Calculating confidence intervals for coefficients in ordinal logistic regression

Question

I've fitted an ordinal logistic regression model in R using MASS::polr and I am looking to compute confidence intervals for the fitted coefficients.

Here is the (cropped) model output:

polr(formula = response ~ group, data = df, 
    Hess = TRUE, method = "logistic")
Coefficients:
        Value       Std. Error t value
group1  -0.5279     0.7007     -0.7533
group2  -1.0144     0.7163     -1.4163

My 2 questions:

1. What distribution should I use? My natural assumption was the t-distribution but I've seen instances online of people using the normal.

2. If the t-distribution, how many degrees of freedom should I use? My hunch was the residual degrees of freedom of the model (n - p) but I've seen instances of people using 1 as well.