rm_ordsum: Fit and format an ordinal logistic regression using polr from...

In Avery-Lisa/myReportRx: Adaptation on ReportRx for Markdown

Description

Fit and format an ordinal logistic regression using polr from the MASS package. The parallel regression assumption can be tested using the Brant test in the Brant package and visually. Only logistic ordinal regression is supported currently.

Usage

rm_ordsum(
  data,
  covs,
  response,
  reflevel,
  caption = NULL,
  showN = T,
  testPO = TRUE,
  digits = 2,
  CIwidth = 0.95,
  excludeLevels
)

Arguments

data	dataframe containing data [REQUIRED]
covs	character vector with the names of columns to include in table [REQUIRED]
response	ordinal outcome variable [REQUIRED]
reflevel	manual specification of the reference level, must match level exactly
caption	Table caption
showN	logical, should sample sizes be shown for each lvel, defaults to TRUE
testPO	logical, should the proportional odds (parallel regression) assumption be tested with the Brant test, defaults to TRUE
digits	number of digits to display, defaults to
CIwidth	level of significance for computing the confidence intervals, default is 0.95
excludeLevels	a named list of levels to exclude from the response variable

Value

A formatted table displaying the odds ratio associated with each covariate

Avery-Lisa/myReportRx documentation built on May 23, 2021, 2:30 a.m.