Deck 20: Item Response Theory

Question 1

In item response theory models, the underlying latent trait is commonly referred to as ___________.

A) Theta
B) Constructs
C) Delta
D) Omega

Answer 1

A

Question 2

The relationship between the underlying latent trait and the probability of answering an item correctly is called an _____________.

A) Item Trace Function
B) Item Functioning Statistic
C) Item Response Function
D) Item Relation Graph

Answer 2

C

Question 3

Which of the following is not one of three parameters in the 3PL? 

A) discrimination
B) difficulty
C) guessing
D) reflection

Answer 3

D

Question 4

After a model has been estimated on your particular data set, which of the following would you examine to see how to specifically improve model-data fit?

A) Adjusted goodness-of-fit index
B) Individual error terms
C) Akaike Information
D) Model modification indexes

Question 5

One strength of item response theory compared to classical test theory is that IRT statistics estimated from one population can be compared to statistics collected from a different population. This property is called _________.

A) parameter invariance
B) population equivalence
C) sample independence
D) equatible comparison

Question 6

In confirmatory factor analysis, what is the preferred term by Shultz, Whitney, and Zickar (2021) for the item variance not explained by latent factors?

A) error
B) residual
C) marker
D) leftover

Question 7

In classical test theory, the concept of reliability is used to quantify measurement precision. In IRT, the analogous concept is called _________.

A) measurement correspondence
B) discrimination
C) information
D) theta overlap

Question 8

What is a minimum suggested sample size required to do most IRT analyses as suggested by Shultz, Whitney, and Zickar (2021)?

A) 100
B) 250
C) 750
D) 3000

Question 9

The idea that items should be uncorrelated with each other after controlling for the latent trait of the model is called ________.

A) Factoral independence
B) Factoral dependence
C) Common interdependence
D) Local independence

Question 10

Which of the following parameters is part of the 3PL model but not the 2PL model?

A) guessing
B) difficulty
C) discrimination
D) randomization

Question 11

What are some key differences between classical test theory and item response theory.

Question 12

What are assumptions and requirements that need to be met to use IRT?

Question 13

Which is better? Classical test theory or IRT?

Question 14

How can you decide what is the best IRT model for a particular data set?

Question 15

Describe the concept of item parameter invariance.