Chapter 14 chi-square and contingency-table functions — chisq

chisq_gof_probs() performs a one-way chi-square goodness-of-fit test to specified category probabilities.

Usage

chisq_gof_probs(
  observed,
  p = NULL,
  labels = NULL,
  min_expected = 5,
  alpha = 0.05,
  digits = 4,
  quiet = FALSE
)

chisq_gof_dist(
  x = NULL,
  observed = NULL,
  dist = c("exp", "norm", "pois"),
  k = NULL,
  breaks = NULL,
  params = NULL,
  estimate = FALSE,
  params_estimated = FALSE,
  min_expected = 5,
  alpha = 0.05,
  digits = 4,
  quiet = FALSE
)

chisq_table(
  observed,
  type = c("independence", "homogeneity"),
  correct = FALSE,
  min_expected = 5,
  alpha = 0.05,
  digits = 4,
  quiet = FALSE
)

table_props(
  observed,
  margin = c("row", "col", "overall"),
  digits = 4,
  quiet = FALSE
)

Arguments

observed: Observed counts. For chisq_gof_probs() and grouped chisq_gof_dist() inputs, use a nonnegative integer vector. For chisq_table() and table_props(), use a nonnegative two-way table or matrix.
p: Probabilities under the null for chisq_gof_probs(). If NULL, equal probabilities are assumed.
labels: Optional category labels for one-way grouped inputs.
min_expected: Minimum expected count threshold used for warnings and, where implemented, class combining.
alpha: Significance level in (0, 1).
digits: Integer number of decimal places used only for printed output.
quiet: Logical; if TRUE, suppress printed output.
x: Raw sample data for chisq_gof_dist().
dist: Distribution for chisq_gof_dist(): "pois", "norm", or "exp".
k: Number of classes used for raw-data GOF tests when breaks is not supplied.
breaks: Class boundaries for grouped chisq_gof_dist() input.
params: Named list of distribution parameters when parameters are not estimated from raw data.
estimate: Logical; if TRUE, estimate parameters from raw data where applicable.
params_estimated: Logical; for grouped GOF inputs, whether supplied parameters should count against the degrees of freedom.
type: For chisq_table(), whether the interpretation is "independence" or "homogeneity".
correct: Logical; if TRUE, apply Yates correction in the 2 x 2 case for chisq_table().
margin: For table_props(), one of "row", "col", or "overall".

Value

For the chi-square test functions, a classed htest_result object with statistic, df, p-value, critical value, rejection region, decision, and diagnostic outputs such as observed counts, expected counts, contributions, and residuals. For chisq_table(), standardized residuals are also returned.

For table_props(), a classed table_props_result object is returned with the original table and the requested proportions.

Details

chisq_gof_dist() performs a chi-square goodness-of-fit test to a distribution, supporting at least Poisson, normal, and exponential models.

chisq_table() performs a chi-square test for independence or homogeneity in an r x c contingency table.

table_props() computes descriptive row, column, or overall proportions for a contingency table.

Functions

chisq_gof_probs(): One-way chi-square GOF test to specified probabilities.
chisq_gof_dist(): Chi-square GOF test to a distribution.
chisq_table(): Chi-square test for a contingency table.
table_props(): Descriptive proportions for a contingency table.