DefinitionComputer based testing is the administration of an examination in fixed or adaptive formats using the power of a computer. Over the past four decades, there has been incremental growth in computer-based testing (CBT) as a viable alternative to paper-and-pencil testing.
Today, CBT is a broad-based industry that encompasses a large variety of assessment types, purposes, test delivery designs, and item types appropriated for educational accountability and achievement testing, college and graduate admission testing, professional certification and licensure testing, psychological testing, intelligence testing, language testing, employment testing, adult education, military use. The delivery of CBT has also undergone many changes from the early days of “dumb” terminals connected to a mainframe or minicomputer. Testing locations or sites include dedicated CBT centers, classrooms or computer labs in schools, colleges, and universities; temporary CBT testing facilities set up at auditoriums, hotels, or other large meeting sites; and even personalized testing in the privacy of one’s home, using a PC with an Internet connection and an online proctoring service.
Modern CBT can be implemented in any of five ways: (a) on a stand-alone personal computer (PC); (b) in dedicated CBT centers; (c) at temporary test centers; (d) in multipurpose computer labs; or (e) using a PC, laptop, netbook, tablet, or hand-held device connected to the Internet, possibly remotely proctored. One of the first large-scale computerized-adaptive testing programs to go operational was the College Board’s ACCUPLACER® testing program, which in 1985 consisted of four tests: Reading Comprehension, Sentence Skills, Arithmetic, and Elementary Algebra. These examinations were introduced to assist in placing entering college students in English and mathematics courses.
Navigation relates to how the examinee moves around in a test. There are two aspects to navigation: (1) visual style of the navigation control and (2) blocking review and/or changing answers to previously seen items. The design and visual style of navigation software controls differs across test delivery drivers (and sometimes across test delivery system platforms). Every test has some navigation mechanisms. Some CBT test delivery drivers merely use “forward/next” and “back” keys or mouse-clickable buttons to move item by item. Some CBT test delivery drivers provide a full-page “review screen” to display all of the items, the examinee’s item response (if any), and any flags left by the examinee indicating that he or she wants to possibly review the item later. Many of the recent genres of CBT graphical user interfaces now provide an “explorer” or “helm” style of navigation control that continuously shows an ordered list of the test items in a narrow scrollable window within some segment of the display screen. This ordered list format is particularly helpful to examinees who want to skip items and go back to them later, if time permits.
One of the most important CBT implementation issues for high stakes examinations is item pool exposure (Haynie & Way, 1994; Stocking, 1993). The inherent flexibility of offering CBT on-demand or over a wide range of test dates potentially exposes the item pools to both small-scale and large-scale efforts aimed at cheating. That is, realizing that item pools may remain active over an extended period of time, examinees can conspire to memorize items, their intent being to reconstruct as much of the pool as possible to advantage re-takers or to share with future first-taker examinees.
“Extracts from a Research Report 2011-12 “A Review of Models for Computer-Based Testing” By Richard M. Luecht and Stephen G. Sireci.”
Julius Macaulay is the Principal Consultant at TECRES Consult (www.tecres.com.ng) providing ICT Training for Schools and Computer Based Testing Tutorials for Senior Secondary School, JAMB and Post-JAMB UTME students. He holds a Masters degree in Information Technology with Distinction and over a decade experience in Information Technology consultancy.