This practice test covers measurement, assessment and evaluation, scales of measurement, validity, reliability and item analysis for teacher recruitment test preparation. Read each question carefully and choose the best answer.
Measurement, Assessment and Evaluation
Q1. A teacher uses a checklist to observe group work and records how often each skill appears. The checklist is best seen as a:
- (A) Final grade
- (B) Value judgement
- (C) Measurement instrument
- (D) Teaching method
Show Answer
Answer: C – An observation checklist is an instrument used to measure behaviour systematically.
Q2. Which question is answered by evaluation rather than by measurement alone?
- (A) How good is this programme compared with its goals?
- (B) How many items did the student answer correctly?
- (C) What percentage did the student obtain?
- (D) How long did the student take to finish?
Show Answer
Answer: A – Evaluation judges quality against goals; the others are measurements.
Q3. A portfolio containing drafts, revisions and reflections over a term is mainly a tool for:
- (A) Single-score ranking
- (B) Assessment of progress over time
- (C) Recording attendance
- (D) Checking textbook coverage only
Show Answer
Answer: B – Portfolios show growth and process, not just one final score.
Q4. After a unit test, a teacher changes the next lesson plan because many students missed the same concept. This use of results is:
- (A) Certification
- (B) Selection for a prize
- (C) Placement in a new school
- (D) Formative use of assessment
Show Answer
Answer: D – Using results to improve upcoming teaching is a formative use.
Q5. A district compares schools by average test scores to decide funding priorities. This is best described as:
- (A) Classroom diagnosis
- (B) Peer assessment
- (C) Self-assessment
- (D) Evaluation for decision making
Show Answer
Answer: D – The scores inform a policy decision, which is an evaluative use.
Q6. Which activity involves measurement but not yet evaluation?
- (A) Declaring a student successful after meeting a standard
- (B) Counting correct responses and reporting a raw score of 34
- (C) Recommending a student for promotion
- (D) Judging a course as effective
Show Answer
Answer: B – Reporting a raw score describes performance without judging its worth.
Q7. A pre-test given to place students into remedial or advanced groups serves mainly a:
- (A) Placement function
- (B) Summative function
- (C) Certification function
- (D) Ranking function only
Show Answer
Answer: A – Placement assessment assigns learners to suitable groups or levels.
Q8. When a teacher asks students to judge their own drafts against a rubric before submission, this is:
- (A) External examination
- (B) Norm-referenced grading
- (C) Self-assessment
- (D) Standardised testing
Show Answer
Answer: C – Learners assessing their own work against criteria is self-assessment.
Q9. A teacher gives marks for classwork, homework and a project, then combines them into a term grade. This combined judgement is closest to:
- (A) Evaluation based on multiple measures
- (B) A single measurement
- (C) A diagnostic label
- (D) An item analysis
Show Answer
Answer: A – Combining several measures to reach a term judgement is evaluative.
Q10. Which statement about assessment is most accurate?
- (A) Assessment is only a final examination
- (B) Assessment always produces a single number
- (C) Assessment must use a standardised test
- (D) Assessment can be formal or informal and ongoing
Show Answer
Answer: D – Assessment includes informal observations as well as formal tests, throughout learning.
Q11. A student is awarded a certificate because she met all course outcomes. The certificate decision depends on:
- (A) Her rank in class
- (B) Her improvement over peers
- (C) Criterion-based evaluation
- (D) Her attendance rank only
Show Answer
Answer: C – Meeting set outcomes is a criterion-based decision.
Q12. A teacher notices, through questioning, that students confuse two science terms and re-teaches them at once. This is an example of:
- (A) Summative testing
- (B) Informal formative assessment
- (C) Final certification
- (D) External moderation
Show Answer
Answer: B – On-the-spot questioning that changes teaching is informal formative assessment.
Scales of Measurement
Q13. Assigning roll numbers to students for identification uses which scale?
- (A) Interval scale
- (B) Nominal scale
- (C) Ratio scale
- (D) Ordinal scale
Show Answer
Answer: B – Roll numbers are labels; a higher number does not mean more of a trait.
Q14. Which example is on a ratio scale?
- (A) Letter grades A, B and C
- (B) School names
- (C) Years of the calendar
- (D) Number of books read in a month
Show Answer
Answer: D – A count of books has a true zero and meaningful ratios.
