diff --git a/README.rst b/README.rst
index 379a2b6..b1d9883 100755
--- a/README.rst
+++ b/README.rst
@@ -161,50 +161,7 @@ queries are structured in three levels:
 
 Response data
 ~~~~~~~~~~~~~
-
-The data from a query can be returned in three different formats: XML, JSON or as
-a Python object.
-
-To get the data in XML format a first positional argument (``raw_data``) has to be
-set to ``True``. This is so because the default raw data response format is XML.
-
-   .. code:: python
-
-           >>> resp = cl.search_palette(True, id=2323)
-           >>> resp
-           '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>\n<palettes numResults="1" totalResults="4567661">\n\t<palette>\n\t\t<id>2323</id>\n\t\t<title><![CDATA[On the lake]]></title>\n\t\t<userName><![CDATA[viatora]]></userName>\n\t\t<numViews>529</numViews>\n\t\t<numVotes>12</numVotes>\n\t\t<numComments>3</numComments>\n\t\t<numHearts>0</numHearts>\n\t\t<rank>0</rank>\n\t\t<dateCreated>2005-08-24 10:44:11</dateCreated>\n\t\t<colors>\n\t\t\t<hex>E6F0F7</hex>\n\t\t\t<hex>97A4B2</hex>\n\t\t\t<hex>5F0609</hex>\n\t\t\t<hex>766F59</hex>\n\t\t\t<hex>989383</hex>\n\t\t</colors>\n\t\t<description><![CDATA[I-MOO\r\n<div style="width: 300px; text-align: center;"><a href="http://www.colourlovers.com/contests/moo/minicard/2291466" target="_blank" style="display: block; margin-bottom: 5px; width: 300px; height: 120px; -moz-box-shadow: 0 1px 4px #d1d1d1; -webkit-box-shadow: 0 1px 4px #d1d1d1; box-shadow: 0 1px 4px #d1d1d1; filter: progid:DXImageTransform.Microsoft.Shadow(Strength=1, Direction=180, Color=]]></description>\n\t\t<url><![CDATA[http://www.colourlovers.com/palette/2323/On_the_lake]]></url>\n\t\t<imageUrl><![CDATA[http://www.colourlovers.com/paletteImg/E6F0F7/97A4B2/5F0609/766F59/989383/On_the_lake.png]]></imageUrl>\n\t\t<badgeUrl><![CDATA[http://www.colourlovers.com/images/badges/p/2/2323_On_the_lake.png]]></badgeUrl>\n\t\t<apiUrl>http://www.colourlovers.com/api/palette/2323</apiUrl>\n\t</palette>\n</palettes>'
-
-To get the data in JSON format, a part from specifying that the response should be returned
-as raw data by setting the first positional argument to ``True``, it is necessary to include
-another keyword parameter in the query specifying that the format of the response should be
-JSON (``format='json'``).
