ff_se_prop: Standard error for derived proportions

ff_se_propR Documentation

Standard error for derived proportions

Description

This function calculates the standard error for derived proportions. The numerator must be a subset of the denominator. For example, the numerator might be the population of african-american males with a college degree, while the denominator is the population of african american males. Use 'ff_se_ratio' if the numerator is not a subset of the denominator.

Usage

ff_se_prop(estimate_num, estimate_den, se_num, se_den)

Arguments

estimate_num

The numerator for the estimated proportion

estimate_den

The denominator for the estimated proprotion

se_num

Standard error for the numerator of the derived proportions.

se_den

Standard error for the denominator of the derived proportions

Details

The function will fail if the value under the square root is negative. In such a case, users will see the following message: 'Warning message: In sqrt(...) : NaNs produced.'

If this occurs, use 'ff_se_ratio'.

Reference for calculations: US Census Bureau, A Compass for Understanding and Using ACS Data, October 2008, A-14

Value

The standard error of the derived proportions

Examples

df <- data.frame(estimate_numerator = rnorm(n = 10, mean = 2000, sd = 500),
                estimate_denominator = rnorm(n = 10, mean = 10000, sd = 2000),
                se_numerator = rnorm(n = 10, mean = 200, sd = 50),
                se_denominator = rnorm(n = 10, mean = 100, sd = 10))

# calculate standard errors of proportions for each observation
ff_se_prop(df$estimate_numerator, df$estimate_denominator,
           df$se_numerator, df$se_denominator)

# add columns to dataframe showing proportions and standard errors of proportions
dplyr::mutate(df,
              proportion = estimate_numerator / estimate_denominator,
              se_proportion = ff_se_prop(estimate_numerator, estimate_denominator,
                                         se_numerator, se_denominator))


forsythfuture/FFtools documentation built on April 5, 2022, 10:02 p.m.