plotor

The goal of plotor is to generate Odds Ratio plots from logistic regression models.

Installation

You can install the development version of plotor from GitHub with:

# install.packages("devtools")
devtools::install_github("craig-parylo/plotor")

You can also install the latest released version from CRAN with:

install.packages("plotor")

Example

In this example we will explore the likelihood of surviving the Titanic disaster based on passenger economic status (class), sex, and age group.

In addition to plotor the packages we will use include dplyr, tidyr and forcats for general data wrangling, the stats package to conduct the logistic regression followed by broom to tidy the output and convert the results to Odds Ratios and confidence intervals, then ggplot2 to visualise the plot.

library(plotor)      # generates Odds Ratio plots
library(datasets)    # source of example data
library(dplyr)       # data wrangling
library(tidyr)       # data wrangling - uncounting aggregated data
library(forcats)     # data wrangling - handling factor variables
library(stats)       # perform logistic regression using glm function
library(broom)       # tidying glm model and producing OR and CI
library(ggplot2)     # data visualisation

Start with getting the data from the datasets package.

df <- datasets::Titanic |> 
  as_tibble() |> 
  # convert counts to observations
  filter(n > 0) |>
  uncount(weights = n) |>
  # convert categorical variables to factors.
  # we specify an order for levels in Class and Survival, otherwise ordering
  # in descending order of frequency
  mutate(
    Class = Class |>
      fct(levels = c('1st', '2nd', '3rd', 'Crew')),
    Sex = Sex |>
      fct_infreq(),
    Age = Age |>
      fct_infreq(),
    Survived = Survived |>
      fct(levels = c('No', 'Yes'))
  )

We now have a tibble of data containing four columns:

Survived - our outcome variable describing whether the passenger survived Yes or died No,
Class - the passenger class, either 1st, 2nd, 3rd or Crew,
Sex - the gender of the passenger, either Male or Female,
Age - whether the passenger was an Adult or Child.

We next conduct a logistic regression of survival (as a binary factor: ‘yes’ and ‘no’) against the characteristics of passenger class, sex and age group. For this we use the Generalised Linear Model function (glm) from the stats package, specifying:

the family as ‘binomial’, and
the formula as survival being a function of Class, Sex and Age.

# conduct a logistic regression of survival against the other variables
lr <- glm(
  data = df, 
  family = 'binomial', 
  formula = Survived ~ Class + Sex + Age
  )

Finally, we can plot the Odds Ratio of survival using the plot_or function.

# using plot_or 
plot_or(glm_model_results = lr)

This plot makes it clear that:

Children were 2.89 times more likely to survive than Adults,
Passengers in 2nd, 3rd class as well as Crew were all less likely to survive than those in 1st class,
Women were 11.25 times more likely to survive than men.

Table outputs

While an odds ratio plot can effectively visualise the direction and magnitude of relationships, a table of results offers additional information such as the value of the odds ratio, p-values and confidence intervals.

Use the table_or() function to return a tibble of details about our Titanic analysis:

# using table_or
table_or(glm_model_results = lr)
#> # A tibble: 8 × 14
#>   label level   rows outcome outcome_rate class  estimate std.error statistic
#>   <fct> <fct>  <int>   <int>        <dbl> <chr>     <dbl>     <dbl>     <dbl>
#> 1 Class 1st      325     203        0.625 factor   NA        NA         NA   
#> 2 Class 2nd      285     118        0.414 factor    0.361     0.196     -5.19
#> 3 Class 3rd      706     178        0.252 factor    0.169     0.172    -10.4 
#> 4 Class Crew     885     212        0.240 factor    0.424     0.157     -5.45
#> 5 Sex   Male    1731     367        0.212 factor   NA        NA         NA   
#> 6 Sex   Female   470     344        0.732 factor   11.2       0.140     17.2 
#> 7 Age   Adult   2092     654        0.313 factor   NA        NA         NA   
#> 8 Age   Child    109      57        0.523 factor    2.89      0.244      4.35
#> # ℹ 5 more variables: p.value <dbl>, conf.low <dbl>, conf.high <dbl>,
#> #   significance <chr>, comparator <dbl>

You can also output these details into a formatted table complete with a mini OR plot, which is ideal for inclusion in reports and publications. To do this, add output = "gt" as part of the table_or() function call.

Assumption checks

New to plotor is a new suite of automated checks. These checks verify the data used in your logistic regression analysis upholds the required assumptions, providing an added layer of confidence in your results.

Assumptions for logistic regression
Assumption	Description	Status
The outcome variable is binary	`plotor` is designed to work with an outcome variable that has only two possible values, i.e. outcome is binary.	✅ Introduced in PR 42
The predictor variables should not be highly correlated with each other	Predictor variables which have high levels of correlation with each other is known as multicollinearity. Where this is the case the odds ratio estimates are likely to be unstable, confidence intervals are likely to be much larger, both of which make it difficult to interpret the results.	✅ Introduced in PR 43
The outcome is not separated by predictors	In logistic regression, separation occurs when a predictor variable (or a combination of predictor variables) perfectly predicts the outcome variable. Separation results in infinite or extremely large odds ratios and possibly issues with non-convergence of the logistic regression model, making it difficult for the model to estimate the coefficients.	✅ Introduced in PR 47
The sample size is large enough	The sample size should be large enough to provide reliable estimates of the odds ratio. A general rule of thumb is to have at least 10 events (or outcomes of interest) per predictor variable.	In development
The observations are independent	Each observation should be independent of the others. This means that the outcome for one observation should not be influenced by the outcome of another observation.	In development
There are no extreme outlier values	The data should not contain outliers or influential observations that can significantly affect the estimates of the odds ratio.	In development
There is a linear relationship between the predictors and the logit	The relationship between the predictor variables and the log odds of the outcome should be linear. This can be checked using diagnostic plots, such as the logit plot.	In development

README

plotor

Installation

Example

Table outputs

Assumption checks