Standard error of a proportion

sqrt(p * (1 - p) / n), the sampling error of a percentage. Packages the inline proportion-error calculation from the report appendix.

Usage

se_prop(p, n, pctp = FALSE)

Arguments

p

The proportion, either as a fraction in [0, 1] or as a percentage in (1, 100] (values above 1 are divided by 100 automatically).

n

The sample size (number of respondents).

pctp

Return the answer in percentage points instead of on the [0, 1] proportion scale. Defaults to FALSE.

Value

The standard error of the proportion, on the same [0, 1] scale as a fractional p, or in percentage points when pctp = TRUE.

Details

For a percentage from a survey, the sampling error is sqrt(p * (1 - p) / n). It is largest when p = 0.5 (a 50/50 split is the hardest to pin down) and shrinks towards the extremes. The result is on the same 0–1 scale as a fractional p; pass pctp = TRUE for percentage points, the form used in report footnotes and the one margin_of_error() needs if you want the margin in points. Inputs above 1 are treated as percentages and divided by 100, so se_prop(33, n) and se_prop(0.33, n) agree.

See also

se_mean(), rse(), margin_of_error(), diagnose().

Other diagnostics: diagnose(), margin_of_error(), precision_summary(), rse(), rse_rating(), se_mean()

Examples

se_prop(0.33, 1184)
#> [1] 0.01366528

se_prop(0.33, 1184, pctp = TRUE)
#> [1] 1.366528