Merge pull request #311 from r-causal/if_else

`ifelse()` -> `if_else()`

chapters/03-po-counterfactuals.qmd

@@ -209,7 +209,7 @@ data_observed <- data |>
     # change the exposure to randomized, generated from
     # a binomial distribution with a probability of 0.5 for
     # being in either group
-    exposure = ifelse(
+    exposure = if_else(
       rbinom(n(), 1, 0.5) == 1, "chocolate", "vanilla"
     ),
     observed_outcome = case_when(
@@ -235,8 +235,8 @@ avg_vanilla <- data_observed |>
 
 data_observed |>
   mutate(
-    y_chocolate = ifelse(exposure == "chocolate", y_chocolate, NA),
-    y_vanilla = ifelse(exposure == "vanilla", y_vanilla, NA),
+    y_chocolate = if_else(exposure == "chocolate", y_chocolate, NA),
+    y_vanilla = if_else(exposure == "vanilla", y_vanilla, NA),
     causal_effect = NA_real_
   ) |>
   select(-observed_outcome, -exposure) |>
@@ -373,12 +373,12 @@ We've exchanged the groups by flipping their assignments, but we can still detec
 set.seed(11)
 
 mix_up <- function(flavor) {
-  ifelse(flavor == "chocolate", "vanilla", "chocolate")
+  if_else(flavor == "chocolate", "vanilla", "chocolate")
 }
 
 data_observed <- data |>
   mutate(
-    exposure = ifelse(
+    exposure = if_else(
       rbinom(n(), 1, 0.5) == 1, "chocolate", "vanilla"
     ),
     exposure = mix_up(exposure),
@@ -404,13 +404,13 @@ data_observed_exch <- data |>
     prefer_chocolate = y_chocolate > y_vanilla,
     exposure = case_when(
       # people who like chocolate more chose that 80% of the time
-      prefer_chocolate ~ ifelse(
+      prefer_chocolate ~ if_else(
         rbinom(n(), 1, 0.8) == 1,
         "chocolate",
         "vanilla"
       ),
       # people who like vanilla more chose that 80% of the time
-      !prefer_chocolate ~ ifelse(
+      !prefer_chocolate ~ if_else(
         rbinom(n(), 1, 0.8) == 1,
         "vanilla",
         "chocolate"
@@ -478,7 +478,7 @@ This is called **conditional exchangeability**: $Y(x) \perp\!\!\!\perp X \mid Z$
 #| fig-cap: "The average potential outcomes by observed exposure group in the presence of confounding. We can still achieve *conditional* exchangeability within levels of the confounder. Here, we also start to see the limits of our sample size, as the potential outcomes, which would be valid in higher numbers, start to fail."
 #| code-fold: true
 data_observed_exch |>
-  mutate(prefer_chocolate = ifelse(
+  mutate(prefer_chocolate = if_else(
     prefer_chocolate,
     "prefers\nchocolate",
     "prefers\nvanilla"
@@ -538,12 +538,12 @@ data_observed_pos <- data |>
   mutate(
     prefer_chocolate = y_chocolate > y_vanilla,
     exposure = case_when(
-      prefer_chocolate ~ ifelse(
+      prefer_chocolate ~ if_else(
         rbinom(n(), 1, 0.8) == 1,
         "chocolate",
         "vanilla"
       ),
-      !prefer_chocolate ~ ifelse(
+      !prefer_chocolate ~ if_else(
         rbinom(n(), 1, 0.8) == 1,
         "vanilla",
         "chocolate"
@@ -573,7 +573,7 @@ In this case, let's say that anyone with an allergy to vanilla who is assigned v
 set.seed(11)
 data_observed_struc <- data |>
   mutate(
-    exposure = ifelse(
+    exposure = if_else(
       rbinom(n(), 1, 0.5) == 1,
       "chocolate",
       "vanilla"
@@ -586,8 +586,8 @@ data_observed_struc <- data_observed_struc |>
     # 30% chance of allergy
     allergy = rbinom(n(), 1, 0.3) == 1,
     # in which case `y_vanilla` is impossible
-    exposure = ifelse(allergy, "chocolate", exposure),
-    y_vanilla = ifelse(allergy, NA, y_vanilla),
+    exposure = if_else(allergy, "chocolate", exposure),
+    y_vanilla = if_else(allergy, NA, y_vanilla),
     observed_outcome = case_when(
       # those with allergies always take chocolate
       allergy ~ y_chocolate,
@@ -869,10 +869,10 @@ data <- tibble(
 set.seed(37)
 data_observed_interf <- data |>
   mutate(
-    exposure = ifelse(
+    exposure = if_else(
       rbinom(n(), 1, 0.5) == 1, "chocolate", "vanilla"
     ),
-    exposure_partner = ifelse(
+    exposure_partner = if_else(
       rbinom(n(), 1, 0.5) == 1, "chocolate", "vanilla"
     ),
     observed_outcome = case_when(
@@ -921,7 +921,7 @@ data_observed_interf |>
       potential_outcome == "y(vanilla)" & exposure == "vanilla" ~ TRUE,
       TRUE ~ FALSE
     ),
-    flavor_match = ifelse(exposure == exposure_partner, "Same Flavors", "Different Flavors") # Collapse to same/different
+    flavor_match = if_else(exposure == exposure_partner, "Same Flavors", "Different Flavors") # Collapse to same/different
   ) |>
   ggplot(aes(happiness, exposure)) +
   geom_jitter(
@@ -985,7 +985,7 @@ set.seed(11)
 ## we are now randomizing the *partnerships* not the individuals
 partners <- tibble(
   partner_id = 1:5,
-  exposure = ifelse(
+  exposure = if_else(
     rbinom(5, 1, 0.5) == 1, "chocolate", "vanilla"
   )
 )

