ATP ATP Psychometrics & Score Interpretation 1 — Questions and Answers
Question 1: What does 'reliability' mean in the context of educational and psychological testing?
- Whether the test measures the right content
- The consistency of test scores across repeated administrations or parallel forms (Correct answer)
- The speed of test delivery
- The number of items in a test
Correct answer: The consistency of test scores across repeated administrations or parallel forms
Reliability refers to the consistency of test scores, meaning the test produces similar results under similar conditions.
Question 2: What is 'Cronbach's alpha' used to measure?
- The average score of all examinees
- Internal consistency reliability of a test (Correct answer)
- The correlation between two different tests
- The standard error of measurement
Correct answer: Internal consistency reliability of a test
Cronbach's alpha measures the internal consistency reliability of a test, indicating how well items measure the same construct.
Question 3: What is the 'standard error of measurement' (SEM)?
- The average number of mistakes made by test administrators
- An estimate of the variability in a test score due to measurement error (Correct answer)
- The standard deviation of all test scores
- The error rate in test printing
Correct answer: An estimate of the variability in a test score due to measurement error
The SEM estimates the amount of error in an individual test score, indicating the range within which a true score likely falls.
Question 4: What is a 'norm-referenced' test?
- A test that measures mastery of specific content standards
- A test where scores are interpreted relative to the performance of a comparison group (Correct answer)
- A test that all examinees must pass
- A test administered under normal conditions only
Correct answer: A test where scores are interpreted relative to the performance of a comparison group
Norm-referenced tests compare an examinee's performance to that of a normative sample rather than to an absolute standard.
Question 5: What does a 'criterion-referenced' test measure?
- Performance relative to peers
- Whether an examinee has mastered a defined set of skills or knowledge (Correct answer)
- The average performance across demographic groups
- The financial cost to develop the test
Correct answer: Whether an examinee has mastered a defined set of skills or knowledge
Criterion-referenced tests evaluate whether an examinee has achieved a predetermined level of performance on a defined content domain.
Question 6: What is a 'percentile rank'?
- The raw score divided by the number of items
- The percentage of scores in a distribution that fall at or below a given score (Correct answer)
- The ranking of test items by difficulty
- The percentage of items answered correctly
Correct answer: The percentage of scores in a distribution that fall at or below a given score
A percentile rank indicates the percentage of scores in a reference group that are at or below a given score.
ATP ATP Psychometrics & Score Interpretation 1
What does 'reliability' mean in the context of educational and psychological testing?