diff --git a/R/remove_empties.R b/R/remove_empties.R
index d6b51d9..5758cee 100644
--- a/R/remove_empties.R
+++ b/R/remove_empties.R
@@ -7,8 +7,8 @@
 #' @param which one of "rows", "cols", or `c("rows", "cols")`. Where no
 #'   value of which is provided, defaults to removing both empty rows and empty
 #'   columns, declaring the behavior with a printed message.
-#' @param cutoff What fraction (>0 to <=1) of rows or columns must be empty to
-#'   be removed?
+#' @param cutoff a row/col should have more than this fraction of non-NA values to be
+#'   retained.  E.g., `cutoff = 0.8` means that rows/cols that are 20% or more missing will be dropped.
 #' @param quiet Should messages be suppressed (`TRUE`) or printed
 #'   (`FALSE`) indicating the summary of empty columns or rows removed?
 #' @return Returns the object without its missing rows or columns.
diff --git a/man/remove_empty.Rd b/man/remove_empty.Rd
index 35252f5..66652c2 100644
--- a/man/remove_empty.Rd
+++ b/man/remove_empty.Rd
@@ -13,8 +13,8 @@ remove_empty(dat, which = c("rows", "cols"), cutoff = 1, quiet = TRUE)
 value of which is provided, defaults to removing both empty rows and empty
 columns, declaring the behavior with a printed message.}
 
-\item{cutoff}{What fraction (>0 to <=1) of rows or columns must be empty to
-be removed?}
+\item{cutoff}{Under what fraction (>0 to <=1) of non-empty rows or columns should
+\code{which} be removed? Lower values keep more rows/columns, higher values drop more.}
 
 \item{quiet}{Should messages be suppressed (\code{TRUE}) or printed
 (\code{FALSE}) indicating the summary of empty columns or rows removed?}