Test security isn’t one-size-fits-all. It means more (or less!) than locking down content or banning calculators. What should it mean for your assessments? Here, we’ll address a few practical questions that move beyond common checklists.
Are they the right goals for your programme?
What does secure actually mean?
We often mean that test takers:
• have not seen the test content before, and
• answer the questions without assistance from people or technology.
Start with your test’s purpose
If you want to measure what a test taker can do independently, with no support, then a tightly controlled, ‘high security’ environment is an appropriate (if challenging) aim.
However, if authenticity is a priority – if you want the test to reflect real-world practice – you might want to allow access to:
• calculators or grammar checkers
• reference materials
• even generative AI tools.
Controlled collaboration may even form a key part of the assessment.
In these cases, the aim of ‘securing’ the test is not to eliminate all support, but to ensure that only the right kind of support is available.
So, what does appropriate security look like for your test?
Begin with a few key questions:
• What do I want the test results to prove?
• For example, do professionals in the target use domain use resources and tools? How do they collaborate?
• Should my assessment reflect those realities – or is there a good reason to isolate individual ability?
From there, you can focus on protecting what matters, not simply following tradition.
It means aiming to ensure that the test taker:
• has not had unauthorised prior access to test materials,
• is engaging in only permitted collaboration, and
• is using only permitted tools.
The practitioner’s challenge
Absolute security is an illusion, whether the test is pen-and-paper or digital.
Appropriate security balances questions of test validity, authenticity, accessibility and practicality.
It is also like a chain – each link is important. Overengineering any one link is pointless. Overlooking another is risky.
In our next post, we’ll tackle a key issue for many testing programmes: Test content exposure and theft – what it means, how it happens and what you can (and can’t!) do about it.