Chapter 14: Criterion-Referenced Language Assessment in Blended Environments

One of the primary considerations in language testing is the frame of reference (or type of score interpretation), according to which tests are divided into norm-referenced tests (NRTs) and criterion-referenced tests (CRTs). In the case of NRTs, each student’s performance is compared to the performances of the other students taking the test, as well as to the performances of the norm group (the group representing the population). The purpose of norm-referenced testing is to see whether or not a given student’s performance is close to what is typical of the entire population of similar students. An ideal NRT produces a set of scores that are normally distributed around a mean (see Figure 14.1a), which results in placing students in rank order of educational achievement. In the case of CRTs, on the other hand, a student’s performance is compared to a predetermined criterion (a standard specifying a minimally acceptable performance at a given level). The purpose of criterion-referenced testing is to see whether, and to what extent, the student has attained mastery of the specified domain. If most of the students taking a test have indeed mastered the domain of content, the scores will be left-tailed, as in Figure 14.1b (for more on the differences between NRTs and CRTs, see, e.g., Bachman, 1990; Miller, Linn, & Gronlund, 2009; Urbina, 2004).