Q1. A norm-referenced score is best interpreted as:

• (A) The number of objectives the pupil has mastered
• (B) The percentage of the syllabus the pupil knows
• (C) The pupil’s exact marks out of the total
• (D) The pupil’s standing relative to the norm group

Q2. Which situation calls for criterion-referenced testing rather than norm testing?

• (A) Choosing the best student in the district for a prize
• (B) Ranking schools for a newspaper list
• (C) Selecting one captain from a large team
• (D) Deciding whether a student is ready to move on after mastering a unit

Q3. A key advantage of criterion-referenced testing is that it

• (A) Gives clear information about what a student can actually do
• (B) Always produces a wide spread of scores
• (C) Forces a fixed share of students to fail
• (D) Hides gaps in learning behind a rank

Q4. A key advantage of norm-referenced testing is that it

• (A) Proves mastery of every taught objective
• (B) Removes the need for any norm group
• (C) Allows comparison when a common standard is hard to fix
• (D) Works without any scoring key

Q5. A norm-referenced test can mislead a parent if the parent reads a percentile as

• (A) A rank among peers
• (B) A percentage of items answered correctly
• (C) A position in the norm group
• (D) A comparison score

Q6. A criterion-referenced test can mislead if the standard is

• (A) Set arbitrarily without a sound basis
• (B) Shared with students in advance
• (C) Linked to clear objectives
• (D) Applied equally to all students

Q7. Can the same test be interpreted in both ways?

• (A) No, a test can never support two interpretations
• (B) Only if the test has no scoring key
• (C) Yes, if both a norm group and a criterion are available for its scores
• (D) Only if the test is oral

Q8. In a class where every student masters the objectives, a norm-referenced test will still

• (A) Rank students against each other
• (B) Report that all have failed
• (C) Become invalid for all use
• (D) Show identical scores for everyone

Q9. A teacher who wants both ranking and mastery information should use

• (A) Only a class rank list
• (B) Only a total mark
• (C) Norm scores plus a criterion report for objectives
• (D) Only attendance records

Q10. Which grading approach best supports a mastery-learning classroom?

• (A) Forced ranking with a fixed number of top grades
• (B) Grading by attendance only
• (C) A single high-stakes test with no feedback
• (D) Criterion-based grading with reassessment opportunities

Q11. A disadvantage of relative grading is that it

• (A) Makes standards clear to every student
• (B) Always raises overall achievement
• (C) Can discourage students from helping peers
• (D) Removes competition completely

Q12. A disadvantage of absolute grading is that it

• (A) Always ranks students unfairly
• (B) May be hard to set fair cut-offs for a new or complex course
• (C) Requires a very large norm group
• (D) Depends on the performance of other classes

Q13. A student argues that a B in a very hard course should count more than a B in an easy course. This argument supports

• (A) Pass-fail grading only
• (B) Weighted grading
• (C) Ranking by age
• (D) Grading by attendance

Q14. Pass-fail grading is most suitable when the main goal is

• (A) Ranking students for scholarships
• (B) Comparing students across schools precisely
• (C) Judging whether a minimum standard is met, not fine ranking
• (D) Awarding honours distinctions

Q15. A teacher combines test scores, projects and classwork into a final grade. This practice improves grading because it

• (A) Makes grading faster to explain
• (B) Samples learning more broadly than one test alone
• (C) Guarantees every student an A
• (D) Removes the need for a syllabus

Q16. If two teachers grade the same essay very differently, the grading lacks

• (A) A grade scale
• (B) A course title
• (C) Reliability
• (D) A pass mark

Q17. A grade that reflects behaviour and neatness more than learning lacks

• (A) A letter symbol
• (B) A numerical value
• (C) Validity as a measure of achievement
• (D) A place on the transcript

Q18. A school changes from letter grades to detailed written reports. A likely benefit is

• (A) Faster ranking of all students
• (B) Richer information about strengths and next steps
• (C) A simpler GPA calculation
• (D) Less work for teachers

Q19. Standardized tests help policymakers mainly by

• (A) Providing comparable data across many schools
• (B) Replacing teacher judgement in every lesson
• (C) Measuring school building quality
• (D) Counting student enrolment

