Merge pull request #77 from r-causal/analysis-as-df

`analysis()` -> `as.data.frame()`

Author: malcolmbarrett

exercises/01-whole-game-exercises.qmd

@@ -272,9 +272,9 @@ Now let's try the bootstrap. First, we need to wrap our model in a function so w
 
 ```{r}
 # fit ipw model for a single bootstrap sample
-fit_ipw_not_quite_rightly <- function(split, ...) {
-  # get bootstrapped data sample with `rsample::analysis()`
-  .df <- analysis(split)
+fit_ipw_not_quite_rightly <- function(.split, ...) {
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit ipw model
   lm(wt82_71 ~ qsmk, data = .df, weights = wts) |>
@@ -285,8 +285,9 @@ fit_ipw_not_quite_rightly <- function(split, ...) {
 The problem is that we need to account for the *entire* modeling process, so we need to include the first step of our analysis -- fitting the inverse probability weights.
 
 ```{r}
-fit_ipw <- function(split, ...) {
-  .df <- analysis(split)
+fit_ipw <- function(.split, ...) {
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit propensity score model
   propensity_model <- glm(

exercises/10-continuous-g-computation-exercises.qmd

@@ -171,8 +171,9 @@ Remember, you need to bootstrap the entire modeling process, including the regre
 set.seed(1234)
 library(rsample)
 
-fit_gcomp <- function(split, ...) { 
-  .df <- analysis(split) 
+fit_gcomp <- function(.split, ...) { 
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit outcome model. remember to model using `.df` instead of `wait_times`
   

exercises/13-bonus-selection-bias-exercises.qmd

@@ -102,8 +102,9 @@ Finish early? Try bootstrapping the G-computation model with censoring weights
 Remember, you need to bootstrap the entire modeling process, including fitting both regression models, cloning the data sets, and calculating the effects.
 
 ```{r}
-fit_gcomp_cens <- function(split, ...) { 
-  .df <- analysis(split) 
+fit_gcomp_cens <- function(.split, ...) { 
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit the censoring model. remember to model using `.df` instead of `nhefs_censored`
 

exercises/14-bonus-continuous-pscores-exercises.qmd

@@ -183,8 +183,9 @@ There's nothing new here. Just remember, you need to bootstrap the entire modeli
 set.seed(1234)
 library(rsample)
 
-fit_model <- function(split, ...) { 
-  .df <- analysis(split) 
+fit_model <- function(.split, ...) { 
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fill in the rest!
 }

slides/raw/01-causal_modeling_whole_game.qmd

@@ -456,9 +456,9 @@ as_tibble(ipw_estimate_robust)
 
 ```{r}
 # fit ipw model for a single bootstrap sample
-fit_ipw_not_quite_rightly <- function(split, ...) { 
-  # get bootstrapped data sample with `rsample::analysis()`
-  .df <- analysis(split)
+fit_ipw_not_quite_rightly <- function(.split, ...) { 
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit ipw model
   lm(wt82_71 ~ qsmk, data = .df, weights = wts) |>
@@ -469,8 +469,9 @@ fit_ipw_not_quite_rightly <- function(split, ...) {
 ## {.small}
 
 ```{r}
-fit_ipw <- function(split, ...) {
-  .df <- analysis(split)
+fit_ipw <- function(.split, ...) {
+  # get bootstrapped data frame
+  .df <- as.data.frame(.split)
   
   # fit propensity score model
   propensity_model <- glm(

slides/raw/09-outcome-model.qmd

@@ -48,8 +48,8 @@ lm(outcome ~ exposure, data = df, weights = wts) |>
 
 ```{r}
 #| code-line-numbers: "|1-2|5-13|16-18|21-22"
-fit_ipw <- function(split, ...) {
-  .df <- analysis(split)
+fit_ipw <- function(.split, ...) {
+  .df <- as.data.frame(.split)
   
   # fit propensity score model
   propensity_model <- glm(