-
-   .. code:: python
-
-           >>> resp = cl.search_palette(True, id=2323, format='json')
-           >>> resp
-           '[
-             {
-              "id":2323,
-              "title":"On the lake",
-              "userName":"viatora",
-              "numViews":529,
-              "numVotes":12,
-              "numComments":3,
-              "numHearts":0,
-              "rank":0,
-              "dateCreated":"2005-08-24 10:44:11",
-              "colors":["E6F0F7","97A4B2","5F0609","766F59","989383"],
-              "description":"I-MOO\\r\\n<div style=\\"width: 300px; text-align: center;\\"><a href=\\"http:\\/\\/www.colourlovers.com\\/contests\\/moo\\/minicard\\/2291466\\" target=\\"_blank\\" style=\\"display: block; margin-bottom: 5px; width: 300px; height: 120px; -moz-box-shadow: 0 1px 4px #d1d1d1; -webkit-box-shadow: 0 1px 4px #d1d1d1; box-shadow: 0 1px 4px #d1d1d1; filter: progid:DXImageTransform.Microsoft.Shadow(Strength=1, Direction=180, Color=",
-              "url":"http:\\/\\/www.colourlovers.com\\/palette\\/2323\\/On_the_lake",
-              "imageUrl":"http:\\/\\/www.colourlovers.com\\/paletteImg\\/E6F0F7\\/97A4B2\\/5F0609\\/766F59\\/989383\\/On_the_lake.png",
-              "badgeUrl":"http:\\/\\/www.colourlovers.com\\/images\\/badges\\/p\\/2\\/2323_On_the_lake.png",
-              "apiUrl":"http:\\/\\/www.colourlovers.com\\/api\\/palette\\/2323"
-             }
-            ]'
-
-Finally, if the parameter raw data is not set or set to ``False`` then the data will be obtained
-as a Python object (which is the recommended way). If the response of a query contains more than
+The data will be obtained as a Python object. If the response of a query contains more than
 one object - for example, when querying for palettes - then each object in the response will be
 mapped to a Python object. Finally, when the response data is obtained as Python objects it will
 always come inside of a list for consistency reasons among methods, even if the response contains
diff --git a/colourlovers/clapi.py b/colourlovers/clapi.py
index 225dc9a..cac074d 100755
--- a/colourlovers/clapi.py
+++ b/colourlovers/clapi.py
@@ -224,18 +224,13 @@ def __public_api_method(self, search_type, data_container):
         type and process and store the obtained data
         """
 
-        def _api_search(raw_data=False, **kwargs):
+        def _api_search(**kwargs):
             """
             This method validates the request parameters and, if all
             of them are valid, builds the request and posts it to the
             API. It will return the data obtained from the response either
             as raw data or as a Python object.
 
-            :param raw_data: Specifies how the data from the response
-            should be treated. If True, the data is returned as it is obtained
-            from the API response without any treatment. If False, the
-            corresponding Python object to store the obtained data is created and
-            returned
             :param kwargs: query parameters for the specified type of request (Pattern,
             Palette, ...). The keyword is the parameter name and the value is the parameter
             value
@@ -245,14 +240,8 @@ def _api_search(raw_data=False, **kwargs):
             # Validate the type of request (new, top, random, ...) taking into account the
             # type of request (pattern, palette, colour, ...) that is to be performed
             processed_request = self.__process_optional_requests(search_type, **kwargs)
-            if not isinstance(raw_data, bool):
-                raise ValueError("Invalid parameter " + str(raw_data))
 
-            if not raw_data:
-                # if user hasn't asked for the raw data of the API
-                # response build container objects. For that, we need
-                # the data in json format
-                processed_request.kwargs["format"] = "json"
+            processed_request.kwargs["format"] = "json"
             # Once request type has been validated make the query to the API
             api_response = self.__query(
                 search_type,
@@ -261,7 +250,7 @@ def _api_search(raw_data=False, **kwargs):
             ).decode()
             # Process the data obtained from the query. We will build container
             # objects by default unless otherwise specified
-            containers = self.__process_response(raw_data, api_response, data_container)
+            containers = self.__process_response(api_response, data_container)
             if containers is not None:
                 return containers
             else:
@@ -372,7 +361,7 @@ def __request(self, search_term, optional_request_term, **kwargs):
         data = response.read()
         return data
 
-    def __process_response(self, raw_data, api_response, request_type_class):
+    def __process_response(self, api_response, request_type_class):
         """
         Once a request to the API has been made, process the obtained data
         and parse it as requested.
@@ -380,8 +369,6 @@ def __process_response(self, raw_data, api_response, request_type_class):
         Note: This method returns more than one type of object. It should be
         refactored so that only one type of data is returned.
 
-        :param raw_data: Boolean specifying whether the obtained data should be
-        processed or returned as is
         :param api_response: Raw data obtained from the API query
         :param request_type_class: The type of data the request asked for
         :return: Either None if no response was obtained, the raw data if
@@ -389,8 +376,6 @@ def __process_response(self, raw_data, api_response, request_type_class):
         """
         if not api_response:
             return None
-        if raw_data:
-            return api_response
         else:
             parsed_json = json.loads(api_response)
             if isinstance(parsed_json, dict):