alright alright, it's augment(newdata = ...) -> `augment(data = ...…

…)` for most of our examples

chapters/02-whole-game.qmd

@@ -421,7 +421,7 @@ In other words, if you used a bed net, the ATE weight is the probability that yo
 library(broom)
 library(propensity)
 net_data_wts <- propensity_model |>
-  augment(newdata = net_data, type.predict = "response") |>
+  augment(data = net_data, type.predict = "response") |>
   # .fitted is the value predicted by the model
   # for a given observation
   mutate(wts = wt_ate(.fitted, net))

chapters/08-propensity-scores.qmd

@@ -64,7 +64,7 @@ We can extract the propensity scores by pulling out the predictions on the proba
 Using the `augment()` function from the [{`broom`}](https://broom.tidymodels.org/) package, we can extract these propensity scores and add them to our original data frame.
 The argument `type.predict` is set to `"response"` to indicate that we want to extract the predicted values on the *probability* scale.
 By default, these will be on the linear logit scale.
-The `data` argument contains the original data frame.
+The `data` argument contains the original data frame; if we leave this blank, we'll only get back a data frame with the variables in the propensity score model, but we need the outcome, too.
 This code will output a new data frame consisting of all components in `df` with six additional columns corresponding to the logistic regression model that was fit.
 The `.fitted` column is the propensity score.
 

chapters/10-estimands.qmd

@@ -275,7 +275,7 @@ finite_sample_wts <- glm(
   data = finite_sample,
   family = binomial("probit")
 ) |>
-  augment(newdata = finite_sample, type.predict = "response") |>
+  augment(data = finite_sample, type.predict = "response") |>
   mutate(wts = wt_ate(.fitted, x))
 
 finite_sample_wts |>
@@ -309,7 +309,7 @@ sim <- function(n) {
     data = finite_sample,
     family = binomial("probit")
   ) |>
-    augment(newdata = finite_sample, type.predict = "response") |>
+    augment(data = finite_sample, type.predict = "response") |>
     mutate(wts = wt_ate(.fitted, x))
   bias <- finite_sample_wts |>
     summarize(
@@ -711,7 +711,7 @@ sim <- function(n) {
     data = finite_sample,
     family = binomial("probit")
   ) |>
-    augment(newdata = finite_sample, type.predict = "response") |>
+    augment(data = finite_sample, type.predict = "response") |>
     mutate(
       wts_ate = wt_ate(.fitted, x),
       wts_ato = wt_ato(.fitted, x)

chapters/15-missingness-and-measurement.qmd

@@ -776,7 +776,7 @@ calib_model <- lm(
 )
 
 seven_dwarves_calib <- calib_model |>
-  augment(newdata = seven_dwarfs_train_2018) |>
+  augment(data = seven_dwarfs_train_2018) |>
   rename(wait_minutes_actual_calib = .fitted) |>
   # convert back to the original scale
   # and fill in real values where they exist