diff --git a/databricks/koalas/frame.py b/databricks/koalas/frame.py
index 9dd3c41630..24ac1b7fce 100644
--- a/databricks/koalas/frame.py
+++ b/databricks/koalas/frame.py
@@ -3979,7 +3979,6 @@ def select_dtypes(self, include=None, exclude=None):
                 inc_ex=set(include).intersection(set(exclude))))
 
         # Handle Spark types
-        columns = []
         include_spark_type = []
         for inc in include:
             try:
@@ -4009,23 +4008,26 @@ def select_dtypes(self, include=None, exclude=None):
             except:
                 pass
 
-        for col in self._internal.data_columns:
+        columns = []
+        column_index = []
+        for idx, col in zip(self._internal.column_index, self._internal.data_columns):
             if len(include) > 0:
                 should_include = (
-                    infer_dtype_from_object(self[col].dtype.name) in include_numpy_type or
-                    self._sdf.schema[col].dataType in include_spark_type)
+                    infer_dtype_from_object(self[idx].dtype.name) in include_numpy_type or
+                    self._internal.spark_type_for(col) in include_spark_type)
             else:
                 should_include = not (
-                    infer_dtype_from_object(self[col].dtype.name) in exclude_numpy_type or
-                    self._sdf.schema[col].dataType in exclude_spark_type)
+                    infer_dtype_from_object(self[idx].dtype.name) in exclude_numpy_type or
+                    self._internal.spark_type_for(col) in exclude_spark_type)
 
             if should_include:
-                columns += col
+                columns.append(col)
+                column_index.append(idx)
 
         return DataFrame(self._internal.copy(
             sdf=self._sdf.select(self._internal.index_scols +
-                                 [scol_for(self._sdf, col) for col in columns]),
-            data_columns=columns))
+                                 [self._internal.scol_for(col) for col in columns]),
+            data_columns=columns, column_index=column_index))
 
     def count(self, axis=None):
         """