Q1. A headteacher studies attendance, pass rates and parent feedback, then decides whether to extend a remedial programme. This decision process is best called:

• (A) Scoring
• (B) Item analysis
• (C) Ranking
• (D) Evaluation

Q2. A teacher asks short oral questions during a lesson to check understanding before moving on. This is:

• (A) Final examination
• (B) Formative assessment
• (C) Certification
• (D) External moderation

Q3. Which example is measurement?

• (A) Judging a school as excellent
• (B) Recommending promotion
• (C) Recording 27 correct answers out of 30
• (D) Deciding a course should close

Q4. A student meets the fixed standard of 60 percent for a skills certificate. This is an example of:

• (A) Criterion-referenced judgement
• (B) Peer ranking
• (C) Norm-referenced judgement
• (D) Self-grading only

Q5. A teacher keeps a diary of observations about a struggling reader across six weeks. This record supports:

• (A) One-off summative grading only
• (B) Ongoing assessment of progress
• (C) Standardised ranking
• (D) Item banking

Q6. Which use of a test is summative?

• (A) Adjusting tomorrow lesson plan
• (B) Giving feedback on a draft
• (C) Diagnosing a misconception mid-unit
• (D) Awarding final grades at term end

Q7. A new student is tested to decide which class level suits her best. This test is used for:

• (A) Certification
• (B) Final grading
• (C) Placement
• (D) Programme closure

Q8. Assessment differs from a single test score because assessment may include:

• (A) Observations, projects and discussions as well as tests
• (B) Only multiple-choice items
• (C) Only final examinations
• (D) Only numerical marks

Q9. A teacher compares Faraz score with the scores of all students who took the same national test. This is:

• (A) Criterion referencing
• (B) Norm referencing
• (C) Diagnostic labelling
• (D) Formative feedback

Q10. Which statement best describes diagnostic assessment?

• (A) It certifies course completion
• (B) It ranks whole schools
• (C) It identifies specific strengths, gaps and causes of difficulty
• (D) It replaces all teaching

Q11. A rubric used to judge project quality against described levels is mainly a tool for:

• (A) Criterion-referenced assessment
• (B) Random ranking
• (C) Attendance recording
• (D) Item difficulty analysis

Q12. The main value of continuous assessment is that it:

• (A) Depends on one examination day
• (B) Ignores classroom work
• (C) Removes feedback entirely
• (D) Reduces reliance on a single high-stakes test

Q13. Classifying textbooks by subject – maths, science, English – uses which scale?

• (A) Nominal scale
• (B) Ordinal scale
• (C) Interval scale
• (D) Ratio scale

Q14. A league table listing schools from first to tenth, with no score gaps shown, is:

• (A) Nominal data
• (B) Ratio data
• (C) Ordinal data
• (D) Interval data

Q15. Scores on an IQ-type scale, where equal point gaps are assumed but zero is arbitrary, are treated as:

• (A) Nominal
• (B) Ordinal
• (C) Categorical only
• (D) Interval

Q16. The number of minutes a student spends reading each day is measured on a:

• (A) Nominal scale
• (B) Ratio scale
• (C) Ordinal scale
• (D) Interval scale only

Q17. Which statement about a ratio scale is true?

• (A) Categories cannot be ordered
• (B) Equal intervals are absent
• (C) Zero is always arbitrary
• (D) A value of zero means absence of the quantity

Q18. For nominal data, the only central tendency that is always meaningful is the:

• (A) Mean
• (B) Mode
• (C) Median
• (D) Standard deviation

Q19. A teacher converts letter grades to ranks for a prize list. The resulting data are:

• (A) Ordinal
• (B) Ratio
• (C) Interval
• (D) Nominal

Q20. On an interval scale, the difference between 30 and 40 is:

• (A) Twice the difference between 10 and 15
• (B) Meaningless
• (C) The same as the difference between 70 and 80
• (D) Half the difference between 20 and 40

Q21. Which variable is ordinal?

• (A) Number of pages in a book
• (B) Weight in grams
• (C) Level of agreement from strongly disagree to strongly agree
• (D) Eye colour categories

Q22. A researcher can say one school library has twice as many books as another because book counts are on a:

• (A) Nominal scale
• (B) Ordinal scale
• (C) Interval scale
• (D) Ratio scale

Q23. Which statistic should NOT be used with ordinal ranks alone?

• (A) Mean, because it assumes equal intervals
• (B) Median
• (C) Mode
• (D) Frequency count

Q24. Temperature in Kelvin, unlike Celsius, is a ratio scale because:

• (A) It uses larger numbers
• (B) Its zero is absolute zero, meaning no thermal energy
• (C) It is measured with thermometers
• (D) It has no equal intervals

Q25. A test of English writing that is scored only on handwriting neatness has a validity problem because it measures:

• (A) Content coverage
• (B) An irrelevant trait instead of writing skill
• (C) Test length
• (D) Scoring speed

Q26. Content validity is best improved by:

• (A) Careful sampling of the whole syllabus when writing items
• (B) Making items harder
• (C) Shortening the test sharply
• (D) Using only true-false items

Q27. A new aptitude test is validated against supervisor ratings collected in the same month. This is:

• (A) Face validity
• (B) Content validity
• (C) Test-retest reliability
• (D) Concurrent validity

Q28. A scholarship test is judged by how well it forecasts university success two years later. This is:

• (A) Face validity
• (B) Content validity
• (C) Predictive validity
• (D) Inter-rater reliability

Q29. A science reasoning test correlates with lab performance but not with spelling skill. This pattern best supports:

• (A) Construct validity
• (B) Scoring reliability only
• (C) Face validity only
• (D) Item difficulty

Q30. Which factor can weaken predictive validity?