Q15. Calendar years such as 2020 and 2024 are best classified as:
- (A) Nominal
- (B) Ordinal
- (C) Interval
- (D) Ratio
Show Answer
Answer: C – Calendar years have equal gaps but no true zero year.
Q16. Which measure is suitable for ordinal data?
- (A) Median
- (B) Mean of raw counts only
- (C) Standard deviation of ranks
- (D) Product-moment correlation of categories
Show Answer
Answer: A – The median summarises the middle rank without assuming equal gaps.
Q17. A teacher ranks debate teams from best to worst. The main limitation of this ranking is that it does not show:
- (A) How large the gaps between teams are
- (B) The order of teams
- (C) Which team is first
- (D) The names of teams
Show Answer
Answer: A – Ordinal ranks show order only; the size of gaps remains unknown.
Q18. Which operation is NOT meaningful for nominal data?
- (A) Counting cases in each category
- (B) Finding the most frequent category
- (C) Calculating a mean category value
- (D) Reporting percentages per category
Show Answer
Answer: C – Category labels have no numerical value, so a mean is meaningless.
Q19. If a runner takes 20 seconds and another takes 40 seconds, the first runner took half the time. This ratio statement works because time is measured on a:
- (A) Nominal scale
- (B) Ordinal scale
- (C) Interval scale
- (D) Ratio scale
Show Answer
Answer: D – Time has a true zero, so ratios are meaningful.
Q20. A test score of zero on a classroom quiz does NOT prove zero knowledge, because classroom scores are usually treated as:
- (A) Nominal measures
- (B) Interval measures
- (C) Ratio measures
- (D) Pure counts
Show Answer
Answer: B – A zero test score is an arbitrary point, not proof of no ability.
Q21. Which data set allows all four operations of ordering, equal intervals and ratios?
- (A) Favourite colours
- (B) Letter grades
- (C) Celsius temperatures
- (D) Distances in metres
Show Answer
Answer: D – Distances have order, equal units and a true zero.
Q22. The most appropriate summary for the favourite games chosen by a class is the:
- (A) Mode
- (B) Mean
- (C) Standard deviation
- (D) Range
Show Answer
Answer: A – Favourite games are nominal categories, summarised by the mode.
Q23. Converting raw ranks into percentile ranks changes ordinal data into a form that is often treated as:
- (A) Nominal
- (B) Purely categorical
- (C) Approximately interval for practical use
- (D) Ratio with a true zero
Show Answer
Answer: C – Percentiles are often treated as near-interval in practice, though strictly they remain rank-based.
Q24. Which scale is the strongest, carrying the most measurement information?
- (A) Nominal
- (B) Ratio
- (C) Ordinal
- (D) Interval
Show Answer
Answer: B – Ratio scales include order, equal intervals and a true zero.
Validity
Q25. A test developer writes a table of specifications linking each item to a learning objective. This step mainly supports:
- (A) Content validity
- (B) Face validity only
- (C) Inter-rater reliability
- (D) Test-retest reliability
Show Answer
Answer: A – A table of specifications ensures systematic coverage of objectives.
Q26. A driving test that only checks road signs but never tests actual driving lacks:
- (A) Scoring consistency
- (B) Test length
- (C) Item difficulty
- (D) Content validity
Show Answer
Answer: D – Missing a major part of the domain weakens content validity.
Q27. Two tests of the same skill, given in the same week, correlate at 0.78. This result best supports:
- (A) Scoring speed
- (B) Concurrent validity
- (C) Face validity
- (D) Item difficulty
Show Answer
Answer: B – Agreement with a current criterion supports concurrent validity.
Q28. A university predicts first-year grades from entry test scores. If the prediction is accurate, the entry test has good:
- (A) Face validity
- (B) Content validity
- (C) Predictive validity
- (D) Split-half reliability
Show Answer
Answer: C – Forecasting a later outcome is predictive validity.
Q29. A test of creativity correlates strongly with a test of artistic skill but weakly with a test of memory. This pattern supports:
- (A) Scoring reliability
- (B) Test length
- (C) Construct validity
- (D) Face validity only
Show Answer
Answer: C – Expected high and low relationships together support the construct.
Q30. Which evidence is weakest for validity?