chapters/05-not-just-a-stats-problem.qmd

@@ -286,7 +286,7 @@ d_mbias <- dagify(
 
 p_coll <- d_coll |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(label == "c", "covariate", NA_character_)) |>
+  mutate(covariate = if_else(label == "c", "covariate", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -307,7 +307,7 @@ p_coll <- d_coll |>
 
 p_conf <- d_conf |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(label == "c", "covariate", NA_character_)) |>
+  mutate(covariate = if_else(label == "c", "covariate", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -327,7 +327,7 @@ p_conf <- d_conf |>
 
 p_med <- d_med |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(label == "c", "covariate", NA_character_)) |>
+  mutate(covariate = if_else(label == "c", "covariate", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -348,7 +348,7 @@ p_med <- d_med |>
 
 p_m_bias <- d_mbias |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(label == "c", "covariate", NA_character_)) |>
+  mutate(covariate = if_else(label == "c", "covariate", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -460,7 +460,7 @@ d_coll <- dagify(
 
 d_coll |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(name == "Z_3", "covariate\n(follow-up)", NA_character_)) |>
+  mutate(covariate = if_else(name == "Z_3", "covariate\n(follow-up)", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -482,7 +482,7 @@ d_coll |>
 
 d_coll |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(name == "Z_2", "covariate\n(baseline)", NA_character_)) |>
+  mutate(covariate = if_else(name == "Z_2", "covariate\n(baseline)", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +
@@ -673,7 +673,7 @@ d_conf2 <- dagify(
 
 p_conf2 <- d_conf2 |>
   tidy_dagitty() |>
-  mutate(covariate = ifelse(name == "Q", "covariate", NA_character_)) |>
+  mutate(covariate = if_else(name == "Q", "covariate", NA_character_)) |>
   ggplot(
     aes(x = x, y = y, xend = xend, yend = yend)
   ) +

chapters/06-stats-models-ci.qmd

@@ -563,7 +563,7 @@ satisfaction_randomized <- tibble(
 plot_estimates <- function(d) {
   unadj_model <- lm(satisfaction ~ update_frequency, data = d) |>
     tidy(conf.int = TRUE) |>
-    mutate(term = ifelse(
+    mutate(term = if_else(
       term == "update_frequencydaily",
       "update_frequency",
       term
@@ -577,7 +577,7 @@ plot_estimates <- function(d) {
     data = d
   ) |>
     tidy(conf.int = TRUE) |>
-    mutate(term = ifelse(
+    mutate(term = if_else(
       term == "update_frequencydaily",
       "update_frequency",
       term
@@ -625,8 +625,8 @@ plot_estimates <- function(d) {
       names_to = "statistic"
     ) |>
     mutate(
-      conf.low = ifelse(statistic == "std.error", NA, conf.low),
-      conf.high = ifelse(statistic == "std.error", NA, conf.high),
+      conf.low = if_else(statistic == "std.error", NA, conf.low),
+      conf.high = if_else(statistic == "std.error", NA, conf.high),
       statistic = case_match(
         statistic,
         "estimate" ~ "estimate (95% CI)",