Q20. A standardized test with poor validity for a local curriculum should be used

• (A) Cautiously, with other evidence, or not at all for that purpose
• (B) As the only measure for promotion
• (C) Without any caution
• (D) To replace the local syllabus

Q21. Test bias is a serious concern because a biased item may

• (A) Raise scores for every student equally
• (B) Make the test shorter
• (C) Lower scores for a group for reasons unrelated to the skill tested
• (D) Improve the test for all groups

Q22. Standardization does NOT guarantee that a test is

• (A) Scored in the same way
• (B) Valid for every purpose a user might choose
• (C) Given under similar conditions
• (D) Interpreted with norms

Q23. A school that prepares students only by drilling past standardized items risks

• (A) Improving all learning automatically
• (B) Teaching to the test instead of the wider curriculum
• (C) Making tests more valid
• (D) Removing test anxiety fully

Q24. Norms become outdated when

• (A) More students take the test
• (B) The test is printed in colour
• (C) The population or curriculum changes over time
• (D) Teachers read the manual

Q25. A standardized achievement test differs from a teacher-made test mainly in that it

• (A) Is always shorter
• (B) Is never used for grading
• (C) Contains no multiple-choice items
• (D) Is developed, tried out and normed on a large sample

Q26. For a high-stakes decision, a single standardized score should be

• (A) Accepted without question
• (B) Ignored completely
• (C) Checked against other evidence before deciding
• (D) Used only if it is low

Q27. A teacher gives a quiz, finds a common error, and reteaches the topic the same week. The quiz functioned as

• (A) Formative assessment
• (B) A final certification test
• (C) An admission test
• (D) A school census

Q28. A final project graded at the end of term and reported to parents functioned as

• (A) Daily formative feedback
• (B) Summative assessment
• (C) A diagnostic interview
• (D) A placement screening

Q29. A counsellor tests a struggling reader’s phonics, vocabulary and fluency separately. This is mainly

• (A) Summative grading
• (B) Diagnostic assessment
• (C) Norm-referenced ranking
• (D) A final board result

Q30. Which statement best separates formative from summative evaluation?

• (A) Formative is always oral; summative is always written
• (B) Formative uses grades; summative never uses grades
• (C) Formative happens once a year; summative happens daily
• (D) Formative guides ongoing learning; summative judges completed learning

Q31. A diagnostic test that only gives a total score without sub-skill detail is weak because it

• (A) Does not show where the difficulty lies
• (B) Is too easy for all students
• (C) Cannot be scored quickly
• (D) Uses too many items

Q32. Peer feedback during group work is formative when students

• (A) Keep it secret from the teacher
• (B) Use it to improve their work before submission
• (C) Use it to rank group members
• (D) Ignore it completely

Q33. A school uses baseline testing at entry to group students for support. This use is closest to

• (A) Placement or diagnostic use before teaching
• (B) Summative certification
• (C) Final grading
• (D) A public ranking

Q34. Summative assessment can still help future learning if

• (A) Scores are filed and never discussed
• (B) Only the top students see their results
• (C) It is given without any instructions
• (D) Results are reviewed to improve the next course or unit

Q35. Which practice mixes the types poorly?

• (A) Using quizzes to guide reteaching
• (B) Using a diagnostic check before remediation
• (C) Using a final test to certify a course
• (D) Using a single end-of-year score alone to plan daily remedial help

Q36. A test constructor who skips the blueprint risks building a test that

• (A) Is automatically more reliable
• (B) Has perfect content coverage
• (C) Over-tests some topics and ignores others
• (D) Needs no tryout

Q37. Writing items before defining outcomes risks building a test that

• (A) Is automatically valid
• (B) Requires no scoring key
• (C) Measures trivia instead of intended learning
• (D) Cannot be printed

Q38. A multiple-choice stem that asks two questions at once is faulty because

• (A) Students may know one part and not the other
• (B) It is always too short
• (C) It cannot have four options
• (D) It is always too easy

Q39. An item whose correct option is much longer than the distractors is faulty because

• (A) Long answers are always wrong
• (B) Test-wise students may pick it by length alone
• (C) Short options are always correct
• (D) Length has no effect on choices

