chisq_test_vs_external_estimate {nrba}

Test of differences in survey percentages relative to external estimates

Description

Compare estimated percentages from the present survey to external estimates from a benchmark source. A Chi-Square test with Rao-Scott's second-order adjustment is used to evaluate whether the survey's estimates differ from the external estimates.

Usage

chisq_test_vs_external_estimate(survey_design, y_var, ext_ests, na.rm = TRUE)

Arguments

survey_design

A survey design object created with the survey package.

y_var

Name of dependent categorical variable.

ext_ests

A numeric vector containing the external estimate of the percentages for each category. The vector must have names, each name corresponding to a given category.

na.rm

Whether to drop cases with missing values

Details

Please see svygofchisq for details of how the Rao-Scott second-order adjusted test is conducted. The test statistic, statistic is obtained by calculating the Pearson Chi-squared statistic for the estimated table of population totals. The reference distribution is a Satterthwaite approximation. The p-value is obtained by comparing statistic/scale to a Chi-squared distribution with df degrees of freedom.

Value

A data frame containing the results of the Chi-Square test(s) of whether survey-based estimates systematically differ from external estimates.

The columns of the output dataset include:

• statistic: The value of the test statistic
  

• df: Degrees of freedom for the reference Chi-Squared distribution
  

• scale: Estimated scale parameter.

• p_value: The p-value of the test of independence
  

• test_method: Text giving the name of the statistical test

• variance_method: Text describing the method of variance estimation

References

• Rao, JNK, Scott, AJ (1984) "On Chi-squared Tests For Multiway Contigency Tables with Proportions Estimated From Survey Data" Annals of Statistics 12:46-60.
  

Examples

library(survey)

# Create a survey design ----
data("involvement_survey_pop", package = "nrba")
data("involvement_survey_str2s", package = "nrba")

involvement_survey_sample <- svydesign(
  data = involvement_survey_str2s,
  weights = ~BASE_WEIGHT,
  strata = ~SCHOOL_DISTRICT,
  ids = ~ SCHOOL_ID + UNIQUE_ID,
  fpc = ~ N_SCHOOLS_IN_DISTRICT + N_STUDENTS_IN_SCHOOL
)

# Subset to only include survey respondents ----

involvement_survey_respondents <- subset(
  involvement_survey_sample,
  RESPONSE_STATUS == "Respondent"
)

# Test whether percentages of categorical variable differ from benchmark ----

parent_email_benchmark <- c(
  "Has Email" = 0.85,
  "No Email" = 0.15
)

chisq_test_vs_external_estimate(
  survey_design = involvement_survey_respondents,
  y_var = "PARENT_HAS_EMAIL",
  ext_ests = parent_email_benchmark
)