Testing if any regions in the dataset have a significantly different prevalence than the overall mean using leave-one-out crossvaliation. This means that the overall mean is re-calculated for each region, leaving out the data from the region in question. If population sizes for the regions are supplied, Fisher's exact test is used to calculate the p-value. If no population data is supplied, as null hypothesis Poisson distributed number of cases per region is assumed. P-values are corrected for multiple testing using the Bonferroni correction.
1 2 | calculate_prevalence_unusual_pval(data, pops = NULL, conf.level = 0.95,
region.head = "region", scale = 1)
|
data |
a dataframe containing the number of cases and total population for all regions in the dataset. |
pops |
dataframe containing the region ID in the first and the population size for each region in the dataset in the second column |
conf.level |
Confidence level to be used for calculating the confidence intervals on the prevalence estimates. |
region.head |
variable name of the incidence column in data. |
scale |
Scaling with which to report prevalence (per head, per 100 000, etc.) |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.