• (A) Clear test directions
• (B) A reliable test
• (C) A relevant criterion
• (D) Changes in teaching or support between test and criterion

Q31. A test appears professional and relevant to candidates, but no other validity evidence exists. It has only:

• (A) Content validity
• (B) Face validity
• (C) Predictive validity
• (D) Construct validity

Q32. Validity is best thought of as a property of:

• (A) The test booklet alone
• (B) The printing quality
• (C) The interpretation and use of scores
• (D) The number of items alone

Q33. A test that is reliable but invalid for a given purpose will:

• (A) Give random scores
• (B) Be easy to use
• (C) Cover the syllabus fully
• (D) Give consistent but misleading results for that purpose

Q34. Which validity question asks whether the test covers the intended domain fairly?

• (A) Does it predict future success?
• (B) Does it look suitable?
• (C) Does it represent the content domain?
• (D) Do scorers agree?

Q35. To establish construct validity, researchers often use:

• (A) Only one easy item
• (B) Factor analysis and group comparisons based on theory
• (C) Counting correct answers only
• (D) Timing the test only

Q36. A criterion that itself is unreliable will usually make validity appear:

• (A) Lower than the true validity
• (B) Higher than the true validity
• (C) Unaffected
• (D) Perfect

Q37. Reliability is necessary but not sufficient for validity because:

• (A) Consistency removes all error
• (B) A valid test is always short
• (C) A test can be consistent yet measure the wrong trait
• (D) Validity needs no consistency

Q38. Two forms of a test with matched content and difficulty are used to estimate:

• (A) Face validity
• (B) Content validity
• (C) Inter-rater agreement
• (D) Parallel-forms reliability

Q39. A single test is split into two halves by odd and even numbers. This estimates:

• (A) Split-half reliability
• (B) Test-retest reliability
• (C) Predictive validity
• (D) Content validity

Q40. KR-20 is suitable when items are scored:

• (A) By partial credit only
• (B) As correct or incorrect
• (C) By judges on a rubric
• (D) By time taken only

Q41. A test-retest study uses a gap of one day for a memory-heavy test. The main risk is:

• (A) Practice and memory effects inflating stability
• (B) Complete change in the trait
• (C) Loss of all items
• (D) Change in test length

Q42. A teacher wants to know if essay marks would be similar if another teacher marked them. She should check:

• (A) Split-half reliability
• (B) Test-retest reliability
• (C) Inter-rater reliability
• (D) Parallel-forms reliability

Q43. Adding many poor-quality items to a test will most likely:

• (A) Raise validity sharply
• (B) Improve content coverage fully
• (C) Reduce testing time
• (D) Fail to improve reliability and may harm it

Q44. A score of 65 with SEM 3 suggests the true score probably lies around:

• (A) Exactly 65
• (B) 62 to 68
• (C) 50 to 80
• (D) 0 to 100

Q45. Which change lowers the standard error of measurement?

• (A) Lower reliability
• (B) Larger score spread only
• (C) More guessing
• (D) Higher reliability

Q46. Guessing on multiple-choice items mainly adds:

• (A) True score variance
• (B) Random error variance
• (C) Content validity
• (D) Item difficulty control

Q47. A reliability coefficient of 0.95 for a short classroom quiz should be viewed with:

• (A) Full trust at once
• (B) Ignoring the result
• (C) Caution, checking method and sample before trusting it
• (D) Rejection of all quizzes

Q48. Standardising directions, timing and scoring mainly improves:

• (A) Reliability by reducing irrelevant variation
• (B) Item difficulty alone
• (C) Test length
• (D) Face appeal only

Q49. If 60 of 200 students answer an item correctly, its difficulty index is:

• (A) 0.30
• (B) 0.60
• (C) 0.20
• (D) 0.40

Q50. A p value of 0.15 indicates an item that is:

• (A) Very easy
• (B) Very difficult
• (C) Moderate
• (D) Perfectly discriminating

Q51. Upper group 27 of 30 correct and lower group 9 of 30 correct gives discrimination of:

• (A) 0.30
• (B) 0.90
• (C) 0.20
• (D) 0.60

Q52. A discrimination index of 0.10 is best described as:

• (A) Excellent
• (B) Good
• (C) Weak
• (D) Negative

Q53. An item with negative discrimination should first be:

• (A) Retained unchanged
• (B) Given double marks
• (C) Placed at the start
• (D) Checked for a wrong key or ambiguous wording

Q54. Which item is most suitable for a mastery test where most students should pass?

• (A) An item with p = 0.85 and clear wording
• (B) An item with p = 0.10
• (C) An item with negative discrimination
• (D) An item with a dead distractor

Q55. Distractor analysis that shows one option chosen by 40 percent of low scorers and 5 percent of high scorers indicates:

• (A) A dead distractor
• (B) A faulty key
• (C) A functioning, plausible distractor
• (D) An item with no discrimination

Q56. The best action for a distractor chosen by almost no one is to:

• (A) Keep it forever
• (B) Replace it with a more plausible option
• (C) Make it the key
• (D) Remove the item stem

Q57. A test with many very easy items will most likely show:

• (A) Strong discrimination throughout
• (B) Low discrimination, because most students score alike
• (C) Higher validity automatically
• (D) No need for item analysis

Q58. Item analysis is normally carried out:

• (A) Before any student takes the test
• (B) Only for essay tests
• (C) Never, if items look good
• (D) After a trial administration, before final use

Q59. An item with p = 0.48, D = 0.42 and all distractors chosen by some students is:

• (A) Too easy for any use
• (B) A flawed item to delete at once
• (C) A sound item to retain
• (D) Valid only for essays

Q60. If high scorers choose a distractor more often than the key, the test developer should suspect:

• (A) A mis-keyed or ambiguous item
• (B) An item that is too easy
• (C) A test that is too short
• (D) A group that is too large