Prepare SaTScan Population File

Usage

prep_pop(df, loc_id, time, pop, covars = NULL)

Arguments

df

Data frame containing population data.

loc_id

Column name for location ID (unquoted).

time

Column name for time/year (unquoted). Population times are anchors.

pop

Column name for population count (unquoted).

covars

Character vector of covariate column names.

Value

A satscan_table object of kind "pop".

Details

Prepares a population file for SaTScan, used as the denominator for Poisson models. Population data is treated as "census anchors" - SaTScan interpolates between these time points.

SaTScan File Specification

The Population File has the following structure: <LocationID> <Year/Time> <Population> <Covariate1> ...

• LocationID: Character or numeric identifier. Match cases and geo.

• Year/Time: The "census year" or time point. For daily analysis, you can still use yearly population anchors (e.g. 2023, 2024). SaTScan interpolates linearly.

• Population: The count of people at risk.

• Covariates: Optional. If used, population must be stratified by these covariates (i.e., you need a row for every combination of Location, Time, and Covariate levels).

See also

ss_pop() for the new ss_tbl interface.

Examples

if (FALSE) { # \dontrun{
# Basic yearly population
head(pop_data)
#   county year  pop
# 1    001 2020 5000
# 2    001 2021 5100

pop_obj <- prep_pop(pop_data, loc_id = county, time = year, pop = pop)

# Stratified by sex
head(pop_strat)
#   county year  pop sex
# 1    001 2020 2500   M
# 2    001 2020 2500   F

pop_obj2 <- prep_pop(pop_strat, loc_id = county, time = year, pop = pop, covars = "sex")
} # }