Deck 7: Using and Interpreting Information About Test Reliability

Question 1

The standard error of measurement is a function of which two factors?

A) reliability of the test and range of test scores
B) variability of test scores and range of test scores
C) reliability of the test and variability of test scores
D) variability of test scores and sample size

Answer 1

C

Question 2

Changes over time in the attribute being measured are considered sources of measurement error in

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

Answer 2

D

Question 3

Reliability of a test is affected by

A) the characteristics of the people taking the test
B) the method used to estimate reliability
C) the intended use of the test scores
D) all of the above

Answer 3

D

Question 4

_________ scores are typically more reliable than ________ scores.

A) difference; composite
B) composite; difference
C) individual test; composite
D) true; observed

Question 5

If individual differences are _______, reliability will be _______.

A) small; low
B) small; high
C) large; low
D) large; zero

Question 6

True Score Theory of Reliability was developed by

A) Thorndike
B) Ghiselli
C) Brown
D) Spearman

Question 7

Reliability of a test can be increased by

A) decreasing inter-item correlations
B) decreasing number of test items
C) increasing number of test items
D) both a and c

Question 8

A test in which items are trivially easy, but which is administered under strict time limits, is called a(n)

A) power test
B) speed test
C) unreliable test
D) timed test

Question 9

The ________ the correlation between X and Y, the ________ the reliability of their difference scores.

A) higher; lower
B) lower; lower
C) higher; higher
D) none of the above

Question 10

A pure speed test should have an odd-even split-half reliability of about

A) 0.00
B) 0.25
C) 1.00
D) -0.50

Question 11

Variability expected on the basis of measurement error is best represented by

A) the standard error of measurement
B) the reliability coefficient
C) the standard deviation
D) the variance of the test

Question 12

If a test has a reliability coefficient of 0.00 and a standard error of 0.00, the test is

A) reliable and accurate
B) unreliable but accurate
C) unreliable and inaccurate
D) reliable but inaccurate

Question 13

In general, _______ could be expected to be higher than _______, because more factors contribute to measurement error in the latter.

A) alternate form correlations; internal consistency estimates
B) test-retest reliability estimates; alternate forms correlations
C) test-retest reliability estimates; internal consistency estimates
D) internal consistency estimates; alternate form correlations

Question 14

If a person's true score is 110 on a test with a standard error of measurement of 3.7 and a mean of 100, we would expect 95% of the person's test scores to fall within

A) 102.75 - 117.25
B) 92.75 - 107.25
C) 90.75 - 120.25
D) 100-110

Question 15

If the standard error of a test increases, the range in which you could be confident that true scores will fall would

A) remain the same
B) decrease
C) increase
D) standard error does not affect the confidence interval

Question 16

A method of estimating the true correlation between X and Y given the correlation between two unreliable measures of X and Y is by using the

A) shrinkage coefficient
B) reliability coefficient
C) correction for attenuation
D) corrected reliability index

Question 17

Which reliability coefficient reflects the more reliable test?

A) 0.30
B) -.85
C) 2.58
D) 0.90

Question 18

The most useful method of assessing the reliability of highly speeded tests is the

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

Question 19

Because of measurement error, the variability of test scores is _________ the variability of true scores.

A) the same as
B) smaller than
C) larger than
D) 2/3 times the variability of true scores

Question 20

A test can be _________, yet not __________.

A) reliable; valid
B) valid; reliable
C) reliable; consistent
D) accurate; consistent

Question 21

In criterion-referenced testing, our principle concern is with the reliability of

A) test scores
B) individual test items
C) decisions
D) reliability is not a concern in criterion-referenced testing

Question 22

The Bandwidth-Fidelity Dilemma refers to the fact that

A) the more accurate the information, the less information conveyed
B) the less accurate the information, the less information conveyed
C) the more accurate the information, the more information conveyed
D) the more accurate the information, the faster it is conveyed