diff --git a/transform.Rmd b/transform.Rmd
index efaa7c9..8c721f0 100644
--- a/transform.Rmd
+++ b/transform.Rmd
@@ -818,7 +818,7 @@ daily %>%
     `not_cancelled %>% count(tailnum, wt = distance)` (without using 
     `count()`).
 
-1.  Our definition of cancelled flights (`!is.na(dep_delay) & !is.na(arr_delay)`
+1.  Our definition of cancelled flights (`is.na(dep_delay) | is.na(arr_delay)`
     ) is slightly suboptimal. Why? Which is the most important column?
 
 1.  Look at the number of cancelled flights per day. Is there a pattern?