Tidy summarizes information about the components of a model. A model component might be a single term in a regression, a single hypothesis, a cluster, or a class. Exactly what tidy considers to be a model component varies across models but is usually self-evident. If a model has several distinct types of components, you will need to specify which components to return.

# S3 method for polr
tidy(
x,
conf.int = FALSE,
conf.level = 0.95,
exponentiate = FALSE,
p.values = FALSE,
...
)

## Arguments

x A polr object returned from MASS::polr(). Logical indicating whether or not to include a confidence interval in the tidied output. Defaults to FALSE. The confidence level to use for the confidence interval if conf.int = TRUE. Must be strictly greater than 0 and less than 1. Defaults to 0.95, which corresponds to a 95 percent confidence interval. Logical indicating whether or not to exponentiate the the coefficient estimates. This is typical for logistic and multinomial regressions, but a bad idea if there is no log or logit link. Defaults to FALSE. Logical. Should p-values be returned, based on chi-squared tests from MASS::dropterm(). Defaults to FALSE. Additional arguments. Not used. Needed to match generic signature only. Cautionary note: Misspelled arguments will be absorbed in ..., where they will be ignored. If the misspelled argument has a default value, the default value will be used. For example, if you pass conf.lvel = 0.9, all computation will proceed using conf.level = 0.95. Additionally, if you pass newdata = my_tibble to an augment() method that does not accept a newdata argument, it will use the default value for the data argument.

## Details

In broom 0.7.0 the coefficient_type column was renamed to coef.type, and the contents were changed as well. Now the contents are coefficient and scale, rather than coefficient and zeta.

Calculating p.values with the dropterm() function is the approach suggested by the MASS package author https://r.789695.n4.nabble.com/p-values-of-plor-td4668100.html. This approach is computationally intensive so that p.values are only returned if requested explicitly. Additionally, it only works for models containing no variables with more than two categories. If this condition is not met, a message is shown and NA is returned instead of p-values.

tidy, MASS::polr()

Other ordinal tidiers: augment.clm(), augment.polr(), glance.clmm(), glance.clm(), glance.polr(), glance.svyolr(), tidy.clmm(), tidy.clm(), tidy.svyolr()

## Value

A tibble::tibble() with columns:

conf.high

Upper bound on the confidence interval for the estimate.

conf.low

Lower bound on the confidence interval for the estimate.

estimate

The estimated value of the regression term.

p.value

The two-sided p-value associated with the observed statistic.

statistic

The value of a T-statistic to use in a hypothesis that the regression term is non-zero.

std.error

The standard error of the regression term.

term

The name of the regression term.

## Examples


library(MASS)

fit <- polr(Sat ~ Infl + Type + Cont, weights = Freq, data = housing)

tidy(fit, exponentiate = TRUE, conf.int = TRUE)
#>
#> Re-fitting to get Hessian#> # A tibble: 8 × 7
#>   term          estimate std.error statistic conf.low conf.high coef.type
#>   <chr>            <dbl>     <dbl>     <dbl>    <dbl>     <dbl> <chr>
#> 1 InflMedium       1.76     0.105       5.41    1.44      2.16  coefficient
#> 2 InflHigh         3.63     0.127      10.1     2.83      4.66  coefficient
#> 3 TypeApartment    0.564    0.119      -4.80    0.446     0.712 coefficient
#> 4 TypeAtrium       0.693    0.155      -2.36    0.511     0.940 coefficient
#> 5 TypeTerrace      0.336    0.151      -7.20    0.249     0.451 coefficient
#> 6 ContHigh         1.43     0.0955      3.77    1.19      1.73  coefficient
#> 7 Low|Medium       0.609    0.125      -3.97   NA        NA     scale
#> 8 Medium|High      2.00     0.125       5.50   NA        NA     scale
glance(fit)
#> # A tibble: 1 × 7
#>     edf logLik   AIC   BIC deviance df.residual  nobs
#>   <int>  <dbl> <dbl> <dbl>    <dbl>       <int> <int>
#> 1     8 -1740. 3495. 3539.    3479.        1673  1681augment(fit, type.predict = "class")
#> # A tibble: 72 × 6
#>    Sat    Infl   Type      Cont  (weights) .fitted
#>    <ord>  <fct>  <fct>     <fct>       <int> <fct>
#>  1 Low    Low    Tower     Low            21 Low
#>  2 Medium Low    Tower     Low            21 Low
#>  3 High   Low    Tower     Low            28 Low
#>  4 Low    Medium Tower     Low            34 High
#>  5 Medium Medium Tower     Low            22 High
#>  6 High   Medium Tower     Low            36 High
#>  7 Low    High   Tower     Low            10 High
#>  8 Medium High   Tower     Low            11 High
#>  9 High   High   Tower     Low            36 High
#> 10 Low    Low    Apartment Low            61 Low
#> # … with 62 more rows
fit2 <- polr(factor(gear) ~ am + mpg + qsec, data = mtcars)
tidy(fit, p.values = TRUE)
#>
#> Re-fitting to get Hessian#> p-values can presently only be returned for models that contain
#>               no categorical variables with more than two levels#> # A tibble: 8 × 6
#>   term          estimate std.error statistic p.value coef.type
#>   <chr>            <dbl>     <dbl>     <dbl> <lgl>   <chr>
#> 1 InflMedium       0.566    0.105       5.41 NA      coefficient
#> 2 InflHigh         1.29     0.127      10.1  NA      coefficient
#> 3 TypeApartment   -0.572    0.119      -4.80 NA      coefficient
#> 4 TypeAtrium      -0.366    0.155      -2.36 NA      coefficient
#> 5 TypeTerrace     -1.09     0.151      -7.20 NA      coefficient
#> 6 ContHigh         0.360    0.0955      3.77 NA      coefficient
#> 7 Low|Medium      -0.496    0.125      -3.97 NA      scale
#> 8 Medium|High      0.691    0.125       5.50 NA      scale