Free research paper on kurpius r stanford m e 2006 testing and measurement a user-friendly guide new york

Introduction

Test item evaluation is important in determining the reliability and effectiveness of the test items in a psychological test. Test item evaluation can take two forms; the criterion referenced or the norm-referenced form. Usually, test evaluations are done in criterion referenced form. If the norm-referenced form is used during the test, the criteria of evaluating discrimination indices and item difficulty changes.

Item difficulty

This is the percentage section of respondents in a test who pass the test. If the many respondents pass the test, the test has a high item difficulty measure. If the norm-referenced approach is used to measure item difficulty, the value of item difficulty is found by dividing the number of correct respondents by the total number of respondents in that test. This approach does not involve the criterion used for each item test evaluation. If a high percentage of respondents get high scores, the item has good item difficulty.

Item discrimination

This is the ability of a test item to discriminate the number of correct respondents from the respondents who do not give correct responses. In criterion referenced approaches, item discrimination considers other factors such as criteria used to grade respondents and the ways of discrimination. However, norm-referenced approaches of item discrimination only focus on comparing the high score respondents with the low scorers. An item that does good discrimination shows that many respondents score highly.

Conclusion

The main difference between criteria of item evaluation in norm referenced and criterion referenced approaches is that the norm referenced approach uses simple comparisons between high or correct scoring respondents with low or poor scoring ones.

Schultz, K. S., & Whitney, D. J. (2005). Measurement theory in action: Case studies and exercises. New York: Thousand Oaks, CA: Sage.