diff --git a/dev/run-tests-jenkins.py b/dev/run-tests-jenkins.py
index 72e32d4e16e14..13be9592d771f 100755
--- a/dev/run-tests-jenkins.py
+++ b/dev/run-tests-jenkins.py
@@ -198,7 +198,7 @@ def main():
     # format: http://linux.die.net/man/1/timeout
     # must be less than the timeout configured on Jenkins. Usually Jenkins's timeout is higher
     # then this. Please consult with the build manager or a committer when it should be increased.
-    tests_timeout = "400m"
+    tests_timeout = "500m"
 
     # Array to capture all test names to run on the pull request. These tests are represented
     # by their file equivalents in the dev/tests/ directory.
diff --git a/python/pyspark/sql/tests/test_arrow.py b/python/pyspark/sql/tests/test_arrow.py
index c3c9fb0f12a25..c59765dd79eb9 100644
--- a/python/pyspark/sql/tests/test_arrow.py
+++ b/python/pyspark/sql/tests/test_arrow.py
@@ -435,11 +435,12 @@ def test_createDateFrame_with_category_type(self):
         assert_frame_equal(result_spark, result_arrow)
 
         # ensure original category elements are string
-        assert isinstance(category_first_element, str)
+        self.assertIsInstance(category_first_element, str)
         # spark data frame and arrow execution mode enabled data frame type must match pandas
-        assert spark_type == arrow_type == 'string'
-        assert isinstance(arrow_first_category_element, str)
-        assert isinstance(spark_first_category_element, str)
+        self.assertEqual(spark_type, 'string')
+        self.assertEqual(arrow_type, 'string')
+        self.assertIsInstance(arrow_first_category_element, str)
+        self.assertIsInstance(spark_first_category_element, str)
 
 
 @unittest.skipIf(
diff --git a/python/pyspark/sql/tests/test_pandas_udf_scalar.py b/python/pyspark/sql/tests/test_pandas_udf_scalar.py
index ae6b8d520f735..2d38efd39f902 100644
--- a/python/pyspark/sql/tests/test_pandas_udf_scalar.py
+++ b/python/pyspark/sql/tests/test_pandas_udf_scalar.py
@@ -910,13 +910,10 @@ def to_category_func(x):
 
         spark_type = df.dtypes[1][1]
         # spark data frame and arrow execution mode enabled data frame type must match pandas
-        assert spark_type == 'string'
+        self.assertEqual(spark_type, 'string')
 
-        # Check result value of column 'B' must be equal to column 'A'
-        for i in range(0, len(result_spark["A"])):
-            assert result_spark["A"][i] == result_spark["B"][i]
-            assert isinstance(result_spark["A"][i], str)
-            assert isinstance(result_spark["B"][i], str)
+        # Check result of column 'B' must be equal to column 'A' in type and values
+        pd.testing.assert_series_equal(result_spark["A"], result_spark["B"], check_names=False)
 
     @unittest.skipIf(sys.version_info[:2] < (3, 5), "Type hints are supported from Python 3.5.")
     def test_type_annotation(self):