Deck 6: Reliability: the Consistency of Test Scores

Question 1

The reliability coefficient is the ratio of

A) true score variance/total variance
B) true score variance/error variance
C) total variance/true score variance
D) total variance/error plus the score variance

Answer 1

A

Question 2

The central assumption of reliability theory is that

A) measurement errors are essentially random
B) measurement errors are not random
C) measurement errors are due to deficiencies in the test
D) measurement errors cannot be minimized

Answer 2

A

Question 3

If a student generally does poorly on math exams, but does extremely well on a math test because the student studied only those things on the exam, we could attribute the source of variability in test scores to

A) temporary and specific characteristics of the individual
B) lasting but specific characteristics of the individual
C) variance not otherwise accounted for (luck)
D) any of the above

Answer 3

D

Question 4

Administering a test to a group of individuals, then computing the average intercorrelation among all items on the test demonstrates the _________ method of estimating reliability.

A) test-retest
B) alternate forms
C) split-half
D) internal consistency

Question 5

True scores and errors are assumed to be

A) highly correlated
B) not highly correlated
C) uncorrelated
D) negatively correlated

Question 6

Administering a test to a group of individuals, readministering that same test to the same group at a later time, and correlating test scores at times 1 and 2 demonstrates which method of estimating reliability?

A) test-retest method
B) alternate forms method
C) split-half method
D) internal consistency method

Question 7

Given the equation X = T + e, X represents ________, T represents ________, and e represents _________.

A) true score; score on the test; error
B) true score; error; score on the test
C) score on the test; error; true score
D) score on the test; true score; error

Question 8

Given that mechanical aptitude is a stable characteristic, scores of a given person on a mechanical aptitude test taken at different periods of time should

A) be similar
B) be different
C) be exactly the same
D) change due to experience

Question 9

"True score" on a measure of anxiety represents

A) a person's real level of anxiety
B) a person's obtained level of anxiety
C) a combination of all factors leading to consistency in anxiety scores
D) both a and c

Question 10

Developing tests that are equivalent in terms of content, response processes, and statistical characteristics is crucial for the ________ method of estimating reliability.

A) test-retest
B) alternate forms
C) split-half
D) internal consistency

Question 11

Administering a test to a group of individuals, dividing the test into two sections, and correlating the scores on each of the sections demonstrates the ________ method of estimating reliability.

A) test-retest
B) alternate forms
C) split-half
D) internal consistency

Question 12

When you are interested in the long-term stability of a measure, you should use the ______ method for estimating reliability.

A) test-retest
B) alternate forms
C) split-half
D) internal consistency

Question 13

Split-half methods compare ______, while internal consistency estimates compare ______.

A) test items; test halves
B) test halves; test forms
C) test halves; test items
D) test forms; test items

Question 14

Thorndike (1949) would place "motivation" under which one of the following possible sources of variability in test scores?

A) lasting and general characteristics of the individual
B) lasting but specific characteristics of the individual
C) temporary but general characteristics of the individual
D) temporary and specific characteristics of the individual

Question 15

Errors on different measures are assumed to be

A) highly correlated
B) not highly correlated
C) uncorrelated
D) negatively correlated

Question 16

Consistency in test scores is important because

A) consistency means a test is measuring what it is designed to measure
B) inconsistent tests must be given many times
C) inconsistent scores don't provide a good measure of stable traits or attributes
D) ability should not change over time

Question 17

Almost all theories of reliability assume that

A) people do not differ in levels of a given characteristic
B) test scores change over time
C) test scores do not change over time
D) test scores are influenced by consistency and inconsistency

Question 18

Methods of studying, defining, and estimating the consistency or inconsistency of test scores deal with the ______ of test scores.

A) validity
B) appropriateness
C) reliability
D) stability

Question 19

The _________ method of estimating reliability involves only one test administration.

A) test-retest
B) alternate forms
C) split-half
D) all involve multiple test administrations

Question 20

Alternate forms demonstrate ________ carry over effects than the test-retest method.

A) more
B) less
C) about the same
D) sometimes less and sometimes more

Question 21

The _______ methods of estimating reliability define measurement error strictly in terms of consistency or inconsistency of the content of a test.

A) test-retest and internal consistency
B) test-retest, alternate forms, and internal consistency
C) split-half, test-retest, and internal consistency
D) internal consistency and split-half

Question 22

Classical reliability theory is simply a special case of

A) true reliability theory
B) analysis of variance methodology
C) generalization theory
D) generalizability theory