Question 1

Identify and explain each component of the formula in the basic conception of Classical Test Theory.

Accepted Answer

Classical Test Theory posits that X (observed score) = T (true score) + E (error). An observed score results from the influence of one's true score plus the amount of random error in the administration of that test.

Question 2

Reliability can be conceptually viewed as a ratio of ________.

Accepted Answer

A)  $\frac{\sigma_{\text {observed }}^{2}}{\sigma_{t r u e}^{2}}$ 
B)  $\frac{\sigma_{\text {true }}^{2}}{\sigma_{\text {observed }}^{2}}$ 
C)  $\frac{\sigma_{\text {observed }}^{2} * \sigma_{\text {true }}^{2}}{\sigma_{\text {true }}^{2}}$ 
D)  $\frac{\sigma_{\text {observed }}^{2}+\sigma_{\text {true }}^{2}}{\sigma_{\text {true }}^{2}}$ 
A)  $\frac{\sigma_{\text {observed }}^{2}}{\sigma_{t r u e}^{2}}$ 
B)  $\frac{\sigma_{\text {true }}^{2}}{\sigma_{\text {observed }}^{2}}$ 
C)  $\frac{\sigma_{\text {observed }}^{2} * \sigma_{\text {true }}^{2}}{\sigma_{\text {true }}^{2}}$ 
D)  $\frac{\sigma_{\text {observed }}^{2}+\sigma_{\text {true }}^{2}}{\sigma_{\text {true }}^{2}}$ B

Question 3

________ is the primary concern of reliability.

Accepted Answer

A)  Prediction 
B)  Transparency 
C)  Consistency 
D)  Impartiality 
A)  Prediction 
B)  Transparency 
C)  Consistency 
D)  Impartiality C

Question 4

Explain the statement: Classical Test Theory is test dependent and sample dependent.

Accepted Answer

These are both recognized limitations of

Question 5

In Classical Test Theory, an individual's true score would be ________.

Accepted Answer

A)  independent of the items composing the test 
B)  consistent across various operationalizations of the measure 
C)  dependent on the other test-takers administered the test 
D)  limited to the specific test administered 
A)  independent of the items composing the test 
B)  consistent across various operationalizations of the measure 
C)  dependent on the other test-takers administered the test 
D)  limited to the specific test administered

Question 6

Classical Test Theory considers how many sources of error at a time?

Accepted Answer

A)  1 
B)  2 
C)  4 
D)  An unlimited number 
A)  1 
B)  2 
C)  4 
D)  An unlimited number

Question 7

In Classical Test Theory, a measure's error is assumed to be ________.

Accepted Answer

A)  constant across administrations 
B)  random 
C)  an indicator of bias 
D)  eliminated through standardization of administration 
A)  constant across administrations 
B)  random 
C)  an indicator of bias 
D)  eliminated through standardization of administration

Question 8

In Classical Test Theory, T represents the ________.

Accepted Answer

A)  hypothetical score the test taker would receive if the measure contained zero error 
B)  observed score 
C)  reliability of the measure 
D)  amount of error relative to score obtained by the test taker. 
A)  hypothetical score the test taker would receive if the measure contained zero error 
B)  observed score 
C)  reliability of the measure 
D)  amount of error relative to score obtained by the test taker.

Question 9

Explain, in terms of true score variance and error variance, the impact on reliability of doubling the length of a test.

Accepted Answer

If the number of items on a test is doub

Question 10

A research measure was found to have a reliability of .85. The researcher concluded that 15% of the variability in observed scores reflected ________.

Accepted Answer

A)  true score variance 
B)  bias 
C)  systematic error 
D)  measurement error 
A)  true score variance 
B)  bias 
C)  systematic error 
D)  measurement error

Question 11

Classical Test Theory makes two very important assumptions about the error in testing. Identify these two assumptions.

Accepted Answer

Error is assumed to be random in Classic

Question 12

If scores from a measure with less than perfect reliability are correlated with scores from a second measure, the resulting correlation can be expected to be ________.

