w_quantile() calculates percentiles (quantiles) using survey weights
for population-representative results. Percentiles divide your data into
equal portions – for example, the 25th percentile is the value below which
25% of your population falls. This is essential for understanding the
distribution of variables like income, age, or satisfaction scores.
Usage
w_quantile(
data,
...,
weights = NULL,
probs = c(0, 0.25, 0.5, 0.75, 1),
na.rm = TRUE
)Arguments
- data
Your survey data (a data frame or tibble)
- ...
The numeric variables you want to analyze. You can list multiple variables or use helpers like
starts_with("income")- weights
Survey weights to make results representative of your population. Without weights, you get the simple sample quantiles. Give a column name (unquoted or as a string), an expression such as
sampling_weight * 2, or a numeric vector with one weight per row.- probs
Which percentiles to calculate, as proportions between 0 and 1. Default:
c(0, 0.25, 0.5, 0.75, 1)for the minimum, 25th percentile, median, 75th percentile, and maximum. Usec(0.1, 0.5, 0.9)for deciles, orc(0.25, 0.5, 0.75)for quartiles only.- na.rm
Remove missing values before calculating? (Default: TRUE). With
FALSE, the result for a variable that contains missing values isNA(as in base R).
Value
Population-weighted quantile(s) with sample size information, including the weighted percentile values, effective sample size (effective N), and the number of valid observations used.
Details
Understanding the Results
Percentile values: The data values at each requested percentile in the weighted population. For example, if the weighted 25th percentile of income is 35,000, then 25% of the population earns less than that.
Effective N: How many independent observations your weighted data represents.
N / Missing: Valid and missing cases. With weights, both are sums of weights (displayed rounded), as SPSS reports them under
WEIGHT BY; Kish's effective N is shown bysummary().
Common percentiles and their meaning:
0% (minimum): The smallest observed value
25% (Q1): One quarter of the population falls below this value
50% (median): Half the population falls below this value
75% (Q3): Three quarters of the population falls below this value
100% (maximum): The largest observed value
When to Use This
Use w_quantile() when:
You want to know at what values the population splits into groups
You need to construct population-representative income brackets or age groups
You want to identify the median or quartiles with proper weighting
You need SPSS-compatible weighted percentile values
See also
quantile for the base R quantile function.
w_median for the weighted median (50th percentile).
w_iqr for the weighted interquartile range (Q3 - Q1).
describe for comprehensive descriptive statistics including quantiles.
Method
Unweighted values use quantile Type 6 (SPSS HAVERAGE). Weighted values
apply the HAVERAGE position to the cumulative weights with linear
interpolation. Both reproduce SPSS FREQUENCIES (with and without
WEIGHT BY); see vignette("spss-compatibility").
Other weighted_statistics:
w_iqr(),
w_kurtosis(),
w_mean(),
w_median(),
w_modus(),
w_range(),
w_sd(),
w_se(),
w_skew(),
w_var()
Examples
# Load required packages and data
library(dplyr)
data(survey_data)
# Basic weighted quantiles (0%, 25%, 50%, 75%, 100%)
survey_data %>% w_quantile(age, weights = sampling_weight)
#>
#> Weighted Quantile Statistics
#> ----------------------------
#> Weights: sampling_weight
#>
#> ---------------------------------------------------------------
#> Variable Min 25% 50% 75% Max N Missing
#> ---------------------------------------------------------------
#> age 18.000 38.000 50.000 63.000 95.000 2516 0
#> ---------------------------------------------------------------
# Custom quantiles
survey_data %>% w_quantile(income, weights = sampling_weight, probs = c(0.1, 0.5, 0.9))
#>
#> Weighted Quantile Statistics
#> ----------------------------
#> Weights: sampling_weight
#>
#> -----------------------------------------------------
#> Variable 10% 50% 90% N Missing
#> -----------------------------------------------------
#> income 2100.000 3500.000 5700.000 2201 315
#> -----------------------------------------------------
# Multiple variables
survey_data %>% w_quantile(age, income, weights = sampling_weight)
#>
#> Weighted Quantile Statistics
#> ----------------------------
#> Weights: sampling_weight
#>
#> ------------------------------------------------------------------------
#> Variable Min 25% 50% 75% Max N Missing
#> ------------------------------------------------------------------------
#> age 18.000 38.000 50.000 63.000 95.000 2516 0
#> income 800.000 2700.000 3500.000 4600.000 8000.000 2201 315
#> ------------------------------------------------------------------------
# Grouped data
survey_data %>% group_by(region) %>% w_quantile(age, weights = sampling_weight)
#>
#> Weighted Quantile Statistics
#> ----------------------------
#> Weights: sampling_weight
#>
#> Group: region = East
#> --------------------
#>
#> --------------------------------------------------------------
#> Variable Min 25% 50% 75% Max N Missing
#> --------------------------------------------------------------
#> age 18.000 40.000 53.000 64.000 95.000 509 0
#> --------------------------------------------------------------
#>
#> Group: region = West
#> --------------------
#>
#> ---------------------------------------------------------------
#> Variable Min 25% 50% 75% Max N Missing
#> ---------------------------------------------------------------
#> age 18.000 38.000 49.000 62.000 95.000 2007 0
#> ---------------------------------------------------------------
# Unweighted (for comparison)
survey_data %>% w_quantile(age)
#>
#> Quantile Statistics
#> -------------------
#>
#> ---------------------------------------------------------------
#> Variable Min 25% 50% 75% Max N Missing
#> ---------------------------------------------------------------
#> age 18.000 38.000 50.000 62.000 95.000 2500 0
#> ---------------------------------------------------------------