##
# Variability in results for schools and colleges in 2018: A level

# Variability in results for schools and colleges in 2018: A level

These interactive graphs show the centre variability at grade A and above for eighteen larger entry subjects in A level. For more explanation of these graphs, click on the Questions and Answers tab, or
read the reports
on A level centre variability 2016 to 2018.

For any feedback on these graphs, please contact
analytics@ofqual.gov.uk.

####
**1. What do the graphs show?**

The graphs show the year-on-year variation seen in centres (schools or colleges) in England at grade A and above for a particular A level subject. The graph on the left shows this year's variation (ie between summer 2018 and summer 2017); while the graph on the right shows the previous year's variation (ie between summer 2017 and summer 2016).

- Each bar represents the number of centres with a particular level of variation, measured in intervals of 2.5 percentage points. For example, the bar immediately to the right of zero on the x-axis represents centres that had an increase of up to 2.5 percentage points (ie centres where up to 2.5% more students received grade A and above, compared to the previous year).
- Bars to the right of zero indicate centres with increases in the percentage of students receiving grade A and above. Bars to the left indicate centres with decreases in the percentage of students receiving grade A and above. The bars further away from zero indicate centres with larger changes in outcomes between years.
- The higher the peaks in the middle and the narrower the distributions, the greater the stability from one year to the next.

####
**2. Which subjects have been included?**

We have included those subjects at A level which have a large enough entry to support the analysis. These are subjects with an entry of 10,000 or more candidates and 100 or more centres with more than 20 students in the subject in question.

####
**3. Which centres are included in the graphs?**

Each graph shows data for centres which have entries in both years covered in that graph. Centres which had an entry in only one of the two years shown in a graph are not included. So a centre with an entry in 2017 but not in 2018 would not be included in the left hand side graph.

We also only include centres that have an entry (in both years) of at least 20 students in the subject in question. We do not include centres with smaller entries because just one or two students' grades can make the centre appear to have quite a lot of variability.

The graphs represent all types of centre (eg comprehensive, academy, free school, independent etc).

####
**4. What age are the students in the centres included in the graph?**

It is possible to see centre variability according to all students in the centres, regardless of age; or to see centre variability just for those students in year 13 (ie aged 18 in the academic year), the target age for A levels.

####
**5. What is meant by centres with a 'stable' entry?**

In these graphs, it is possible to have 'all centres' or 'centres with a stable entry'. Centres with stable entry are those centres which entered very similar numbers of students in the subject in both years shown in a graph. By very similar numbers we mean within +/- 15% of the previous year.

These centres typically show less variation in results. This is probably because large changes in entry in a subject within a centre probably means other things have changed such as the overall profile of students or the nature of provision for the teaching of the subject.

####
**6. What does the mean value on graphs represent?**

The mean on each graph is the average difference between the two years for all of the centres presented in the graph. For example, a mean of -0.6% indicates that, on average, centres had a 0.6% drop in the number of students attaining a grade A and above at A level.

####
**7. What is the standard deviation (SD) shown on graphs?**

The standard deviation (SD on the graphs) is a measure of the spread of the variation - a lower SD means there is less variation overall whereas a higher standard deviation means there is more variation. A small SD means that most centres had low levels of variability.

####
**8. How should we interpret the SD?**

We include two graphs so it helps to interpret the SD. In general, where specifications and overall cohorts have not changed, we might expect that the SDs in the two graphs are similar. Some of the subjects at A level are being assessed for the first time and so we might expect to see increased variability in these subjects as centres become familiar with the new content and assessments. More information on the A level reform timetable is available
here.

####
**9. Some graphs don't have many centres in them. Does that matter?**

Where graphs have a small number of centres presented (less than 50), the graphs need to be interpreted with caution. Where the combinations selected (of subject, stability, centre size, etc.) produce graphs with a relatively small number of centres, these graphs say little about the overall picture, and more about the individual centres. For this reason, only graphs with more than 20 centres are displayed.

####
**10. Where does this data come from?**

Data is supplied to Ofqual by examination boards shortly before results day. At the point in time Ofqual received the data, all certificates and entries may not have been fully processed. Additionally, the data does not reflect any changes to grades from post-results enquiries.

Return to the
Ofqual Analytics
home page.

Wherever you use this information, please attribute it to Ofqual and provide a link back to this page where possible.

If you need an accessible version of this information to meet specific accessibility requirements, please email
publications@ofqual.gov.uk
with details of your request.