Achievement Tests
Achievement Tests
An achievement test measures what a child has actually learnt in a subject after teaching. It differs from an aptitude test, which predicts future capacity, and from a diagnostic test, which locates the precise point where learning broke down. Confusing the three is the commonest error in this block.
Three tests, three purposes
| Test | Question it answers | Timing |
|---|---|---|
| Achievement test | What has the child learnt? | After teaching |
| Aptitude test | What can this child become good at? | Before training or selection |
| Diagnostic test | Exactly where and why is the child failing? | After a weakness appears |
| Prognostic test | How well will the child do in this subject ahead? | Before starting a course |
Qualities of a good test
- Validity — it measures what it claims to measure. A reading comprehension passage full of unfamiliar technical words tests vocabulary, not comprehension.
- Reliability — it gives consistent results on repetition or across markers.
- Objectivity — two examiners give the same score to the same answer.
- Usability — practical in time, cost and ease of scoring.
- Comprehensiveness — covers the whole content and range of objectives, not only recall.
A test can be highly reliable and still invalid — a perfectly consistent test of the wrong thing. Validity is therefore the first quality, and reliability is a condition for it.
Building a fair achievement test
A blueprint maps every question against content and objective, so that a Class 5 EVS paper does not become twenty recall questions from two chapters. Objectives are usually taken from Bloom's taxonomy — knowledge, comprehension, application, analysis, synthesis and evaluation in the original, revised as remember, understand, apply, analyse, evaluate and create.
- Mix objective type, very short answer, short answer and long answer questions.
- Include application items, not only recall: 'why does the ice melt faster here?' rather than 'define melting'.
- Keep the language simpler than the concept being tested.
- Weight the marks to the teaching time actually spent on each unit.
✗ A test with a high pass percentage is a good test. | ✓ A good test is valid, reliable and objective; the pass percentage says nothing about its quality.
संकेत: उपलब्धि परीक्षण = क्या सीखा; अभिक्षमता परीक्षण = क्या सीख सकता है; निदानात्मक परीक्षण = कहाँ अटका है।
One question per paper, usually asking which type of test a described situation calls for, or which quality of a good test is being described. The repeated trap uses the word 'diagnostic' where the situation actually describes an ordinary achievement test. Remember the direction: achievement looks backward at what has been taught, aptitude looks forward at potential, diagnostic looks inward at the point of failure. Blueprint and Bloom's taxonomy also appear as direct one-line questions.
Frequently asked questions
What is the difference between an achievement test and an aptitude test?
An achievement test measures what a child has already learnt after being taught a particular content. An aptitude test estimates the potential to learn or succeed in an area in future, and is used for guidance or selection rather than for grading.
What makes a test valid?
A test is valid when it measures exactly what it claims to measure. A maths word problem written in difficult language partly measures reading ability, which lowers its validity as a test of mathematics.
What is a blueprint in test construction?
A two-way table that maps the marks of a paper against content units on one axis and instructional objectives on the other, ensuring balanced coverage and preventing a paper made only of recall questions from a few chapters.
60-second recap
- Achievement measures what was learnt; aptitude predicts potential; diagnostic locates the breakdown.
- Good test qualities: validity, reliability, objectivity, usability, comprehensiveness.
- Validity is first — a reliable test of the wrong thing is useless.
- A blueprint balances content against objectives from Bloom's taxonomy.
- Weight marks according to the teaching time each unit received.