Computes the post-stratification factors natively in base R: each
unit in stratum \(h\) gets \((N_h / N) / (n_h / n)\), the ratio of
the population share to the sample share (the adjustment
survey::postStratify() applies to equal starting weights). No package
is called.
Usage
morie_survey_poststratify(df, strata_col, population_counts)
Arguments
- df
A data.frame holding the (unit-level) sample data plus
any weight/strata/cluster/domain columns referenced by name.
- strata_col
Character; column name of the stratum identifier
in df.
- population_counts
Named numeric vector of population counts
by stratum (post-stratification target).
Value
A numeric vector of post-stratification weights.
Examples
set.seed(1)
df <- data.frame(g = rep(c("a", "b"), 15))
str(morie_survey_poststratify(df, "g", c(a = 60, b = 40)), max.level = 1)
#> num [1:30] 1.2 0.8 1.2 0.8 1.2 0.8 1.2 0.8 1.2 0.8 ...
