Written question asked by David Laws (Liberal Democrat) on Wednesday, 21 May 2008, in the House of Commons. It was due for an answer on Monday, 2 June 2008. It was answered by Lord Knight of Weymouth (Labour) on Thursday, 5 June 2008 on behalf of the Department for Children, Schools and Families.
Education: Assessments
- Question
- To ask the Secretary of State for Children, Schools and Families (1) what margin of error was permitted by markers passing the standardisation test to mark (a) Key Stage 2 and (b) Key Stage 3 national tests in (i) science, (ii) mathematics and (iii) English in each year since 2005; and if he will make a statement;
- Answer
-
The National Assessment Agency (NAA) is responsible for the standardisation of markers for National Curriculum tests. I have therefore asked David Gee, Managing Director, NAA, to write to you in response to your questions. A copy of his letter has been placed in the Library.Letter from David Gee, dated 29 May 2008:"Further to your two parliamentary questions to the Secretary of State the National Assessment Agency (NAA) has been asked to place this letter in the House of Commons library as a formal response.""Marker Standardisation & Mentoring""The process for the standardisation of markers for National Curriculum Tests has changed in 2008 and therefore there are differences in the way the margin of errors are recorded for the period from 2005 - 2007 and 2008. These are defined in the table enclosed.""From 2005 - 2007 standardisation and the first sample, submitted at the same time, were the main method for detecting markers who were unable to consistently apply the mark scheme. The second sample, two-thirds of the way through the process, was a final check, but it was rare for a marker to be stopped at this stage. This manual process involved supervising markers checking the marking of those in their team and was very dependent on the individual consistency of the team leaders.""From 2008, the introduction of the online standardisation and benchmarking processes means that monitoring of markers has become an ongoing process and poor marking can be identified and corrected at several stages throughout the marking activity. The procedures have been designed to detect those markers that are unable to consistently apply the mark scheme and stop them. It also allows team leaders to identify where markers may require mentoring support to improve their marking to ensure that all markers consistently apply the mark scheme.""The online system for standardisation and benchmarking allows ETS, the test delivery agency, to monitor all markers against an agreed national standard set of scripts at all points in the process; monitoring both marking quality, through benchmarking, and marking progress on a daily basis. This will highlight were additional support may be necessary.""Team leaders should provide feedback on the standardisation papers completed to all markers in their team. Any markers who failed their first attempt at standardisation will receive mentoring ""before a second attempt is made. For those markers who have passed standardisation, feedback should take place either before marking starts or very early in the process. Higher priority is given to those markers who have only just managed to pass standardisation. If after mentoring, a marker fails the second attempt at standardisation or benchmarking, the marker is stopped and removed from marking.""If you are interested further I would be delighted to meet with you once this years tests are completed to give you a full briefing on the work of the NAA in all areas of assessment and school support.""Please contact my office if you would like to arrange this."
Standardisation and Sampling Metrics 2005 to 2007¹ 2005 to 2007¹ Absolute mark difference allowed Key stage Subject Number of scripts in sample Discretional second sample Compulsory second sample 2 English reading 10 6-7 8+ English writing 10 13-22 23+ Mathematics 10 16-19 20+ Science 10 11-23 24+ 3 English reading 10 13-22 23+ English writing 10 24-33 34+ Mathematics 10 10-19 20+ Science 10 19-30 31+ ¹ Data derived from 2007 NCT Policy Document. Standardisation and benchmarking metrics 2008 2008 Standardisation 2008 Benchmarking (current 27 May) and may be subject to further tightening. Benchmarking is conducted typically every 90 scripts Key Stage Subject Number of scripts in sample Percentage exact agreement Percentage adjacent Percentage discrepant Number of scripts in sample Percentage exact agreement Percentage adjacent Percentage discrepant 2 English 5 reading and 10 writing 80 20 6 90 10 Mathematics 5 90 0 10 4 97 0 3 Science 5 90 0 10 4 95 0 5 3 English reading 10 80 20 6 85 15 English writing 10 80 20 6 90 10 Mathematics 6 90 0 10 4 97 0 3 Science 6 90 0 10 4 93 0 7
Secondary information
- Type
- Written question
- Reference
- 476 c1158-60W; 207868
- Session
- 2007-08
- Subjects
- Assessments National curriculum tests
- Link
- View this Written question on www.publications.parliament.uk
Librarians' tools
- Timestamp
- 2022-06-07 11:25:35 +0100
- URI
- http://data.parliament.uk/pimsdata/Hansard/PARLIAMENTARY_QUESTION_1327866
- In Indexing
- http://indexing.parliament.uk/Content/Edit/1?uri=http://data.parliament.uk/pimsdata/Hansard/PARLIAMENTARY_QUESTION_1327866
- In Solr
- https://search.parliament.uk/claw/solr/?id=http://data.parliament.uk/pimsdata/Hansard/PARLIAMENTARY_QUESTION_1327866