Because globalization typically requires translations and adaptations of test materials, the deployment of tests internationally poses problems for establishing their validity. Although many organizations assume that a test that has sufficient validity for selection purposes in one country also has sufficient validity when translated and used in other countries, the assumption may not be true if test-takers are not familiar with the test format or content, or when the translation and adaption processes modify the construct being measured.
The problems with translation alone are well known. Translations and back-translations often significantly distort the original meaning of test content and are likely to create significant problems for test users. This is particularly apparent when personality tests are involved. Professional standards (e.g., AERA et al., 2014; ITC, 2005) emphasize the importance of using both effective translation and adaptation procedures. Although often technically infeasible, establishing the equivalence of various versions of a test supports the notion that validity extends across these test versions.
The familiarity of different populations with different item types and content can affect the validity of a test in certain cultures, even when the test-takers possess the skill being measured. For example, analogies seem to be more familiar to American applicants than to other foreign nationals. Items that involve calculations using the metric system may be more valid in one country than in another simply because of the applicants’ familiarity with it.
Although the translations and adaptations of tests using unfamiliar item types or content may be adequate, the item itself may work to the advantage of those who are more familiar.