- (A) A test that merely looks suitable to non-experts
- (B) Correlation with a relevant criterion
- (C) Expert review of content
- (D) Theory-based studies of the trait
Show Answer
Answer: A – Appearance alone is weak evidence; expert review, criterion links and theory studies are stronger.
Q31. If a maths test is written in language far above the students reading level, scores will mainly reflect reading ability. This is a threat called:
- (A) Low reliability
- (B) Poor item difficulty
- (C) Short test length
- (D) Construct-irrelevant variance
Show Answer
Answer: D – Reading demand unrelated to maths adds irrelevant variance.
Q32. A test claims to measure problem solving but only asks recall of facts. The main problem is:
- (A) Low face validity
- (B) Construct under-representation
- (C) High reliability
- (D) Long testing time
Show Answer
Answer: B – The test misses key parts of the problem-solving construct.
Q33. Which pair belongs under criterion-related validity?
- (A) Face and content validity
- (B) Concurrent and predictive validity
- (C) Split-half and test-retest reliability
- (D) Internal consistency and stability
Show Answer
Answer: B – Concurrent and predictive validity both relate scores to a criterion.
Q34. A test with high reliability but poor validity is best described as:
- (A) Useful for all purposes
- (B) Free of error
- (C) Easy to score
- (D) Consistently measuring the wrong thing
Show Answer
Answer: D – Reliability alone does not guarantee that the right trait is measured.
Q35. To judge content validity, the most suitable judges are:
- (A) Subject-matter experts
- (B) Students who took the test
- (C) Parents of students
- (D) Test printers
Show Answer
Answer: A – Experts can judge whether items represent the subject domain.
Q36. Restriction of range in the criterion group will usually:
- (A) Increase item difficulty
- (B) Increase test length
- (C) Lower the observed validity coefficient
- (D) Improve face validity
Show Answer
Answer: C – A narrow range of scores reduces the correlation with a criterion.
Reliability
Q37. A bathroom scale that shows a different weight each time a person steps on it lacks:
- (A) Content coverage
- (B) Face appeal
- (C) Predictive power
- (D) Reliability
Show Answer
Answer: D – Inconsistent repeated measures show low reliability.
Q38. Which reliability method needs only one test administration?
- (A) Test-retest method
- (B) Parallel-forms method
- (C) Split-half method
- (D) Stability method over time
Show Answer
Answer: C – Split-half uses a single administration divided into halves.
Q39. The Spearman-Brown formula is used in split-half reliability to:
- (A) Correct for the shorter length of each half
- (B) Find item difficulty
- (C) Judge content coverage
- (D) Score essay answers
Show Answer
Answer: A – It estimates reliability for the full-length test from half-test correlation.
Q40. A test-retest interval that is too short may inflate reliability because students:
- (A) Forget the items completely
- (B) Remember items and answers from the first sitting
- (C) Change in ability greatly
- (D) Receive new teaching between sittings
Show Answer
Answer: B – Memory effects can make scores look more stable than they are.
Q41. Cronbach alpha is most similar to which approach?
- (A) Test-retest over months
- (B) Internal consistency across items
- (C) Agreement between judges
- (D) Prediction of future grades
Show Answer
Answer: B – Alpha estimates how consistently items measure the same trait.
Q42. If two markers give very different marks for the same scripts, the marking suffers from low:
- (A) Content validity
- (B) Predictive validity
- (C) Item difficulty
- (D) Inter-rater reliability
Show Answer
Answer: D – Disagreement between raters is an inter-rater reliability problem.
Q43. A clear marking scheme for essays is most likely to improve:
- (A) Test length
- (B) Syllabus coverage
- (C) Inter-rater reliability
- (D) Item difficulty
Show Answer
Answer: C – Clear criteria help different markers agree.
Q44. A test with SD 8 and reliability 0.75 has a standard error of measurement of:
- (A) 4
- (B) 6
- (C) 2
- (D) 8
Show Answer
Answer: A – SEM = 8 x square root of (1 – 0.75) = 8 x 0.5 = 4.
Q45. A smaller standard error of measurement means scores are:
- (A) More difficult
- (B) More valid automatically
- (C) More precise
- (D) Less consistent
Show Answer
Answer: C – A small SEM means less uncertainty around the true score.
Q46. Which condition tends to increase reliability?
