calculate_prevalence_unusual_pval: calculate_prevalence_unusual_pval

In Hackout2/mapData: Tools to help with mapping outbreak data

Description

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.

Usage

calculate_prevalence_unusual_pval(data, pops = NULL, conf.level = 0.95,
  region.head = "region", scale = 1)

Arguments

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.)

Hackout2/mapData documentation built on May 6, 2019, 9:48 p.m.