Assessment models in examination PE courses - Part 1

This post is written for any PE teacher or course manager working in an examination PE course. The focus of the six assessment models is on exam assessment rather than written, oral or performed NEA. If this is of interest to you, please read on.

Dear PE colleagues,

For the purposes of this post, I am going to write six alternative assessment models for the exam components of a PE course. I will write three in Part 1 and three in Part 2. I am choosing to write about AQA GCSE PE, as it is the most popular exam-centred PE course in the UK, but these assessment models could be applied to any course, whether a different GCSE, an A-level or a technical qualification of any kind.

A few FYIs before we dive in…

Because I am going to focus on AQA GCSE PE, it’s worth you knowing the following in case it is not your course. AQA GCSE PE:

• has two exam components…
• …totalling 156 raw marks…
• …making up 60% of the entire course assessment;
• both exam components contain two pieces of extended writing;
• both exam components contain one piece of synoptic writing.

Therefore, the need to get the exam assessment model right for one’s students is very significant.

The six models I am going to present to you are:

Part 1:

• Model 1: End-of-unit core assessments
• Model 2: Skill-based, end-of-unit core assessments
• Model 3: Spaced, end-of-unit core assessments

Part 2:

• Model 4: Interleaved, end-of-unit core assessments
• Model 5: Unannounced to announced core assessments
• Model 6: end-of-unit core assessments

Model 1: End-of-unit core assessments

This is probably the standard or most common model of assessments within a course. The idea is that a range of content is developed over a series of lessons and homeworks and then a test is sat and marked to provide insight into the degree of learning that has occurred. For example, the model might look like this within an AQA GCSE PE course in early Year 10:

The model will then repeat to some degree:

The cycle could then continue and teachers can expect that there would be in the region of two core assessments per half-term (based on a seven-week half-term). As a result, a teacher may develop a mark book that looks something like this:

So, let’s evaluate this model. It certainly has advantages. They are:

• It’s easy to implement.
• Standardised core assessments can be developed over time.
• There are clear expectations due to regular and predictable schedules.

But the model also has some serious limitations:

• Core assessments occur after the game (learning) has ended. This provides post-game data only.
• Content is assessed, rather than content and skills.
• Content from weeks 1, 2, and 3 is not reassessed until a mock exam.
• Each block of content is assessed independently. For example, movement analysis and levers are assessed separately from joints and the musculoskeletal system.

Model 2: Skill-based, end-of-unit core assessments

Model 2 is much like model 1 but it suggests that teachers need to be more aware of the skills as well as the content focus of core assessments. Take a look at this example:


This is the analysis of how classification of skill has been assessed on AQA GCSE PE between 2018 and 2023. Furthermore, AQA asked this question in 2024:

So, what assumptions can be drawn about the nature of practice assessments and core assessments which contain the topic of classification on AQA GCSE PE? All of the following seem accurate. Classification of skill is:

• assessed on every paper;
• assessed across AO1, AO2 and AO3;
• assessed more heavily on AO2 (application and placement) than AO1 (knowledge);
• assessed more heavily on AO3 (justifying placement) than AO1 (knowledge);
• assessed with the biggest weighting of marks for the ‘Justify’ command.

Now, all of this detail allows a teacher to more accurately present core assessment activities with skills in mind. It is very likely that, with knowledge of the data above, a PE teacher will structure their core assessments with a specific skill focus. Yes, a teacher would ask for an AO1 description of a gross skill, say, but they are highly likely to spend more assessment as well as learning focus, achieving a range of skills and, specifically, aiming for the ‘Justify’ skill as the peak experience.

Now, you may be thinking that this is all well and good but you don’t have the data to build these skill-based assessment models. You do! Every client of The EverLearner with access to our ExamSimulator software also has access to our Data Centre. The Data Centre specifically informs PE teachers about how to create skill-based assessments.

Finally, for this model and by way of provocation, how do you feel if I write the statement below:

The aim of core assessment is not to assess knowledge of content. It is to use knowledge of content to assess and, ultimately, develop the writing, drawing and calculating skills of our students.

I could take this further:

The aim of our teaching is not to teach knowledge of content. It is to use knowledge of content to teach and develop the writing, drawing and calculating skills of our students so that they can be used for anything throughout the student’s life.

With conversations like this, we always need to consider reality and my role in writing these statements is not to cause you to massively deviate in your practice. Rather, ponder these suggestions philosophically first and then, trial different approaches before making wholesale changes. No one needs to throw the baby out with the bathwater.

Now, let’s evaluate the skill-based assessment model:

• Explicitly develops skills as well as content.
• Skills are assessed and assessed in context over time.
• Standardised core assessments can be developed over time.
• There are clear expectations due to regular and predictable schedules.

But the model also has some serious limitations:

• Data analysis is required –please talk to us at The EverLearner Ltd about this.
• Core assessments occur after the game (learning) has ended. This provides post-game data only.
• Content from weeks 1, 2, and 3 is not reassessed until a mock exam.
• Each block of content is assessed independently. For example, movement analysis and levers are assessed separately from joints
• and the musculoskeletal system.

Model 3: Spaced, end-of-unit core assessments

The idea of spacing is that students can learn and be exposed to concepts over a longer period of time. According to Thalheimer, this means that students will take longer to forget the learned concepts.

And this is not the end of it. By creating a structure that causes students to engage with previously learned concepts as part of the core experience, you will be causing benefit by spacing.

So, how can it work?

Put bluntly, if the musculoskeletal system is the first topic in your course, you need to ensure that ongoing assessments cause the students to return to those early learning concepts again and again.

So a non-spaced assessment model might look like this:

Notice that there is no spacing going on here. A spaced assessment model might look more like this:

Notice how the musculoskeletal system will be touched upon again and again and again. Therefore, the tendency to simply forget those concepts is much lower and, if they are forgotten, the experience of forgetting can be tracked, observed and acted upon.

What is pictured above probably relates to exam assessments involving prose-based writing and drawing but it is also worth considering that this model does not only need to be applied to exam practice. The same can be applied to quizzing and even teaching. There is no need to view any course experience as “one-and-done”. Spacing helps us to challenge this assumption.

The following are some of the advantages of spacing:

• It keeps old learning alive for longer.
• It can be modelled to other learning experiences including quizzing.
• Standardised core assessments can be developed over time.
• It keeps students on their toes.
• It honours the probability of forgetting taking place and provides a structure to identify the nature of forgetting.

But the model also has some serious limitations:

• Assessments get longer over time.
• Some of the content that is assessed together has no association. Look at the interleaved assessment model to challenge this weakness.
• Core assessments occur after the game (learning) has ended. This provides post-game data only.

Conclusions

So, there we have it for Part 1. I strongly encourage you to return for Part 2, specifically for the interleaved and the unannounced models, as these are two that I personally use. 

As mentioned at the start of the post, I do not consider these six models to be exhaustive. There are loads, right? Also, they can be blended and intertwined and I hope you may consider developing an assessment model that suits you and your course best.

Finally, I want to make a really important point that I will return to in Part 2:

The aim of assessment is not to track data/performance. No! It is to provide the relevant foundation for learning-based interventions on a structured basis.

Thank you for reading. Have a lovely day.

James