@@ -737,6 +737,10 @@ astropy.table
   as long as the custom unit is enabled during reading (otherwise, the unit
   will become an ``UnrecognizedUnit``). [#9015]
 
+- Fix bug where string values could be truncated when inserting into a
+  ``Column`` or ``MaskedColumn``, or when adding or inserting a row containing
+  string values. [#9559]
+
 astropy.tests
 ^^^^^^^^^^^^^
 
 
@@ -125,6 +125,41 @@ def __setitem__(self, item, val):
                              .format(self.__class__.__name__))
 
 
+def _expand_string_array_for_values(arr, values):
+    """
+    For string-dtype return a version of ``arr`` that is wide enough for ``values``.
+    If ``arr`` is not string-dtype or does not need expansion then return ``arr``.
+
+    Parameters
+    ----------
+    arr : np.ndarray
+        Input array
+    values : scalar or array-like
+        Values for width comparison for string arrays
+
+    Returns
+    -------
+    arr_expanded : np.ndarray
+
+    """
+    # If this is not a string type then just return original arr
+    if arr.dtype.kind not in ('U', 'S'):
+        return arr
+
+    # Find the length of the longest string in the new values.
+    values_str_len = np.char.str_len(values).max()
+
+    # Determine character repeat count of self.dtype.  Returns a positive
+    # int or None (something like 'U0' is not possible in numpy).  If new values
+    # are longer than current then make a new (wider) version of self.
+    arr_str_len = dtype_bytes_or_chars(arr.dtype)
+    if arr_str_len and values_str_len > arr_str_len:
+        arr_dtype = arr.dtype.byteorder + arr.dtype.kind + str(values_str_len)
+        return arr.astype(arr_dtype)
+    else:
+        return arr
+
+
 class ColumnInfo(BaseColumnInfo):
     """
     Container for meta information like name, description, format.
@@ -1030,10 +1065,9 @@ def insert(self, obj, values, axis=0):
             data = np.insert(self, obj, None, axis=axis)
             data[obj] = values
         else:
-            # Explicitly convert to dtype of this column.  Needed because numpy 1.7
-            # enforces safe casting by default, so .  This isn't the case for 1.6 or 1.8+.
-            values = np.asarray(values, dtype=self.dtype)
-            data = np.insert(self, obj, values, axis=axis)
+            self_for_insert = _expand_string_array_for_values(self, values)
+            data = np.insert(self_for_insert, obj, values, axis=axis)
+
         out = data.view(self.__class__)
         out.__array_finalize__(self)
         return out
@@ -1343,16 +1377,14 @@ def insert(self, obj, values, mask=None, axis=0):
             new_data = np.insert(self_ma.data, obj, None, axis=axis)
             new_data[obj] = values
         else:
-            # Explicitly convert to dtype of this column.  Needed because numpy 1.7
-            # enforces safe casting by default, so .  This isn't the case for 1.6 or 1.8+.
-            values = np.asarray(values, dtype=self.dtype)
+            self_ma = _expand_string_array_for_values(self_ma, values)
             new_data = np.insert(self_ma.data, obj, values, axis=axis)
 
         if mask is None:
             if self.dtype.kind == 'O':
                 mask = False
             else:
-                mask = np.zeros(values.shape, dtype=bool)
+                mask = np.zeros(np.broadcast(values).shape, dtype=bool)
         new_mask = np.insert(self_ma.mask, obj, mask, axis=axis)
         new_ma = np.ma.array(new_data, mask=new_mask, copy=False)
 
 
@@ -319,6 +319,15 @@ def test_insert_axis(self, Column):
         c1 = c.insert(1, [5, 6], axis=1)
         assert np.all(c1 == [[1, 5, 2], [3, 6, 4]])
 
+    def test_insert_string_expand(self, Column):
+        c = Column(['a', 'b'])
+        c1 = c.insert(0, 'abc')
+        assert np.all(c1 == ['abc', 'a', 'b'])
+
+        c = Column(['a', 'b'])
+        c1 = c.insert(0, ['c', 'def'])
+        assert np.all(c1 == ['c', 'def', 'a', 'b'])
+
     def test_insert_multidim(self, Column):
         c = Column([[1, 2],
                     [3, 4]], name='a', dtype=int)
 
@@ -693,7 +693,7 @@ def test_add_with_list(self, table_types):
         assert len(t) == 4
         assert np.all(t['a'] == np.array([1, 2, 3, 4]))
         assert np.allclose(t['b'], np.array([4.0, 5.1, 6.2, 7.2]))
-        assert np.all(t['c'] == np.array(['7', '8', '9', '1']))
+        assert np.all(t['c'] == np.array(['7', '8', '9', '10']))
 
     def test_add_with_dict(self, table_types):
         self._setup(table_types)
 
@@ -228,7 +228,7 @@ Adding a new row of data to the table is as follows.  Note that the unit
 value is given in ``cm / s`` but will be added to the table as ``0.1 m / s`` in
 accord with the existing unit.
 
-  >>> t.add_row([-8, -9, 10 * u.cm / u.s, 11])
+  >>> t.add_row([-8, 'string', 10 * u.cm / u.s, 10])
   >>> len(t)
   4
Original file line number	Diff line number	Diff line change
`@@ -228,7 +228,7 @@ Adding a new row of data to the table is as follows. Note that the unit`
`228`	`228`	value is given in ``cm / s`` but will be added to the table as ``0.1 m / s`` in
`229`	`229`	`accord with the existing unit.`
`230`	`230`
`231`		`- >>> t.add_row([-8, -9, 10 * u.cm / u.s, 11])`
	`231`	`+ >>> t.add_row([-8, 'string', 10 * u.cm / u.s, 10])`
`232`	`232`	`>>> len(t)`
`233`	`233`	`4`
`234`	`234`