The tables below show how points are allocated to pupils' Key Stage 3 and GCSE/GNVQ results.
| Key Stage 3 results |
|
Level |
Points |
|
Exceptional Performance (EP) |
9 |
|
8 |
8 |
|
7 |
7 |
|
6 |
6 |
|
5 |
5 |
|
4 |
4 |
|
3 |
3 |
|
2 |
2 |
|
1 |
1 |
|
Working towards level 1 |
0 |
|
Failed to register a level (N) |
0 |
|
Disapplied |
0 |
|
Absent |
Disregarded |
|
Missing |
Disregarded |
|
GCSE results |
|
GNVQ results |
|
GCSE grade |
No of points |
|
GNVQ grade |
Full GNVQ |
Part One GNVQ |
|
A* |
8 |
|
Intermediate
Distinction |
30 |
15 |
|
A |
7 |
|
Intermediate
Merit |
24 |
12 |
|
B |
6 |
|
Intermediate
Pass |
20 |
10 |
|
C |
5 |
|
Foundation
Distinction |
16 |
8 |
|
D |
4 |
|
Foundation
Merit |
12 |
6 |
|
E |
3 |
|
Foundation
Pass |
6 |
3 |
|
F |
2
|
|
|
G |
1
|
|
For GCSE (Short Course) grades, the number of points is divided by 2.
| The input measure |
| The Key Stage 3 input measure for each 15 year old pupil (including pupils disapplied from assessment, but excluding pupils absent from all of the tests) on roll at the time of the 1998 Schools' Census is the average level achieved in the tests/tasks in each of the core subjects. The following conditions apply:
pupils absent from all of the tests are disregarded for the purposes of calculating value added;
disapplied pupils, pupils working towards level 1, pupils who failed to achieve a level for other reasons, or pupils who took a task and the level is unknown are regarded as level "zero". |
| The input measure is calculated as the arithmetic average of the test or task levels achieved:
for a pupil with test/task levels for each of the 3 core subjects - if the test results were 5, 5 and 6, the average for that pupil is the sum of 5, 5 and 6 divided by 3 which is 5.33;
if a pupil is absent from one or two tests then the test(s) is/are ignored for the purposes of the calculation of the average. For example, if a pupil has only two test levels of 5 and 6, the average for that pupil is the sum of 5 and 6 divided by 2 which is 5.5, or if a pupil has only one test level, the average is taken as that single level.
|
| The output measure |
| The output measure for each 15 year old pupil on roll on the day of the 1998 Schools' Census is calculated as the total GCSE/GNVQ point score using the scores given above. For example,
a pupil achieving 5 grade As, 2 grade Cs and an Intermediate GNVQ Part One (Merit), the total GCSE/GNVQ point score will be (5x7) + (2x5) + (1x12)x2 = 57 points.
|
| Calculation of the pupil value added measures |
| Full value added is based on comparing a pupil's GCSE/GNVQ performance with that of other pupils with similar Key Stage 3 results. The value added measure for an individual pupil is calculated as the difference between their total GCSE/GNVQ point score and the median total GCSE/GNVQ point score for all pupils nationally with a similar average Key Stage 3 score. The median is the middle value - with half of pupils having a GCSE/GNVQ point score at or below the median, and half at or above. |
| During initial investigations into the value added measures to be published in this pilot, it was concluded that special schools should be treated separately from mainstream schools for the purpose of calculating value added measures. The charts below give the pattern of median GCSE/GNVQ total point scores across the range of average Key Stage 3 scores for mainstream schools and special schools - joining the median gives the 'national median line'.
National median line
mainstream schools
 | National median line
special schools
 |
|
| For example, School A, a mainstream school, has 11 pupils with Key Stage 3 and GCSE/GNVQ results shown below. Median total GCSE/GNVQ point scores have been allocated according to the national median line for mainstream schools shown above, and a value added score calculated for each pupil.
