- Foundation
- About 14 min
- 5 practice questions
Qualities of a good test: reliability, validity, objectivity, usability
Define reliability, validity, objectivity and usability in simple words, spot them in everyday examples, and see why a good test needs all four.
If something looks wrong, please tell us.
Not every test is a good test. A test can give different results each time, or measure the wrong thing, or depend on who marks it. Teaching-aptitude exams expect you to know four qualities that a good test should have. This lesson explains each with a simple everyday comparison.
Reliability: consistent results
A test is reliable if it gives consistent results. If the same learners took the same test again under the same conditions, without learning anything new in between, they should get roughly the same scores. Think of a weighing scale: if you stand on it three times and it shows 50 kg, 58 kg and 43 kg, it is not reliable. Unclear questions, very short tests and careless marking all reduce reliability.
Validity: measuring the right thing
A test is valid if it measures what it claims to measure. A mathematics word-problem test written in very difficult English may partly measure English reading, not mathematics, so its validity as a mathematics test falls. Think of the weighing scale again: a scale that measures your weight perfectly cannot measure your height. Validity is usually considered the most important quality, because a test that measures the wrong thing is useless however consistent it is.
Objectivity: free from the marker's opinion
A test is objective when the score does not depend on who marks it. Two fair examiners marking the same answer sheet should give the same score. Multiple-choice questions with a fixed key are highly objective. Essay answers marked without clear guidelines are less objective, because one examiner may value style and another may value content. A clear marking scheme or rubric improves objectivity, and better objectivity also improves reliability.
Usability: practical to use
Usability, also called practicability, means the test is practical in the real situation: it takes a sensible time, costs little, is easy to give, easy to mark and easy to understand. A perfect test that needs three hours per student and special equipment is not usable in an ordinary classroom.
| Quality | Simple question | Everyday comparison |
|---|---|---|
| Reliability | Are the results consistent? | A scale showing the same weight each time |
| Validity | Does it measure what it should? | Using a scale for weight, not a ruler |
| Objectivity | Is the score the same whoever marks it? | Two judges giving the same score |
| Usability | Is it practical to use? | A scale that is cheap and easy to read |
How they connect
A test can be reliable but not valid: a clock that is always exactly ten minutes fast is consistent but wrong. A test cannot be truly valid if it is not reliable, because results that jump around cannot measure anything accurately. So reliability is necessary for validity, but it is not enough on its own.
Worked example 1
A teacher wants to test Class 3 students' addition. She gives the sums as long stories with difficult words. Many good adders score poorly. Step 1: what was the test meant to measure? Addition. Step 2: what else did it measure? Reading of hard words. Answer: the test's validity is low, because scores depend on something other than addition. Simpler wording would raise validity.
Worked example 2
Two teachers mark the same science essay. One gives 7 out of 10, the other 4 out of 10, because there was no marking scheme. Step 1: does the score depend on the marker? Yes. Answer: the test lacks objectivity, and therefore its reliability also suffers. A clear marking scheme listing the points expected would improve both.
Reliability: consistent. Validity: measures the right thing, and matters most. Objectivity: same score whoever marks. Usability: practical. Reliability is needed for validity but does not guarantee it.
Key takeaways
- Reliability means consistency of results.
- Validity means measuring what is intended; it is the most important quality.
- Objectivity means scores do not depend on the marker; marking schemes help.
- Usability means the test is practical in time, cost and ease.
- A test can be reliable without being valid, but not valid without being reliable.
Check your understanding
-
1. A test gives nearly the same scores when the same students take it twice under the same conditions. The test is:
- Valid
- Objective
- Usable
- Reliable
Show answer
Answer: D. Consistency of results across repeated use is reliability.
-
2. A science test uses such difficult language that it mostly measures reading skill. Which quality is most affected?
- Usability
- Validity
- Objectivity
- Length
Show answer
Answer: B. The test is not measuring what it claims (science), so validity is low.
-
3. Which type of question is usually the most objective to mark?
- Long essay questions without guidelines
- Open discussion questions
- Multiple-choice questions with a fixed answer key
- Oral interviews without criteria
Show answer
Answer: C. With a fixed key, every marker gives the same score, so objectivity is high.
-
4. Which statement is correct?
- A test can be reliable without being valid
- A test can be valid without being reliable
- Reliability and validity mean the same thing
- Objectivity has nothing to do with reliability
Show answer
Answer: A. Consistent results can still measure the wrong thing, but results that are not consistent cannot measure accurately.
-
5. A test needs costly equipment and four hours per student, so schools cannot use it. It lacks:
- Validity
- Objectivity
- Reliability
- Usability
Show answer
Answer: D. Being impractical in time and cost is a lack of usability.
Sign in to keep track of the lessons you finish.