Table OR — table_or • plotor

Produces a formatted table showing the outputs from the Odds Ratio analysis, including details on covariate characteristics and model results.

Usage

table_or(
  glm_model_results,
  conf_level = 0.95,
  output = "tibble",
  confint_fast_estimate = FALSE
)

Arguments

glm_model_results: Results from a binomial Generalised Linear Model (GLM), as produced by stats::glm().
conf_level: Numeric value between 0.001 and 0.999 (default = 0.95) specifying the confidence level for the confidence interval.
output: String describing of the output type (default = 'tibble'). Options include 'tibble' and 'gt'.
confint_fast_estimate: Boolean (default = FALSE) indicating whether to use a faster estimate of the confidence interval. Note: this assumes normally distributed data, which may not be suitable for your data.

Value

The returned object depends on the output parameter:

If output = 'tibble', the function returns an object of class "tbl_df", "tbl" and "data.frame".
If output = 'gt', the function returns an object of class "gt_tbl" and "list"

Details

The table includes the following information:

Covariate characteristics:
- Number of observations for each characteristic
- Number of observiations resulting in the outcome of interest
- Conversion rate of outcome by the number of observations
Model results:
- Estimated Odds Ratio, standard error and p-value
- Calculated confidence interval for the specified confidence level
A visualisation of the OR plot is also provided for an at-a-glance view of the model results

Includes details on the characteristics of the covariates, such as:

the number of observations for each characteristic,
the number of observations resulting in the outcome of interest,
the conversion rate of outcome by the number of observations,

Details are calculated showing the:

estimated Odds Ratio, standard error and p-value,
calculated confidence interval for the confidence level,

Also included is a visualisation of the OR plot to provide an at-a-glance view of the model results.

Examples

# Load the Titanic dataset
df <- datasets::Titanic |>
  dplyr::as_tibble() |>
  # convert aggregated counts to individual observations
  dplyr::filter(n > 0) |>
  tidyr::uncount(weights = n) |>
  # convert character variables to factors
  dplyr::mutate(dplyr::across(dplyr::where(is.character), as.factor))

# Perform logistic regression using `glm`
lr <- stats::glm(
  data = df,
  family = 'binomial',
  formula = Survived ~ Class + Sex + Age
)

# Produce the Odds Ratio table as a tibble
table_or(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   Female   470     344        0.732 factor  NA         NA         NA   
#> 6 Sex   Male    1731     367        0.212 factor   0.0889     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>

# Produce the Odds Ratio table as a gt object
table_or(lr, output = 'gt')
#> Warning: Ignoring unknown parameters: `linewidth`
#> Warning: Ignoring unknown parameters: `linewidth`
#> Warning: Ignoring unknown parameters: `linewidth`
#> Warning: Ignoring unknown parameters: `linewidth`
#> Warning: Ignoring unknown parameters: `linewidth`


  Survived
    
Odds Ratio summary table with 95% Confidence Interval
    

      
        Characteristic¹
      
      
        Odds Ratio (OR)²
      
      
        95% Confidence Interval (CI)³
      
      OR Plot
    
Level
      N
      n
      Rate
      Class
      OR
      SE
      p
      Lower
      Upper
      Significance
    
Class
1st
325
203
62.46%
factor
—
—
—
—
—
Comparator
 
2nd
285
118
41.4%
factor
0.3613
0.1960
2.05 × 10⁻⁷
0.2453
0.5292
Significant
3rd
706
178
25.21%
factor
0.1690
0.1716
3.69 × 10⁻²⁵
0.1203
0.2358
Significant
Crew
885
212
23.95%
factor
0.4241
0.1573
5.00 × 10⁻⁸
0.3115
0.5775
Significant
Sex
Female
470
344
73.19%
factor
—
—
—
—
—
Comparator
 
Male
1,731
367
21.2%
factor
0.08892
0.1404
1.43 × 10⁻⁶⁶
0.06725
0.1166
Significant
Age
Adult
2,092
654
31.26%
factor
—
—
—
—
—
Comparator
 
Child
109
57
52.29%
factor
2.891
0.2440
1.36 × 10⁻⁵
1.792
4.671
Significant
¹ Characteristics are the explanatory variables in the logistic regression analysis. For categorical variables the first characteristic is designated as a reference against which the others are compared. For numeric variables the results indicate a change per single unit increase.
Level - the name or the description of the explanatory variable.
N - the number of observations examined.
n - the number of observations resulting in the outcome of interest.
Rate - the proportion of observations resulting in the outcome of interest (n / N).
Class - description of the data type.
    
