Adaptive assessments - clever exams

Adaptive tests are sophisticated assessments that can modify the level of content depending on the candidate’s abilities. This means that if the candidate is finding the questions difficult, the level will adjust to a more appropriate level; likewise if the content is too easy the level of difficulty will increase. The goal of adaptive assessments are to find the candidate’s optimum difficulty level; this is often coupled with a standardised score - so results are reflective of how the performance compares to other candidates. All of the tests at APT are fully adaptive. Our aim is to identify areas of proficiency and weakness and drill into specific skills to identify academic strengths (or gaps) to determine.


There are numerous advantages to testing in this way. It can significantly reduce the amount of time it takes to effectively judge a level of ability than traditional tests and examinations. Candidates tend to be better motivated by an assessment that doesn’t continually ask questions that are either too challenging or too easy, and a single adaptive test can potentially measure a broad range of ability. Security is a great advantage - with large enough question banks, the likelihood of two candidates experiencing the same sequence of questions is highly unlikely. Given some of the challenges faced by schools and colleges in arranging exams to be sat during the pandemic you can understand why interest in adaptive testing has risen sharply in recent months.


Adaptive tests do not necessarily work in all scenarios, and they are best suited to skills-based profiling rather than learned knowledge or interpretive examinations. They have been used extensively for literacy and maths assessments over the past 20 years where there is a wide range of ability across all age levels. More sophisticated tests of this nature, such as the APT, can be designed to provide additional insight into specific sub-skills by associating each question with the skills it is designed to assess. This allows a single test to evaluate a broad range of skills using just one short assessment - in the past this would have taken multiple assessments over time to build a similar picture of a student's academic skills.



There are a few significant challenges in developing adaptive tests:

  • An adaptive test is only as strong as the engine behind it, and they employ quite sophisticated programming and algorithms. Therefore there are barriers to developing such tests without the software and expertise to create one that works effectively.

  • Given that the previous answer determines which question will appear next, a significant number of questions are required in these kinds of tests. Not only that, but each question needs to have been levelled (for difficulty) and validated to fit within the battery.

  • You need to be confident with the difficulty of each item and what skills it assesses (it is often more than one). APT is proud to be working with Atom Learning allowing us to provide tests with huge test batteries, teacher-written content and AI-validated questions.

  • Another challenge is that candidates can’t go back and check their work or change an answer - once an answer is picked - you have to move on; this can be hard for students who are not confident or rushing through because of time restraints.

  • The question format is generally multiple choice - meaning that the candidate always has a 1 in 4 (or 1 in 5 depending on the multiple choice format) chance of randomly selecting a correct answer - this means the test has to be carefully calibrated to account for such occurrences.


High stakes or Low stakes - what does this mean?


Adaptive tests work well for low stakes tests. The ‘stakes’ of a test are the possible uses and outcomes of it. An entrance exam or formal qualification such as a GCSE or A-LEvel is considered high stakes because there are future outcomes dependent on these results. Typically, the more high-stakes a test is, the more has to be done to ensure the result is valid, a fair test, and a universal experience is delivered to all candidates. These tests also have to take additional measures against potential cheating as there are potentially life-altering outcomes up for grabs.


In comparison, a personality test or skills-based evaluation is considered low stakes - little rests on the outcome of such measures, but this doesn’t diminish the potential value such insight can deliver, particularly when coupled with expert analysis and professional feedback. At APT we do everything in our power to keep our tests low-stakes - to reduce stress on our candidates and focus attention on the value of the results generated. APT is only available with expert analysis built-in thanks to our post-assessment feedback sessions which provide high quality insight, supporting candidates with their immediate academic focus, future choices and educational journeys.


At APT we do everything in our power to keep our tests low-stakes - to reduce stress on our candidates and focus attention on the value of the results generated.

We have seen the popularity of adaptive tests increase dramatically during the pandemic. They can be delivered through online systems - where candidates only need an internet connection to access these sophisticated tests. We know that many parents are worried about the impact of two-years disruption to their children’s schooling and as a result, such tests are so needed to identify gaps in learning and skills. APT has been developed specifically to address these needs and is supporting families considering where to focus additional support and to plan for their children’s future. If you would like to find out more, feel free to contact us and we will be happy to assist in any way we can.