Q40. A tryout sample should resemble

• (A) Only the highest achievers
• (B) Only students from one school
• (C) The students who will take the final test
• (D) Adults who no longer study the subject

Q41. Item analysis after tryout helps a constructor to

• (A) Increase the pass mark secretly
• (B) Change the syllabus after testing
• (C) Award grades without scoring
• (D) Keep, revise or drop items based on evidence

Q42. A test that is too long for the time allowed will mainly harm

• (A) The difficulty of individual items alone
• (B) The total number of pages alone
• (C) Validity, because unfinished items do not measure learning
• (D) The speed of scoring alone

Q43. Assembling the final test requires attention to

• (A) Item order, layout, instructions and time allowed
• (B) Only the total number of pages
• (C) Only the name of the school
• (D) Only the date of printing

Q44. A field test that shows a key error in one item should lead to

• (A) Keeping the error to test attention
• (B) Removing the whole topic from teaching
• (C) Giving all students full marks for the paper
• (D) Correcting the key before final use

Q45. A teacher finds the mean is much higher than the median. This suggests the data are

• (A) Perfectly symmetrical
• (B) Pulled upward by some high extreme scores
• (C) All identical
• (D) Impossible to summarise

Q46. A class has scores 60, 62, 64, 66, 68. The standard deviation is

• (A) Very large, because scores increase
• (B) Zero, because scores differ
• (C) Impossible to estimate
• (D) Small, because scores cluster tightly

Q47. If a score has a z-score of +1.5, it lies

• (A) One and a half standard deviations above the mean
• (B) One and a half marks above the mean
• (C) At the 150th percentile
• (D) Fifteen points above the highest score

Q48. A percentile rank of 25 and a percent score of 25 are

• (A) Different: rank is position, percent is proportion correct
• (B) The same in all cases
• (C) Both equal to 25 marks
• (D) Both impossible on one test

Q49. In a normal distribution, the mean, median and mode are

• (A) Far apart at the extremes
• (B) Approximately equal and at the centre
• (C) Equal to the range
• (D) Always zero

Q50. A skewed distribution with a long tail to the right has a mean that is

• (A) Less than the median
• (B) Equal to the mode
• (C) Unrelated to the median
• (D) Greater than the median

Q51. The interquartile range measures the spread of

• (A) The middle 50 percent of scores
• (B) All scores from lowest to highest
• (C) Only the top quarter of scores
• (D) Only scores above the mean

Q52. A teacher wants a single value for typical performance in a class with one extreme outlier. The best choice is the

• (A) Mean
• (B) Range
• (C) Highest score
• (D) Median

Q53. A frequency distribution is most useful for showing

• (A) Only the name of the top student
• (B) How scores are spread across values or intervals
• (C) Only the pass mark
• (D) The date of the test

Q54. Two classes have the same mean but different standard deviations. The class with the larger deviation has scores that are

• (A) More clustered
• (B) Higher on average
• (C) Lower on average
• (D) More spread out

Q55. Tyler’s model works best when objectives are

• (A) Kept secret from teachers
• (B) Clear, measurable and agreed in advance
• (C) Changed after results are known
• (D) Written only after teaching ends

Q56. A weakness of Tyler’s model for a new programme is that it

• (A) Requires no objectives at all
• (B) Forbids the use of tests
• (C) Gives little guidance on needs assessment before objectives are set
• (D) Works only for mathematics

Q57. CIPP is called a decision-oriented model because its evaluations are meant to

• (A) Support decisions at planning, implementation and outcome stages
• (B) Produce a single final grade
• (C) Rank students for prizes
• (D) Replace all classroom tests

Q58. In CIPP, a decision to continue, change or stop a programme mainly uses

• (A) Only attendance data
• (B) Only textbook prices
• (C) Only staff opinions
• (D) Product evaluation evidence

Q59. Which of the following is an example of input evaluation?

• (A) Judging final examination results after a programme ends
• (B) Checking daily attendance during a programme
• (C) Asking parents how they felt after a programme ended
• (D) Comparing two reading schemes’ costs and materials before adopting one

Q60. Process evaluation helps decision makers to

• (A) Detect implementation problems early and adjust the programme
• (B) Set admission standards for next year
• (C) Judge only final results
• (D) Choose a school building