Monitoring Accuracy

Monitoring Accuracy

The monitoring accuracy tool is designed to help you quickly identify subjects where the accuracy of predicting might be a cause for concern.

Senior Leaders, Subject Leaders and subject staff can track historical and in-year data within their subject area and teaching set. The subject leader can monitor how individual teaching sets are performing against the overall Subject Alps Grade.

Note: throughout this article I will refer to gradepoints – these are individual points across the year where your Data Manager would have added data to Connect. They will either be:

  • Monitoring gradepoints – these are in-year gradepoints based on predicted teacher assessed grades / raw mock grades etc. These will be labelled according to a convention in your school / college, e.g. 21.Y13Jan would be my January predictions for the 2020/21 Year 13 group
  • Examination gradepoints – these are the actual examination outcomes for that Key Stage and will be labelled by curriculum year, e.g. 2020/21. There will only ever be one of these per Academic Year.

Alternatively, you can watch our short videos on Monitoring Accuracy below:

Accessing Monitoring Accuracy

On your Connect Homepage, you will see an option called Monitoring Accuracy:

You now have three options - hover over the navigation header to click between the pages:

  • Trends – Difference: This area allows you to track between any monitoring point and an examination point. 
  • Trends – Variance: This area allows you to track across up to 4 in-year monitoring points.
  • Trends - Exam Vs Monitoring: This area allows you to view a comparison between your selected exam Gradepoint and the latest in-year monitoring Gradepoint for the same cohort of students.

The purpose of this article is to describe the powerful potential of these two areas to help identify any issues with the accuracy of predictions, so that you can set priorities going forward.

Let’s look at in-year trends first.

1. Trends - Variance

The best way to use this section is to track the accuracy of predicting across your in-year monitoring gradepoints. You can use examination data, and I will show you how you might use a ‘monitoring sandwich’ in this section.

In this section you can compare multiple gradepoints. Step 1 is to choose your gradepoints. Click on the gradepoint selector at the top of this page.

This will take you to the gradepoint pop-up. Here I have selected 4monitoring gradepoints for Year 13 that I have in Connect. 

Option 1 – In-year monitoring gradepoint tracking

The gradepoints selected above allow me to track across my monitoring points in Year 13 as shown below:

Let’s analyse this page. You can see that the Business Studies department have been consistent in their predicting. You might want to do a monitoring sandwich below to check if these are in line with examination performance. Economics on the other hand, have jumped considerably from the first monitoring point – you can see this in the arrows in the boxes, and by the largest difference column – at 0.29, that is approximately every student performing a grade and a half higher than at the beginning of the year between monitoring points 1 & 4.

The graphs

You can interact with the graphs in the trends column to get even more analysis by clicking on the subject in the table.

I am going to demonstrate this by clicking in the Biology subject page.

In this view you will now be able to see any teaching sets and teachers, if your data manager has added these to Connect.

Important Note: if teachers teach more than one set, the values are amalgamated, and teachers may not have taught a group the whole way through, therefore sticking with teaching sets is possibly more informative.

By using the dropdown, you can add teaching sets to the graph and the Alps grades for that set are superimposed onto the graph. This way the Subject Leader or teacher can get a feel for how that set is performing in relation to the group, and indeed, if there are causes for concern in monitoring accuracy.

Option 2 – the monitoring sandwich

By choosing gradepoints sandwiching monitoring gradepoints between consecutive examination outcomes, you can identify issues relating to monitoring trends. In the examples below, I have chosen a monitoring point for Year 13 - 21.Y13 Pred and selected to view this sandwiched between their outcomes for 2020/21 and the previous year’s results in 2019/20 for comparison.

2. Trends – Difference OR Exam Vs Monitoring

Select any two Gradepoints and compare the difference between them. 

You can mix and match gradepoints - by this we mean that you can choose
  1. two monitoring points to compare
  2. two exam gradepoints
  3. one monitoring and one exam gradepoint 

Select your gradepoints manually by clicking in the gradepoint selector in the top corner of the screen. 

Comparing retrospective monitoring gradepoints against examination outcomes

In this example, we are looking at a monitoring gradepoint for last year’s Year 11 (21.Y11 Pred), and comparing this to their final outcomes in 2020/21. I can use the filters on the gradepoint selection tool to manually select these options. 

You can now see both gradepoints next to each other and compare the accuracy of the Alps prediction grades against the actual outcome grade for each subject.

A quick review of this table highlights some talking points:

  • I can quickly see that in Computer Science the Alps predicted grade was an Alps 6, yet the final outcome was an Alps 4 with a change in Alps Score of 0.22. This means that their predictions were harsher than their final outcomes - perhaps this led to smart interventions? Or perhaps their interventions were so successful that the student made rapid progress toward the end of the course. Either way – worth investigating.
  • Your English Literature department was spot on, and perhaps they could lead on some CPD sharing good practice across the school / college.

Note how in this view the %student grade column gives a value for the proportion and number of students with a different grade. This is because both gradepoints contain the same students. 

Comparing current year progress against previous examination outcomes

In this example, I am comparing the monitoring point of a Year 13 group with the examination results from the previous year. Yes, this is a different cohort, but it gives me a feel for future subject trends.

This time there are no values in the % Student difference column as these gradepoints are comparing different student cohorts. 


There are lots of ways in which you can use this feature. This article has been written to give you a few scenarios which might help get you started, but you will find more ways to set up custom gradepoints to suit your own needs.

This area will help you to be smarter with your time when identifying priorities around predicting accuracy and in the tracking of teacher assessment grades across the year.


    • Related Articles

    • Monitoring accuracy across all subjects

      We have devised a series of video shorts to help you get the most from the tools and features in Connect. The Monitoring accuracy area allows you build up a picture of the accuracy of predicting across your subjects empowering you to ask searching ...
    • Monitoring accuracy within a subject area

      We have devised a series of video shorts to help you get the most from the tools and features in Connect. Use the monitoring accuracy areas to identify discrepancies between the predicted outcomes of teaching sets. ​
    • Creating your Monitoring Point Zero

      The Monitoring Point Zero (MPZ) gradepoint in Connect is an important starting point for your Year 13 students and helps staff set ambitious targets for their groups in the final year leading up to their A levels. This gradepoint is based on the AS ...
    • Creating a Monitoring Point

      Before creating a monitoring point, you first need to create a student cohort in Connect Data for your current year. Click here to see how to create a student cohort. You will know if you haven’t created a student cohort for your current year if the ...
    • Consortium best practice for monitoring

      This article provides guidance on how best to manage your in-year Consortium monitoring data to allow you to analyse your ‘on roll’ and ‘taught at’ groups separately.  We also have an article on how you can analyse your Consortium data using Connect ...