| |
Average KS3 level |
Total GCSE/GNVQ point score |
Median Total GCSE/GNVQ point score for KS3 bin |
Value Added score |
|
Pupil 1 |
5 |
37 |
41 |
-4 |
|
Pupil 2 |
5.33 |
46.5 |
46.5 |
0 |
|
Pupil 3 |
5.33 |
56 |
46.5 |
9.5 |
|
Pupil 4 |
5.33 |
58 |
46.5 |
11.5 |
|
Pupil 5 |
4.67 |
39 |
36 |
3 |
|
Pupil 6 |
4.67 |
49 |
36 |
13 |
|
Pupil 7 |
5.33 |
43 |
46.5 |
-3.5 |
|
Pupil 8 |
4.33 |
25 |
31 |
-6 |
|
Pupil 9 |
4.33 |
35.5 |
31 |
4.5 |
|
Pupil 10 |
4.33 |
33 |
31 |
2 |
|
Pupil 11 |
5 |
34.5 |
41 |
-6.5 |
|
| Calculation and presentation of the school value added scores |
| A school's value added score will be a simple average (arithmetic mean) of the value added measures for all the pupils in the school. For example, the value added score for School A above is the sum of its value added scores divided by 11 (the number of pupils). This gives School A a value added score of 23.5 /11 = +2.1. |
| The pilot offers an important opportunity to explore the options for presenting value added in performance tables and uses banding and numeric scores. The bands use letters A to E indicating whether a school fell in the top 5% of schools, the next 20%, the middle 50%, the lower 20%, or the bottom 5%. |
| The cut-offs for the bands used in this pilot for both mainstream and special schools are shown below and relate to a school's value added score, e.g. School A with its value added score of +2.1 is a Band B school.
| |
A |
B |
C |
D |
E |
|
Mainstream |
5.5 and over |
Between 1.2 and 5.5 |
Between 1.2 and -3.1 |
Between -3.1 and -7.1 |
Less than -7.1 |
|
Special |
5.0 and over |
Between 2.9 and 5.0 |
Between 2.9 and -10.3 |
Between -10.3 and -13.5 |
Less than -13.5 |
|
| Different measures of value added |
| More than one measure of value added has been published for each school: measures are shown by gender and differing performance at Key Stage 3. Numerical scores and bands have been published for the overall value added measure, and for measures calculated for pupils with differing performance at Key Stage 3. |
| Differing levels of Key Stage 3 performance were calculated from the national distribution of pupils' average Key Stage 3 levels. The average Key Stage 3 level for an individual pupil can only take the form of a whole number of levels, or a half, a third or two thirds of a level, i.e. an average level of 4.5 can be achieved, but not a level of 4.4. Therefore, the distribution of average Key Stage 3 level is made up of discrete values. |
| The average Key Stage 3 level distribution was divided into four parts (quartiles) with approximately the same number of pupils in each quartile. An exact 25% in each quartile could not be achieved because of the discrete nature of the distribution. The cut-off points for the three groups of differing Key Stage 3 performance were chosen to be the upper quartile and lower quartile values, with 25% of pupils being in the 'low' performance group, 50% of pupils in the 'medium' performance group, and 25% of pupils in the 'high' performance group. |
| Separate cut-off points for defining differing Key Stage 3 performance were calculated for mainstream and special schools and are shown in the table below.
| |
Performance at Key Stage 3 |
|
'Low' |
'Medium' |
'High' |
|
Mainstream schools |
4 and below |
4 and 5.66 |
5.66 and above |
|
Special schools |
2.5 and below |
Between 2.5 to 4 |
4 and above |
|
| Each school, if it had pupils in the relevant category, has a value added measure for each of the three groups of Key Stage 3 performance. The score is calculated as a simple average of the value added measures for all the school's pupils within the relevant Key Stage 3 performance group. |
| These scores are also shown as A-E bands, using the same cut-off values given above. The distribution of school value added scores within each Key Stage 3 performance group will not be the same as the national distribution used to calculate the cut-offs. Therefore, the distribution of the A-E bandings within each Key Stage 3 performance group need not represent the same 5, 20, 50, 20, 5 per cent distribution of schools in the A-E bands nationally.
|
| Calculation of stability measure |
| The stability measure provides an indication of the mobility of pupils, i.e. it shows the proportion of pupils that has remained in a school between Key Stage 3 and GCSE/GNVQ. It is calculated as:
the number of 15 year old pupils who took Key Stage 3 tests at a school minus those who left before January 1998 as a proportion of the total number of 15 year old pupils on roll in 1998.
|
| Calculation of value added coverage |
| The value added coverage shows how many 15 year old pupils were included in the calculation of each value added measure. Pupils who are absent or missing from all three subjects at Key Stage 3 are excluded. It is calculated as:
total number of 15 year olds included in the calculation of the value added score divided by the total number of 15 year old pupils on roll.
|
| Calculation of national average |
| The national value added score is a simple average (arithmetic mean) of the value added measures for all pupils in the sample of pilot schools. The national average can be negative because the distribution of pupil value added scores about the median value is negatively skewed - poorer performing pupils are further below the median line on average than better performers are above. |
