Description Usage Arguments Value See Also Examples
Calculates the value of a t statistic that has been adjusted for a cluster randomized design. See the vignette "Construction of the library functions" for a description of the assumptions and construction of this t statistic.
1 | t_test_clustered_stat(data_experiment)
|
data_experiment |
a dataframe with three columns: group, cluster and
response. The response column holds the values of the measured responses. The
group column assigns a group number (must be either 1 or 2) to each
response. The cluster column assigns a cluster number to each response.
Each different cluster must have a different number, even if they are in
different groups. See the examples section for an example of a valid
|
a number - the value of the cluster adjusted t statistic.
t_test_clustered_pval
for p value of the cluster
adjusted t test, power_t_test_clustered
for power
calculation of the cluster adjusted t test and
simulate_power_t_test_clustered
for power simulation of
the cluster adjusted t test.
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 | #***************************************************************************
# a valid instance of a data_experiment.
# * this example fits with the underlying model - see XXXX - only in the
# uninteresting case, where the cluster correllation coefficient and the
# treatment effect are both 0
# * although this construction orders the rows of the data.frame according
# to groups and clusters, this ordering does not have to be present for the
# data.frame to be a valid data_experiment
clusters_group_1 <- c(60,50,55)
clusters_group_2 <- c(50,45,50,55)
clusters <- c(clusters_group_1,clusters_group_2)
group_1_size <- sum(clusters_group_1)
group_2_size <- sum(clusters_group_2)
total_size <- group_1_size+group_2_size
group <- as.factor(c(rep(1,group_1_size),rep(2,group_2_size)))
cluster <- vector()
for (i in 1:length(clusters)) {
cluster <- c(cluster,rep(i,clusters[i]))
}
cluster <- as.factor(cluster)
response <- rnorm(total_size)
test_data <- data.frame(group,cluster,response)
#***************************************************************************
#***************************************************************************
# the value of the cluster adjusted t statistic on the data_experiment
# constructed above
# * (since the intracluster correlation coefficient is 0 for this
# data_experiment, the cluster adjusted t statistic is actually
# unnecessary here...)
t_test_clustered_stat(data_experiment=test_data)
#***************************************************************************
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.