Description Usage Arguments Value Examples
View source: R/eefAnalyticPerm_modified_05_02_2016.R
caceMSTBoot
performs CACE analysis of multisite randomised trials.
1 | caceMSTBoot(formula, random, intervention, compliance, nBoot, data)
|
formula |
model specification of the form posttest ~ pretests+Intervention+.... the model to be analysed. It is of the form y~x1+x2+..., where y is the outcome variable and X's are the predictors. |
random |
a string variable specifying the "clustering" variable as contained in the data. This must be put between quotes. For example, "school". |
intervention |
the name of the intervention variable as appeared in formula. This must be put in quotes. For example "intervention" or "treatment" or "group". |
compliance |
percentages of sessions attended by pupils. |
nBoot |
number of bootstrap required to generate bootstrap confidence interval. This must be specified. |
data |
data frame containing the data to be analysed. |
S3 object; a list consisting of
CACE
. Estimated CACE effect size based on percentages of sessions attended by pupils.
The percentage data is converted into the following grids (0, 10, 20, 30, 40, 50, 60, 70, 80, 90)
and CACE effect size is calculated for each grid.
Compliers
. A summary table of the percentage of pupils in the intervention and control groups that
attended more than a pre-specified percentage of sessions. The values for the control group should be zeros if
there is no dilution in which a pupil or school in the control group receives intervention.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 | data(catcht)
########################################################
## MLM analysis of multisite trials + 1.96SE ##
########################################################
output1 <- mstFREQ(Posttest~ Intervention+Prettest,random="School",
intervention="Intervention",data=catcht)
### Fixed effects
beta <- output1$Beta
beta
### Effect size
ES1 <- output1$ES$Intervention1
ES1
## Covariance matrix
covParm <- output1$covParm
covParm
### random effects for schools
randOut <- output1$"SchEffects"
randOut <- randOut[order(randOut$Estimate),]
barplot(randOut$Estimate,ylab="Deviations from Overall Average",
names.arg=randOut$Schools,las=2)
###############################################
## MLM analysis of multisite trials ##
## with bootstrap confidence intervals ##
###############################################
output2 <- mstFREQ(Posttest~ Intervention+Prettest,random="School",
intervention="Intervention",nBoot=1000,data=catcht)
### Effect size
ES2 <- output2$ES
ES2
#######################################################################
## MLM analysis of mutltisite trials with permutation p-value##
#######################################################################
output3 <- mstFREQ(Posttest~ Intervention+Prettest,random="School",
intervention="Intervention",nPerm=1000,data=catcht)
#### Distribution under the null
perm <- output3$Perm
str(perm )
#### Permutation P-value using total variance
obsg <- output3$ES$Intervention1[2,1]
obsg
p_value <- ifelse(mean(perm$"Intervention1Total" > obsg)==0,"<0.001",
mean(perm$"Intervention1Total" > obsg) )
p_value
hist(perm$"Intervention1Total", breaks=40, col="white",
border="blueviolet",
xlab="Distribution Under Null Hypothesis",
main=paste("P(X|NULL)= ",p_value,sep=""),
xlim=range(c(perm$"Intervention1Total",obsg),
na.rm=TRUE))
abline(v=obsg,lwd=2,col=4)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.