Chapter 7 Flashcards
Item Difficulty Index (P)
EX: p= # of examinees correctly answering item
__________________________________
Number of examinees
Class of 30 students, 20 answer correct divide by total examinees to get .66 20/30=.66
the percentage or proportion of the test taker who correctly answer the item.
Reliability and Validity of test scores
are dependent on the quality of the items on the test. If you can improve the item quality you improve the test.
Optimum Item Difficulty level
.50 p of .50
optimum for reliability and validity
Mastery tests are criterion-referenced and are pass or fail so p=.90
Item Discrimination
2: Discrimination Index
Item-total correlation
how well item can discriminate in test takers who differ on the construct being measured.
EX: reflects the items ability to distinguish between good and poor examinee skills.
Discrimination Index 0.0 to .10 poor .11 to .29 fair over .30 value is acceptable .40 and up is excellent
selecting two groups for total test performance.
Select top and bottom 27% of test takers excluding middle 46%
D= Pt-Pb
Item-total correlations Coefficients
Now easier to compute
Correlate performance of item with total test score.
Item-discrimination on mastery tests
- Instruction vs. No instruction
- Administer test to sample before and after instruction
- Mastery vs. non mastery
Distracter analysis
allows you to examine how many examinees in the top and bottom groups selected each option on a multiple choice.
Qualitative item analysis
taking a few days to cool off then review the test.
have a colleague review the test.
Have examinees provide feedback.
Should you use qualitative and quantitive approaches to improve test items?
yes.
Item Response Theory (IRT)
Theory or model of mental measurements that holds responses to items are accounted by latent traits.
Item Characteristics curves
a graph with ability to reflect on the horizontal axis and the Probability of a correct response reflected on the vertical axis.
pg 237
