 7 608.4 629.7 -297.2 1 vs 2 0.412 0.8138
## PERMAHAPPY2Model 3 8 606.4 630.7 -295.2 2 vs 3 4.009 0.0453
## fullModel 4 10 605.8 636.3 -292.9 3 vs 4 4.556 0.1025
summary(fullModel)
## Warning: NaNs produced
## Linear mixed-effects model fit by maximum likelihood
## Data: data.test
## AIC BIC logLik
## 605.8 636.3 -292.9
##
## Random effects:
## Formula: ~1 | ID
## (Intercept)
## StdDev: 1.583
##
## Formula: ~1 | GROUP %in% ID
## (Intercept)
## StdDev: 0.0003785
##
## Formula: ~1 | wave %in% GROUP %in% ID
## (Intercept) Residual
## StdDev: 3.878e-05 0.9866
##
## Fixed effects: PERMA17 ~ GROUP * wave
## Value Std.Error DF t-value p-value
## (Intercept) 6.678 0.4255 89 15.696 0.0000
## GROUP1 -0.503 0.6206 0 -0.811 NaN
## GROUP2 -2.538 1.7401 89 -1.459 0.1482
## wave 0.074 0.2383 59 0.312 0.7564
## GROUP1:wave 0.522 0.3574 59 1.460 0.1497
## GROUP2:wave 1.786 1.0110 59 1.766 0.0825
## Correlation:
## (Intr) GROUP1 GROUP2 wave GROUP1:
## GROUP1 -0.684
## GROUP2 -0.245 0.167
## wave -0.787 0.540 0.193
## GROUP1:wave 0.535 -0.806 -0.131 -0.661
## GROUP2:wave 0.186 -0.127 -0.786 -0.236 0.156
##
## Standardized Within-Group Residuals:
## Min Q1 Med Q3 Max
## -2.65539 -0.31343 0.07236 0.37791 2.15309
##
## Number of Observations: 155
## Number of Groups:
## ID GROUP %in% ID wave %in% GROUP %in% ID
## 91 92 154