Review questions

1. What are criterion levels? What do criterion levels have to do with CRTs? How are percentage scores different from percentile scores and how are these concepts related to the differences between CRTs and NRTs?

2. When is it reasonable to expect a CRT distribution to be positively skewed? Negatively skewed? How are CRTs different from NRTs in this regard?

3. What are three ways of statistically describing the central tendency of a distribution of numbers? How is each statistic defined? How are they different from each other?

4. What are two ways of statistically describing the dispersion of a distribution of numbers? How is each statistic defined? How are they different from each other?

5. What is the item facility index? How do you calculate it? How do you interpret the results of your calculations?

6. What is the item discrimination index? How do you calculate it? How do you interpret the results of your calculations?

7. What is the item difference index? What role does item facility play in calculating item difference indices? What is the difference in the interpretation of a difference index when it is based on an intervention study and when it is based on a differential groups study?

8. What is the B-index? How is it calculated? How does it differ from the difference index?

9. What is the agreement statistic? How is it calculated? How is it related to the difference index and B-index?

10. What is item phi? How is it calculated? How is it related to the difference index, B- index, and agreement statistic?

11. How can item response theory be used for developing equivalent forms and for item banking? What are the differences between the one-parameter, two-parameter, and three-parameter models in IRT?

How could you use FACETS analysis to address rater severity problems in composition grading?