² Odds Ratios estimate the relative odds of an outcome with reference to the Characteristic. For categorical data the first level is the reference against which the odds of other levels are compared. Numerical characteristics indicate the change in OR for each additional increase of one unit in the variable.
OR - The Odds Ratio point estimate - values below 1 indicate an inverse relationship whereas values above 1 indicate a positive relationship. Values shown to 4 significant figures.
SE - Standard Error of the point estimate. Values shown to 4 significant figures.
p - The p-value estimate based on the residual Chi-squared statistic.
    
³ Confidence Interval - the range of values likely to contain the OR in 95% of cases if this study were to be repeated multiple times. If the CI touches or crosses the value 1 then it is unlikely the Characteristic is significantly associated with the outcome.
Lower & Upper - The range of values comprising the CI, shown to 4 significant figures.
Significance - The statistical significance indicated by the CI, Significant where the CI does not touch or cross the value 1.

	Characteristic¹					Odds Ratio (OR)²			95% Confidence Interval (CI)³			OR Plot
Survived
Odds Ratio summary table with 95% Confidence Interval
	Level	N	n	Rate	Class	OR	SE	p	Lower	Upper	Significance	OR Plot
Class	1st	325	203	62.46%	factor	—	—	—	—	—	Comparator
	2nd	285	118	41.4%	factor	0.3613	0.1960	2.05 × 10⁻⁷	0.2453	0.5292	Significant
	3rd	706	178	25.21%	factor	0.1690	0.1716	3.69 × 10⁻²⁵	0.1203	0.2358	Significant
	Crew	885	212	23.95%	factor	0.4241	0.1573	5.00 × 10⁻⁸	0.3115	0.5775	Significant
Sex	Female	470	344	73.19%	factor	—	—	—	—	—	Comparator
Sex	Male	1,731	367	21.2%	factor	0.08892	0.1404	1.43 × 10⁻⁶⁶	0.06725	0.1166	Significant
Age	Adult	2,092	654	31.26%	factor	—	—	—	—	—	Comparator
Age	Child	109	57	52.29%	factor	2.891	0.2440	1.36 × 10⁻⁵	1.792	4.671	Significant
¹ Characteristics are the explanatory variables in the logistic regression analysis. For categorical variables the first characteristic is designated as a reference against which the others are compared. For numeric variables the results indicate a change per single unit increase. Level - the name or the description of the explanatory variable. N - the number of observations examined. n - the number of observations resulting in the outcome of interest. Rate - the proportion of observations resulting in the outcome of interest (n / N). Class - description of the data type.
² Odds Ratios estimate the relative odds of an outcome with reference to the Characteristic. For categorical data the first level is the reference against which the odds of other levels are compared. Numerical characteristics indicate the change in OR for each additional increase of one unit in the variable. OR - The Odds Ratio point estimate - values below 1 indicate an inverse relationship whereas values above 1 indicate a positive relationship. Values shown to 4 significant figures. SE - Standard Error of the point estimate. Values shown to 4 significant figures. p - The p-value estimate based on the residual Chi-squared statistic.
³ Confidence Interval - the range of values likely to contain the OR in 95% of cases if this study were to be repeated multiple times. If the CI touches or crosses the value 1 then it is unlikely the Characteristic is significantly associated with the outcome. Lower & Upper - The range of values comprising the CI, shown to 4 significant figures. Significance - The statistical significance indicated by the CI, Significant where the CI does not touch or cross the value 1.