quest (version 0.2.0)

props_test: Test for Multiple Sample Proportion Against Pi (Chi-square Tests of Goodness of Fit)

Description

props_test tests for multiple sample proportion difference from population proportions with chi-square tests of goodness of fit. The default is that the goodness of fit is consistent with a population proportion Pi of 0.50. The function also calculates the descriptive statistics, various standardized effect sizes (e.g., Cramer's V), and can provide the 1x2 contingency tables. props_test is simply a wrapper for prop.test plus some extra calculations.

Usage

props_test(
  data,
  dum.nm,
  pi = 0.5,
  yates = TRUE,
  ci.level = 0.95,
  rtn.table = TRUE,
  check = TRUE
)

Value

list of data.frames containing statistical information about the proportion differences from pi: 1) nhst = chi-square test of goodness of fit stat info in a data.frame, 2) desc = descriptive statistics stat info in a data.frame, 3) std = various standardized effect sizes in a data.frame, 4) count = data.frame containing the rbinded 1x2 tables of counts with an additional column for the total (if rtn.table = TRUE), 5) percent = data.frame containing the rbinded 1x2 tables of overall percentages with an additional column for the total (if rtn.table = TRUE)

1) nhst = chi-square test of goodness of fit stat info in a data.frame

est

proportion difference estimate (i.e., sample proportion - pi)

se

NA (to remind the user there is no standard error for the test)

X2

chi-square value

df

degrees of freedom (will always be 1)

p

two-sided p-value

2) desc = descriptive statistics stat info in a data.frame

prop

sample proportion

pi

popularion proportion provided by the user (or 0.50 by default)

sd

standard deviation

n

sample size

lwr

lower bound of the confidence interval of the sample proportion itself

upr

upper bound of the confidence interval of the sample proportion itself

3) std = various standardized effect sizes in a data.frame

cramer

Cramer's V estimate

h

Cohen's h estimate

4) count = data.frame containing the rbinded 1x2 tables of counts with an additional column for the total (if rtn.table = TRUE). The colnames are 1. "0", 2. "1", 3. "total"

5) percent = data.frame containing the rbinded 1x2 tables of overall percentages with an additional column for the total (if rtn.table = TRUE). The colnames are 1. "0", 2. "1", 3. "total"

Arguments

data

data.frame of data.

dum.nm

character vector of length 1 specifying the colnames in data of the variables used to calculate the proportions. The variables must only have values of 0 or 1 (or missing values), or be otherwise known as dummy variables. See is.dummy.

pi

numeric vector of length = length(dum.nm) or length 1 specifying the population proportion values to compare the sample proportions against. The order of the values should be the same as the order in dum.nm. When length 1, the same population proportion value is used for all the variables.

yates

logical vector of length 1 specifying whether the Yate's continuity correction should be applied for small samples. See chisq.test for details.

ci.level

numeric vector of length 1 specifying the confidence level. ci.level must range from 0 to 1.

rtn.table

logical vector of lengh 1 specifying whether the return object should include the rbinded 1x2 contingency table of counts with totals and the rbinded 1x2 overall percentages table. If TRUE, then the last two elements of the return object are "count" containing a data.frame of counts and "percent" containing a data.frame of overall percentages.

check

logical vector of length 1 specifying whether the input arguments should be checked for errors. For example, if data[dum.nm] are all dummy variables that only take on values of 0 or 1 (or missing values). This is a tradeoff between computational efficiency (FALSE) and more useful error messages (TRUE).

See Also

prop.test the workhorse for prop_test, prop_test for a single dummy variables, props_diff for chi-square tests of independence,

Examples

# multiple variables
mtcars2 <- mtcars
mtcars2$"gear_dum" <- ifelse(mtcars2$"gear" > 3, yes = 1L, no = 0L)
mtcars2$"carb_dum" <- ifelse(mtcars2$"carb" > 3, yes = 1L, no = 0L)
vrb_nm <- c("am","gear_dum","carb_dum") # dummy variables
lapply(X = vrb_nm, FUN = function(nm) {
   table(mtcars2[nm])
})
props_test(data = mtcars2, dum.nm = c("am","gear_dum","carb_dum"))
props_test(data = mtcars2, dum.nm = c("am","gear_dum","carb_dum"),
   rtn.table = FALSE)

# single variable
props_test(data = mtcars2, dum.nm = "am")
props_test(data = mtcars2, dum.nm = "am", rtn.table = FALSE)

# error from non-dummy variables
if (FALSE) {
props_test(data = mtcars2, dum.nm = c("am","gear","carb"))
}