chapters/07-prep-data.qmd

@@ -269,7 +269,7 @@ seven_dwarfs_9 <- seven_dwarfs_train |>
       wait_minutes_posted_avg,
       wait_minutes_actual_avg
     ),
-    \(.x) ifelse(is.nan(.x), NA, .x)
+    \(.x) if_else(is.nan(.x), NA, .x)
   )) |>
   # outcome definition:
   # only keep the average wait time between 9 and 10

chapters/11-outcome-model.qmd

@@ -260,8 +260,8 @@ library(PSW)
 
 seven_dwarfs_9 <- seven_dwarfs_9 |>
   mutate(
-    park_ticket_season_regular = ifelse(park_ticket_season == "regular", 1, 0),
-    park_ticket_season_value = ifelse(park_ticket_season == "value", 1, 0)
+    park_ticket_season_regular = if_else(park_ticket_season == "regular", 1, 0),
+    park_ticket_season_value = if_else(park_ticket_season == "value", 1, 0)
   )
 psw(
   data = seven_dwarfs_9,

chapters/15-missingness-and-measurement.qmd

@@ -80,7 +80,7 @@ edges_with_aes <- function(..., edge_color = "grey85", shadow = TRUE) {
     },
     geom_dag_edges_link(
       aes(edge_color = path),
-      data = \(.x) mutate(.x, path = ifelse(is.na(to), NA, path))
+      data = \(.x) mutate(.x, path = if_else(is.na(to), NA, path))
     )
   )
 }
@@ -259,7 +259,7 @@ y <- case_when(
   x == "e" ~ 5 + rnorm(n),
 )
 
-x_measured <- ifelse(
+x_measured <- if_else(
   x %in% c("c", "d"),
   sample(c("c", "d"), size = n, replace = TRUE),
   x
@@ -394,7 +394,7 @@ outcome <- exposure + confounder + rnorm(n)
 true_model <- lm(outcome ~ exposure * confounder)
 
 # mismeasure confounder
-confounder <- ifelse(
+confounder <- if_else(
   outcome > 0,
   confounder,
   confounder + 10 * rnorm(n)
@@ -650,9 +650,9 @@ actual <- coef * posted + rnorm(365, mean = 0, sd = 2)
 posted_60 <- posted / 60
 missing_dag_1 <- rbinom(365, 1, .3) |>
   as.logical()
-missing_dag_2 <- ifelse(posted_60 > .50, rbinom(365, 1, .95), 0) |>
+missing_dag_2 <- if_else(posted_60 > .50, rbinom(365, 1, .95), 0) |>
   as.logical()
-missing_dag_3 <- ifelse(actual > 22, rbinom(365, 1, .99), 0) |>
+missing_dag_3 <- if_else(actual > 22, rbinom(365, 1, .99), 0) |>
   as.logical()
 # the same structure, but it's `posted` that gets the resulting missingness
 missing_dag_4 <- missing_dag_2
@@ -702,7 +702,7 @@ dag_stats <- bind_rows(
 
 dag_stats |>
   mutate(
-    true_value = ifelse(dag == "No missingness", "True value", "Observed value"),
+    true_value = if_else(dag == "No missingness", "True value", "Observed value"),
     dag = factor(dag, levels = c(paste("DAG", 5:1), "No missingness")),
     stat = factor(
       stat,

chapters/16-sensitivity.qmd

@@ -402,7 +402,7 @@ seven_dwarfs_sim2 <- seven_dwarfs_train_2018 |>
   mutate(
     u = rnorm(n(), mean = 10, sd = 3),
     wait_minutes_posted_avg = wait_minutes_posted_avg + u,
-    park_extra_magic_morning = ifelse(
+    park_extra_magic_morning = if_else(
       u > 10,
       rbinom(1, 1, .1),
       park_extra_magic_morning
@@ -809,7 +809,7 @@ curvatures[5] <- 0.3
 emm_wait_dag |>
   tidy_dagitty() |>
   node_status() |>
-  mutate(linetype = ifelse(name == "park_temperature_high", "dashed", "solid")) |>
+  mutate(linetype = if_else(name == "park_temperature_high", "dashed", "solid")) |>
   ggplot(
     aes(x, y, xend = xend, yend = yend, color = status, edge_linetype = linetype)
   ) +