- (A) Vague directions
- (B) Homogeneous items measuring one trait
- (C) Very few items
- (D) Noisy testing room
Show Answer
Answer: B – Homogeneous, well-written items increase internal consistency.
Q47. A reliability coefficient is best interpreted as:
- (A) The percentage of students who pass
- (B) The average item difficulty
- (C) The number of items in the test
- (D) The proportion of score variance due to true differences
Show Answer
Answer: D – Reliability reflects the share of variance from true score differences rather than error.
Q48. A test given to a group with a very wide range of ability will usually show:
- (A) Higher reliability, because true score variance is larger
- (B) Lower reliability, because scores vary too much
- (C) No change in reliability
- (D) Lower validity in all cases
Show Answer
Answer: A – Greater true variance raises the reliability coefficient.
Item Analysis
Q49. In item analysis, the upper and lower groups are usually formed from the top and bottom:
- (A) 10 percent each
- (B) 50 percent each
- (C) 27 percent each
- (D) 5 percent each
Show Answer
Answer: C – The upper and lower 27 percent are commonly used for item analysis.
Q50. An item answered correctly by 135 of 150 students has difficulty index:
- (A) 0.15
- (B) 0.90
- (C) 0.45
- (D) 0.10
Show Answer
Answer: B – p = 135 divided by 150 = 0.90.
Q51. An item with p = 0.20 is best described as:
- (A) Very easy
- (B) Moderate
- (C) Too easy to discriminate
- (D) Difficult
Show Answer
Answer: D – Only 20 percent answered correctly, so the item is difficult.
Q52. A discrimination index of 0.40 or above is generally regarded as:
- (A) Very good discrimination
- (B) Poor discrimination
- (C) Negative discrimination
- (D) No discrimination
Show Answer
Answer: A – Values at or above 0.40 indicate strong discrimination.
Q53. Upper group 20 of 25 correct, lower group 10 of 25 correct. The discrimination index is:
- (A) 0.40
- (B) 0.20
- (C) 0.80
- (D) 0.60
Show Answer
Answer: A – D = (20 – 10) divided by 25 = 0.40.
Q54. An item with discrimination near zero should be:
- (A) Retained as a model item
- (B) Revised or discarded, since it does not separate strong and weak students
- (C) Made compulsory for all
- (D) Used only in easy tests
Show Answer
Answer: B – Near-zero discrimination means the item adds no useful separation.
Q55. Which distractor pattern shows a working distractor?
- (A) Chosen by no one
- (B) Chosen mainly by high scorers
- (C) Chosen mainly by low scorers, with few high scorers
- (D) Chosen equally by all groups
Show Answer
Answer: C – A plausible wrong option should attract weaker students more than strong ones.
Q56. If the key is chosen by fewer low scorers than a distractor, the item most likely has:
- (A) Ideal difficulty
- (B) Perfect discrimination
- (C) A very easy key
- (D) A faulty key or misleading distractor
Show Answer
Answer: D – When a distractor beats the key among low scorers, check the key and wording.
Q57. An item analysis shows p = 0.55 and D = 0.35. The best action is to:
- (A) Delete the item at once
- (B) Make the item much harder
- (C) Make the item much easier
- (D) Retain the item, with minor review if needed
Show Answer
Answer: D – Moderate difficulty and good discrimination make this a sound item.
Q58. The point-biserial correlation for an item is used to show:
- (A) The time taken per item
- (B) How many students chose each option
- (C) How well the item relates to total test performance
- (D) The number of items in the test
Show Answer
Answer: C – A positive point-biserial means good overall performers tend to answer the item correctly.
Q59. A test developer wants maximum discrimination for a norm-referenced test. The best average difficulty to aim for is about:
- (A) 0.10
- (B) 0.50
- (C) 0.90
- (D) 0.95
Show Answer
Answer: B – Items near 0.50 spread students most and discriminate best.
Q60. After item analysis, an item with p = 0.88, D = 0.08 and one unused distractor is best handled by:
- (A) Revising the item and its distractors, or removing it
- (B) Keeping it unchanged as a core item
- (C) Using it as the first item in every test
- (D) Doubling its marks
Show Answer
Answer: A – An easy, weakly discriminating item with a dead distractor needs revision or removal.
📥 Prefer pen and paper? Take this test offline.
Download Printable PDF – Test 2
Free printable version: all 60 questions + answer key. Explanations stay here on the website.