Accepted Answer

A)  near zero 
B)  attenuated 
C)  enhanced 
D)  a more accurate estimate of the population correlation 
A)  near zero 
B)  attenuated 
C)  enhanced 
D)  a more accurate estimate of the population correlation

Question 13

A graduate student is concerned that the long length of his dissertation questionnaire will discourage participation. He believes, however, that many of the items on a self-created 40 item measure are repetitive, and he is considering decreasing the scale to 20 items. His original 40-item scale was pre-tested and found to have a reliability of .75. What would you recommend he do before proceeding with his plan to remove items from the measure?

Accepted Answer

Many practical issues can impact researc

Question 14

If a measure randomly assigned scores to each participant, the measure would possess a reliability equal to ________.

Accepted Answer

A)  -1.0 
B)  0 
C)  0.5 
D)  1.0 
A)  -1.0 
B)  0 
C)  0.5 
D)  1.0

Question 15

All else being equal, increasing the length of a test would be expected to ________ the test's reliability.

Accepted Answer

A)  increase 
B)  decrease 
C)  have no effect upon 
A)  increase 
B)  decrease 
C)  have no effect upon

Identify and explain each component of the formula in the basic conception of Classical Test Theory.

Reliability can be conceptually viewed as a ratio of ________.

________ is the primary concern of reliability.

Explain the statement: Classical Test Theory is test dependent and sample dependent.

In Classical Test Theory, an individual's true score would be ________.

Classical Test Theory considers how many sources of error at a time?

In Classical Test Theory, a measure's error is assumed to be ________.

In Classical Test Theory, T represents the ________.

Explain, in terms of true score variance and error variance, the impact on reliability of doubling the length of a test.

A research measure was found to have a reliability of .85. The researcher concluded that 15% of the variability in observed scores reflected ________.

Classical Test Theory makes two very important assumptions about the error in testing. Identify these two assumptions.

If scores from a measure with less than perfect reliability are correlated with scores from a second measure, the resulting correlation can be expected to be ________.

If a measure randomly assigned scores to each participant, the measure would possess a reliability equal to ________.

All else being equal, increasing the length of a test would be expected to ________ the test's reliability.

Introduction and Overview

Statistics Review for PSYC Measurement

Sychological Scaling

Test Preparation and Specification

Estimating Reliability

Content Validation

Criterion-Related Validation

Construct Validation

VG Psychometric Meta-Analysis Copy

Tests Bias Unfairness and Equivalence

Developing Tests of Maximal Performance

Classical Test Theory Item Analysis

Scoring Tests

Developing Measures of Typical Performance

Response Biases

Multiple Regression

Exploratory Factor Analysis

Confirmatory Factor Analysis

Item Response Theory

Applications of Item Response Theory: Computer Adaptive Testing and Differential Item Functioning

Generalizability Theory

Filters

Exam 5: Reliability Overview: Classical Test Theory

Identify and explain each component of the formula in the basic conception of Classical Test Theory.

Reliability can be conceptually viewed as a ratio of ________.

________ is the primary concern of reliability.

Explain the statement: Classical Test Theory is test dependent and sample dependent.

In Classical Test Theory, an individual's true score would be ________.

Classical Test Theory considers how many sources of error at a time?

In Classical Test Theory, a measure's error is assumed to be ________.

In Classical Test Theory, T represents the ________.

Explain, in terms of true score variance and error variance, the impact on reliability of doubling the length of a test.

A research measure was found to have a reliability of .85. The researcher concluded that 15% of the variability in observed scores reflected ________.

Classical Test Theory makes two very important assumptions about the error in testing. Identify these two assumptions.

If scores from a measure with less than perfect reliability are correlated with scores from a second measure, the resulting correlation can be expected to be ________.

If a measure randomly assigned scores to each participant, the measure would possess a reliability equal to ________.

All else being equal, increasing the length of a test would be expected to ________ the test's reliability.

Introduction and Overview

Statistics Review for PSYC Measurement

Sychological Scaling

Test Preparation and Specification

Estimating Reliability

Content Validation

Criterion-Related Validation

Construct Validation

VG Psychometric Meta-Analysis Copy

Tests Bias Unfairness and Equivalence

Developing Tests of Maximal Performance

Classical Test Theory Item Analysis

Scoring Tests

Developing Measures of Typical Performance

Response Biases

Multiple Regression

Exploratory Factor Analysis

Confirmatory Factor Analysis

Item Response Theory

Applications of Item Response Theory: Computer Adaptive Testing and Differential Item Functioning

Generalizability Theory

Filters