diff --git a/sdks/python/apache_beam/dataframe/frames.py b/sdks/python/apache_beam/dataframe/frames.py
index 0d9b22ae3f9e..756e75b9583c 100644
--- a/sdks/python/apache_beam/dataframe/frames.py
+++ b/sdks/python/apache_beam/dataframe/frames.py
@@ -4907,9 +4907,9 @@ def __setitem__(self, index, value):
 
 
 class _DeferredStringMethods(frame_base.DeferredBase):
-  @frame_base.with_docs_from(pd.core.strings.StringMethods)
-  @frame_base.args_to_kwargs(pd.core.strings.StringMethods)
-  @frame_base.populate_defaults(pd.core.strings.StringMethods)
+  @frame_base.with_docs_from(pd.Series.str)
+  @frame_base.args_to_kwargs(pd.Series.str)
+  @frame_base.populate_defaults(pd.Series.str)
   def cat(self, others, join, **kwargs):
     """If defined, ``others`` must be a :class:`DeferredSeries` or a ``list`` of
     ``DeferredSeries``."""
@@ -4949,8 +4949,8 @@ def func(*args):
             requires_partition_by=requires,
             preserves_partition_by=partitionings.Arbitrary()))
 
-  @frame_base.with_docs_from(pd.core.strings.StringMethods)
-  @frame_base.args_to_kwargs(pd.core.strings.StringMethods)
+  @frame_base.with_docs_from(pd.Series.str)
+  @frame_base.args_to_kwargs(pd.Series.str)
   def repeat(self, repeats):
     """``repeats`` must be an ``int`` or a :class:`DeferredSeries`. Lists are
     not supported because they make this operation order-sensitive."""
@@ -4987,8 +4987,8 @@ def repeat(self, repeats):
       raise TypeError("str.repeat(repeats=) value must be an int or a "
                       f"DeferredSeries (encountered {type(repeats)}).")
 
-  @frame_base.with_docs_from(pd.core.strings.StringMethods)
-  @frame_base.args_to_kwargs(pd.core.strings.StringMethods)
+  @frame_base.with_docs_from(pd.Series.str)
+  @frame_base.args_to_kwargs(pd.Series.str)
   def get_dummies(self, **kwargs):
     """
     Series must be categorical dtype. Please cast to ``CategoricalDtype``
@@ -5070,9 +5070,9 @@ def func(s):
             requires_partition_by=partitionings.Arbitrary(),
             preserves_partition_by=partitionings.Arbitrary()))
 
-  @frame_base.with_docs_from(pd.core.strings.StringMethods)
-  @frame_base.args_to_kwargs(pd.core.strings.StringMethods)
-  @frame_base.populate_defaults(pd.core.strings.StringMethods)
+  @frame_base.with_docs_from(pd.Series.str)
+  @frame_base.args_to_kwargs(pd.Series.str)
+  @frame_base.populate_defaults(pd.Series.str)
   def split(self, **kwargs):
     """
     Like other non-deferred methods, dtype must be CategoricalDtype.
@@ -5081,9 +5081,9 @@ def split(self, **kwargs):
     """
     return self._split_helper(rsplit=False, **kwargs)
 
-  @frame_base.with_docs_from(pd.core.strings.StringMethods)
-  @frame_base.args_to_kwargs(pd.core.strings.StringMethods)
-  @frame_base.populate_defaults(pd.core.strings.StringMethods)
+  @frame_base.with_docs_from(pd.Series.str)
+  @frame_base.args_to_kwargs(pd.Series.str)
+  @frame_base.populate_defaults(pd.Series.str)
   def rsplit(self, **kwargs):
     """
     Like other non-deferred methods, dtype must be CategoricalDtype.
@@ -5161,17 +5161,17 @@ def func(df, *args, **kwargs):
   return func
 
 for method in ELEMENTWISE_STRING_METHODS:
-  if not hasattr(pd.core.strings.StringMethods, method):
+  if not hasattr(pd.Series.str, method):
     # older versions (1.0.x) don't support some of these methods
     continue
   setattr(_DeferredStringMethods,
           method,
           frame_base._elementwise_method(make_str_func(method),
                                          name=method,
-                                         base=pd.core.strings.StringMethods))
+                                         base=pd.Series.str))
 
 for method in NON_ELEMENTWISE_STRING_METHODS:
-  if not hasattr(pd.core.strings.StringMethods, method):
+  if not hasattr(pd.Series.str, method):
     # older versions (1.0.x) don't support some of these methods
     continue
   setattr(_DeferredStringMethods,
@@ -5179,7 +5179,7 @@ def func(df, *args, **kwargs):
           frame_base._proxy_method(
               make_str_func(method),
               name=method,
-              base=pd.core.strings.StringMethods,
+              base=pd.Series.str,
               requires_partition_by=partitionings.Arbitrary(),
               preserves_partition_by=partitionings.Singleton()))
 
diff --git a/sdks/python/apache_beam/dataframe/frames_test.py b/sdks/python/apache_beam/dataframe/frames_test.py
index fa121aa85c30..d2d2b7339c10 100644
--- a/sdks/python/apache_beam/dataframe/frames_test.py
+++ b/sdks/python/apache_beam/dataframe/frames_test.py
@@ -2936,7 +2936,7 @@ class DocstringTest(unittest.TestCase):
       (frames.DeferredDataFrame, pd.DataFrame),
       (frames.DeferredSeries, pd.Series),
       #(frames._DeferredIndex, pd.Index),
-      (frames._DeferredStringMethods, pd.core.strings.StringMethods),
+      (frames._DeferredStringMethods, pd.Series.str),
       (
           frames._DeferredCategoricalMethods,
           pd.core.arrays.categorical.CategoricalAccessor),