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.

766<br />

DEOCS: A New and Improved MEOCS<br />

Stephen A. Truhon<br />

Department of Social Sciences<br />

Winston-Salem State University<br />

Winston-Salem, NC 27110<br />

truhons@wssu.edu<br />

ABSTRACT<br />

The Defense Equal Opportunity Management Institute (DEOMI) has studied<br />

equal opportunity through the use the <strong>Military</strong> Equal Opportunity Climate Survey<br />

(MEOCS) for more than a decade. In the process of updating the MEOCS a new version<br />

called the DEOCS (DEOMI Equal Opportunity Climate Survey) has been developed,<br />

which uses items from the MEOCS-EEO (Equal Employment Opportunity version) that<br />

have been neutralized (i.e., direct references to a majority and minority race and male and<br />

female gender have been removed). A three-step process was used to compare the<br />

DEOCS items with their counterparts in the MEOCS-EEO. Item response theory (IRT)<br />

using the MULTILOG program was performed to calculate difficulty and discrimination<br />

parameters. These parameters were matched to a common scale through the use of the<br />

EQUATE program. Differential item functioning (DIF) was then performed through the<br />

use of the DFIT and SIBTEST programs to discover any item bias. Results showed that<br />

few items displayed item bias (i.e., DIF), usually when those items were extensively<br />

reworded for the DEOCS. In most cases those items displaying DIF the version in the<br />

DEOCS had superior psychometric properties compared to their versions in the MEOCS-<br />

EEO.<br />

INTRODUCTION<br />

A major research project for the Defense Equal Opportunity Management<br />

Institute (DEOMI) has been the development and testing of the <strong>Military</strong> Equal<br />

Opportunity Climate Survey (MEOCS; Landis, Dansby, & Faley, 1993). This project<br />

includes revising the MEOCS and keeping it up to date.<br />

Suggested revisions to the MEOCS have included shortening it and making its<br />

items more neutral (i.e., replacing references to “majority,” “minority,” “men,” and<br />

“women” with more general terms “race” and “gender” and then using demographic<br />

information to determine the respondent’s specific race and gender). Various methods for<br />

shortening the MEOCS have been examined including confirmatory factor analysis<br />

(McIntyre, 1999), cluster analysis (Truhon, 1999), and item response theory (IRT;<br />

Truhon, 2000, 2002).<br />

Fifty-one items from the MEOCS-EEO have been rewritten to be more neutral in<br />

the DEOCS. The purpose of the current study was to compare the revised items with their<br />

original versions through the use of IRT and DIF.<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!