Assessment for the course was based on three extremely tough sit-down examinations as well as a term paper. At the very beginning of the course, O’Grady explicitly justified his use of examinations by saying that we were future teachers of linguistics and so we needed to know the material in detail. I happen to disagree with this argument, and under other circumstances I might have polemicized with O’Grady; but I was very favorably impressed with the fact that the professor had questioned his own assessment procedures and had made decisions about how to evaluate the students not out of an unthinking adherence to custom but out of a conviction (I would further suggest, a moral conviction) that it was the good and right thing to do, and furthermore, that he respected his students as thinking adults enough to explain his position to them. This brings me to another key component of our role as teachers in determining assessment procedures. All educational work is fundamentally rooted in context. In the case of assessment, then, I suggest that we have an ongoing moral duty to interrogate the context in which assessment is being used, in order to determine what is the good and right way to proceed with these particular learners at this particular point in their learning. In the case of Professor O’Grady, a key element in the equation were the needs of the students in their coming professional lives: His learners were all doctoral students in linguistics, and so the examination format was, in his view, appropriate. In any given case, this equation will be a complex one, including the future needs of students, their expectations, the nature of what is being assessed (vocabulary? writing skills? communicative ability?), systemic requirements, cultural preferences, and so on. Moreover, calling the decision-making process an equation is also inaccurate, because in reality the weighing of many factors is nearly always done in a holistic and flexible way. It is equally important to interrogate the nature and ostensible and real purpose of existing tests: that is, to take what Shohamy (1998), after Pennycook (1994) and others, called a “critical” approach to language testing. According to Shohamy, among other things such a stance:
• “views language tests as…deeply embedded in cultural, educational, and political arenas where different ideological and social forms struggle for dominance” (p. 332);
• “asks questions about what sort of agendas are delivered through tests and whose agendas they are” (p. 332);
• “challenges psychometric traditions and considers interpretive ones” (p. 332);
• “asks questions about whose knowledge the tests are based on,” including whether what is included in tests can be “negotiated, challenged and appropriated” (p. 333);
• suggests that “the notion of ‘just a test’ is an impossibility because it is impossible to separate language testing from the many contexts in which it operates” (p. 333). 77 Values in English Language Teaching The one key word that Shohamy (1998) did not use, although she implies it in all the foregoing, is value. What values are enshrined in, or presupposed by, the kinds of tests we use and the ways in which we use them?

Continue Reading

The preceding discussion suggests another central moral paradox of language teaching. On the one hand, without some form of evaluation our students cannot be sure of their progress or how this matches up with the requirements of the systems within which they are studying; furthermore, although administrative convenience may seem a paltry motivation for particular forms of testing, it is also true that it represents one domain in which learners are treated in some way equally. (One could also argue that, in principle, an efficiently functioning administration should also be in the best interests of students.) On the other hand, any form of testing or assessment that we use is unavoidably only a partial, indirect, and subjectively judged reflection of the student’s actual abilities; this is true both because of the inherent qualities of assessment procedures and the impossibility of ever conclusively determining what it means to know a language. Thus, we have a moral imperative to offer some form of assessment, yet any form of assessment is morally suspect and fallible. This clearly leaves us as teachers in an uncomfortable position. Is there any way out of it? Well, at one level there is not; like the other moral paradoxes and dilemmas we have The Morality of Testing and Assessment 76 examined in this book, the paradox of testing simply represents a constant factor in our work. It is better seen as a dynamic rather than a problem; that is, it is simply a permanent characteristic of what we do, rather than some obstacle that will eventually be overcome. On another level, however, I believe there are ways forward. First of all, I suggest it is incumbent on each of us as teachers to continually reflect on our own values and continue to question whether these values are accurately reflected in our assessment procedures. A few years ago, when I was in graduate school, I took a course in transformational syntax from William O’Grady, a well-known theoretical linguist.

Continue Reading

To what extent is subjectivity an advantage, and how can I know when it becomes a hindrance? In some teaching situations, teachers address this problem by being highly specific about the components of the portfolio. My daughter’s sixth-grade teacher used this approach. For example, draft writing components must include a 1-page outline (100–120 words), two drafts, and a final draft. Furthermore, many teachers simply note the presence or absence of certain items, not paying attention to their quality. This simplifies the procedure and renders grading more uniform and less subjective, but it also has the effect of shifting the portfolio back toward traditional assessment in its inflexibility and reliance on quantity and not quality. Each of these decisions carries complex and usually contradictory moral consequences. At this stage in the book I hope it is not necessary to point out that the preceding discussion is not intended to argue against portfolios. As I mentioned, I continue to use portfolios myself, and I see them as being considerably superior to traditional forms of testing. My point in this section is that even if we choose to use portfolios or other forms of alternative assessment, we still very much need to be aware of the moral consequences of our decision and the moral complexities with which it is fraught. It is only through reflection on these issues that we can move toward what for us is a morally grounded approach to evaluation.

Continue Reading