diff --git a/python/pyspark/core/rdd.py b/python/pyspark/core/rdd.py
index 803bb6d5b882b..573cc7ae165c5 100644
--- a/python/pyspark/core/rdd.py
+++ b/python/pyspark/core/rdd.py
@@ -4843,7 +4843,8 @@ def sumApprox(
         jrdd = self.mapPartitions(lambda it: [float(sum(it))])._to_java_object_rdd()
         assert self.ctx._jvm is not None
         jdrdd = self.ctx._jvm.JavaDoubleRDD.fromRDD(jrdd.rdd())
-        r = jdrdd.sumApprox(timeout, confidence).getFinalValue()
+        partial = jdrdd.sumApprox(timeout, confidence)
+        r = partial.initialValue()
         return BoundedFloat(r.mean(), r.confidence(), r.low(), r.high())
 
     def meanApprox(
diff --git a/python/pyspark/tests/test_rdd.py b/python/pyspark/tests/test_rdd.py
index 4c43d6181cc36..507d13c74d815 100644
--- a/python/pyspark/tests/test_rdd.py
+++ b/python/pyspark/tests/test_rdd.py
@@ -638,6 +638,18 @@ def test_distinct(self):
         self.assertEqual(result.getNumPartitions(), 5)
         self.assertEqual(result.count(), 3)
 
+    def test_count_approx_respects_timeout(self):
+        rdd = self.sc.range(1000000, numSlices=8)
+        start = time.time()
+        result = rdd.countApprox(timeout=100)
+        elapsed = time.time() - start
+        self.assertLess(elapsed, 10)
+        self.assertIsNotNone(result)
+
+    def test_count_approx_returns_exact_when_completed(self):
+        rdd = self.sc.parallelize(range(1000), 8)
+        self.assertEqual(rdd.countApprox(timeout=5000), 1000)
+
     def test_external_group_by_key(self):
         self.sc._conf.set("spark.python.worker.memory", "1m")
         N = 2000001