Quality Control
A practical advantage of CBT is the ability to quickly correct any errors found on a test. In PBT, an error in an item may require time-consuming and expensive reprinting of booklets. In a CBT environment, fixing a problem with a test item may be a matter of minutes, at practically no cost at all. On one operational testing program with which we are familiar (the Massachusetts Adult Proficiency Tests), occasionally examinees or test proctors find a typo on an item and contact the testing vendor, who fixes the problem the same day Of course we do not expect errors to be present on test items, but when they are, it is typically easier to correct them quickly in a digital environment.
Testing Students With Disabilities and English Learners
A valuable benefit of CBTs is that accommodations can be built into the tests and items to make them more accessible for certain types of test takers, such as students with disabilities (SWD) and linguistic minorities, such as English learners (ELs). Many examinees in credentialing or achievement testing arenas have disabilities that prevent them from performing optimally on a test when the constraints of standardization are in place. Other examinees may have limited proficiency in the language in which the test is administered. For these reasons, accommodations to standardized tests are often given to remove any barriers associated with the test administration that prevent examinees from demonstrating their true proficiencies.
Although test accommodations are designed to promote fairness for SWD and ELs (Abedi, this volume; Elliott & Ketller, this volume; Laitusis & Cook, 2007; Sireci, Scarpati & Li, 2005), two validity concerns arise whenever accommodations to standardized tests are granted. First, the accommodation may change the construct measured. Second, the accommodations might provide an unfair advantage to the examinees who receive them. Therefore, rather than provide accommodations to standardized tests, it is preferable to design the tests to be as accessible as possible, so that accommodations are not needed. CBTs can facilitate such design.
One of the greatest potential advantages of computerized item formats is the possibility of promoting access or “building in” accommodations for all examinees (Thurlow, Lazarus, Albus & Hodgson, 2010). For example, by clicking on an icon or submenu, examinees could access glosses that define a word; can increase the font size; can have an item, passage or directions read aloud; can translate a page into a desired language; or can request clarification of directions. Such assistance can be built into the digital test delivery system and be made available to all examinees, thus keeping the playing field level for all (Almond et al., 2010). Of course, any such accommodations should be evaluated with respect to potential alteration of the construct measured.
There are several interesting examples of innovative computerized item formats designed to improve the assessment of SWD and ELs. Russell and his colleagues have used avatars for sign- language accommodations (Russell, Kavanaugh, Masters, Higgins & Hoffman, 2009), and used “talking tactile tablets” for students with visual impairments (Landau, Russell & Erin, 2006). Kopriva and her colleagues (e.g., Carr & Kopriva, 2009; Kopriva, 2010; Kopriva & Bauman, 2008) demonstrated the benefits of technology-enhanced item formats and computerized accommodations for ELs. In addition, computer-based assessments in history and math that provide online accommodations, such as glosses, have been piloted for ELs (Crotts-Roohr & Sireci, 2014).
