09.12.2012 Views

2003 IMTA Proceedings - International Military Testing Association

2003 IMTA Proceedings - International Military Testing Association

2003 IMTA Proceedings - International Military Testing Association

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

There is a link between test construction and test equating. This is illustrated in a<br />

statement by Mislevy (1992),<br />

Test construction and equating are inseparable. When they are applied in concert,<br />

equated scores from parallel test forms provide virtually exchangeable evidence about<br />

students’ behavior on the same general domain of tasks, under the same specified<br />

standardized conditions. When equating works, it is because of the way the tests are<br />

constructed…(italics in original, p. 37; from Kolen & Brennan, 1995, p. 246).<br />

Dodd, De Ayala, & Koch (1995) have described a procedure to use CRFs to<br />

calculate item information functions (IIFs). These IIFs can be added together to produce<br />

test information functions (TIFs). Wu (2000) developed a model management system to<br />

develop comparable tests making use of TIFs. These equating procedures can then be<br />

used to develop comparable forms of the DEOCS. In this way IRT can be used to<br />

develop alternate forms of the MEOCS and DEOCS (see Cortina, 2001) rather than<br />

compare forms after the fact, as was done in this study.<br />

The kinds of analyses described above set future directions for the DEOCS. They<br />

are frequently used in computerized adaptive testing (CAT). Most CAT has been done in<br />

the area of ability testing. However, CAT has been successfully applied to attitude testing<br />

(Koch & Dodd, 1990). A CAT-version of the DEOCS could establish a person’s response<br />

level on the different scales in the DEOCS with a minimal number of items.<br />

Finally, the DEOCS has begun to go on-line. DIF analyses should be used to<br />

compare on-line responses to paper-and-pencil responses to the DEOCS. While previous<br />

DIF research suggests that administration format does not make a difference for attitude<br />

scales (Donovan, Drasgow, & Probst, 2000) and evaluations (Penny, <strong>2003</strong>), it would<br />

useful to verify this.<br />

REFERENCES<br />

Baker, F. B. (1995). EQUATE 2.1: A computer program for equating two metrics in item<br />

response theory [Computer program]. Madison: University of Wisconsin,<br />

Laboratory of Experimental Design.<br />

Bolt, D. M. (2002). A Monte Carlo comparison of parametric and nonparametric<br />

polytomous DIF detection methods. Applied Measurement in Education, 15, 113-<br />

141.<br />

Chang, H., Mazzeo, J., & Roussos, L. (1996). Detecting DIF for polytomously scored<br />

items: An adaptation of the SIBTEST procedure. Journal of Educational<br />

Measurement, 33, 333-353.<br />

Cortina, L. M. (2001). Assessing sexual harassment among Latinas: Development of an<br />

instrument. Cultural Diversity and Ethnic Minority Psychology, 7, 164-181.<br />

Dodd, B. G., De Ayala, R. J., & Koch, W. R. (1995). Computerized adaptive testing with<br />

polytomous items. Applied Psychological Measurement, 19, 5-22.<br />

769<br />

45 th Annual Conference of the <strong>International</strong> <strong>Military</strong> <strong>Testing</strong> <strong>Association</strong><br />

Pensacola, Florida, 3-6 November <strong>2003</strong>

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!