Skip to main content
Understanding Alva’s Test Methodology

FAQs on Alva's Test Methodology

Updated this week

Introduction

At Alva, we believe that psychometric testing should be fair, accurate, and useful — both for candidates and employers. That’s why our tests are built using modern test theory, instead of traditional methods based on static norm groups. This Q&A explains what that means in practice, and why it makes a difference.

1. Why doesn’t Alva use a traditional norm group like many other tests?

Traditional tests often rely on fixed norm groups that don’t update over time and may represent only a limited population. This can lead to unfair comparisons and outdated results. Instead, Alva uses a global standard and a dynamic, machine-learning-based model that continuously updates with new data. This ensures results are accurate, fair, and consistently comparable across roles, industries, and regions.

2. What does it mean that Alva uses modern test theory?

Modern test theory (IRT) goes beyond simply counting correct answers or adding up scores. Instead, it looks at how each individual question functions — for example, how difficult it is and how well it distinguishes between people with different levels of ability or traits. This allows us to estimate a candidate’s actual logical ability or personality trait level, rather than just comparing them to others.

3. Who are candidates compared to when taking Alva’s tests?

All candidates are benchmarked against a global working population, rather than a narrow or local sample. This makes results relevant across different regions, industries, and job roles.

4. How do we know the tests are accurate and fair for different people?

During the continuous development of Alva’s tests, we’ve used multiple large and diverse samples to build and refine our models. For example, more than 400,000 individuals have participated in building and standardising the personality test. The samples used include people from a wide range of industries, roles, countries, and backgrounds — helping us ensure that the tests work reliably and fairly for a truly global working population.

In conclusion

By using modern test theory and continuously learning from new data, Alva ensures that test results are:

  • Fair – Everyone is measured against the same global standard

  • Accurate – Question-level data leads to more precise results

  • Up-to-date – The model adapts as new data comes in

  • Easy to use – One consistent scale makes interpretation simple

This approach makes Alva’s tests a reliable foundation for hiring decisions – today and in the future.

Want to dive deeper?

You can explore the samples and methodology in more detail in our technical manual here and here.

Did this answer your question?