diff --git a/python/pyspark/pandas/base.py b/python/pyspark/pandas/base.py
index bdec3ac3f46f5..bd1d13ce200fb 100644
--- a/python/pyspark/pandas/base.py
+++ b/python/pyspark/pandas/base.py
@@ -1326,8 +1326,8 @@ def value_counts(
                 sdf = sdf.orderBy(F.col("count").desc())
 
         if normalize:
-            sum = sdf_dropna.count()
-            sdf = sdf.withColumn("count", F.col("count") / SF.lit(sum))
+            drop_sum = sdf_dropna.count()
+            sdf = sdf.withColumn("count", F.col("count") / SF.lit(drop_sum))
 
         internal = InternalFrame(
             spark_frame=sdf,