What is the a-value in digital assessment?

Statistical analysis helps you make a good (digital) test. One of the statistical concepts you can use is the a-value. But what exactly is the a-value of a question? And what use does a-value have in a (digital) test?

What is the a-value?

The a-value, or “Distractor-value,” represents the percentage or proportion of candidates who choose a distractor for a question in a multiple-choice format. In addition to the correct answer, each question has several distractors. By analyzing the a-value of each distractor, you can measure their quality, allowing you to fine-tune your tests for maximum effectiveness.

Item-level analysis: your key to insights and knowledge

Digital summative assessments aim to provide a final judgment on a candidate’s knowledge and insights. To ensure a test result accurately reflects the candidates’ preparation, examining and analyzing statistical data at the item level is crucial. The a-value offers valuable insights that empower you to improve the quality of individual test items.

A Real-World Example

Imagine an exam with 20 multiple-choice questions. Some questions may need to be stronger in terms of testing. By examining the a-value, you’ll know the percentage of candidates choosing a distractor. For instance:

Question: What colors does the Optimum Assessment logo have?

1. Blue and green
2. Yellow and black
3. Red and black
4. Red and Yellow

Answer 3 is correct, and the remaining answers are distractors. If the statistical analysis shows candidates rarely choose option A, it indicates that the answer option isn’t well-aligned with the question. In such cases, consider changing the distractor.

Understanding chance score

As a result, the probability score increases. The probability score is the score a candidate can obtain if he guesses/games the correct answer but does not actually know. An item with a low a-value contributes to your test not reflecting your candidates’ knowledge level. You can choose to change the distractor in that case.

A-value spectrum between 0 and 1

The a-value lies between 0 and 1. For example, suppose you have one correct answer option chosen by the exam candidates in half of the cases; then there is still a value of 0.5 distributed between the distractors. So you always calculate an a-value over answer options.

A low a-value increases the chance score—the score a candidate can achieve by guessing the correct answer without actually knowing it. Items with a low a-value contribute to a less accurate reflection of the candidates’ knowledge level. In such cases, consider changing the distractor.

More inspiration

Eager to explore more about the a-value or other statistical data related to digital testing? Visit our knowledge bank for information and inspiration. Stay ahead of the curve by following us on LinkedIn, and keep up-to-date with the latest developments in digital assessment.