Fix regression for loc and __setitem__ when one-dimensional tuple was given for MultiIndex

doc/source/whatsnew/v1.2.0.rst

@@ -470,6 +470,7 @@ Indexing
 - Bug in :meth:`Index.where` incorrectly casting numeric values to strings (:issue:`37591`)
 - Bug in :meth:`Series.loc` and :meth:`DataFrame.loc` raises when numeric label was given for object :class:`Index` although label was in :class:`Index` (:issue:`26491`)
 - Bug in :meth:`DataFrame.loc` returned requested key plus missing values when ``loc`` was applied to single level from :class:`MultiIndex` (:issue:`27104`)
+- Fixed regression in :meth:`DataFrame.loc` and :meth:`Series.loc` for ``__setitem__`` when one-dimensional tuple was given to select from :class:`MultiIndex` (:issue:`37711`)
 
 Missing
 ^^^^^^^

pandas/core/indexing.py

@@ -650,11 +650,12 @@ def _ensure_listlike_indexer(self, key, axis=None, value=None):
         if self.ndim != 2:
             return
 
-        if isinstance(key, tuple):
+        if isinstance(key, tuple) and not isinstance(self.obj.index, ABCMultiIndex):
             # key may be a tuple if we are .loc
-            # in that case, set key to the column part of key
+            # if index is not a MultiIndex, set key to column part
             key = key[column_axis]
             axis = column_axis
+            pass
 
         if (
             axis == column_axis

pandas/tests/indexing/multiindex/test_loc.py

@@ -288,6 +288,14 @@ def convert_nested_indexer(indexer_type, keys):
 
         tm.assert_series_equal(result, expected)
 
+    def test_multiindex_loc_one_dimensional_tuple(self, frame_or_series):
+        # GH#37711
+        mi = MultiIndex.from_tuples([("a", "A"), ("b", "A")])
+        obj = frame_or_series([1, 2], index=mi)
+        obj.loc[("a",)] = 0
+        expected = frame_or_series([0, 2], index=mi)
+        tm.assert_equal(obj, expected)
+
 
 @pytest.mark.parametrize(
     "indexer, pos",