Stephen van Vlack

Sookmyung Women`s University

Graduate School of TESOL

Teaching Reading

Week 15 - Answers

Nuttall (1996), Chapters 13 and 14

1.Nuttall (1996) claims that it makes pedagogical sense to make a test deliberately easy. Do you agree or disagree and why?

This is a valid idea in contrast to many teachers who try purposefully to try to trick students into getting wrong answers as a way of punishment or as a way of differentiating students (to what purpose??), so we wholeheartedly salute it`s sentiment, but, at the same time it is also a vast overgeneralization. Making a test easy depends on many different variables and considerations. Well, like everything else it is all a matter of degree. We absolutely salute the sentiment, for many reasons, not the least of which is that testing must follow classroom instruction, and we really don`t want there to be built in tricks in what we do during the class, but still we need to think about the purposes of the test. Why do we test, really? Is it not possible to make our tests easy? Sometimes maybe. A test that is too easy will certainly not suit all our testing purposes. Think, for example, of a placement test. Now, we certainly don`t want it to be easy for if it is we will not be able to determine where the student has done well or done not so well. A placement test needs to show what the testee can and cannot do so that he/she can be placed accordingly.

The thing that really bothers me about testing, and particularly in relation to writing, is that it is so far from the classroom experience or at least what we think we would like to do in the classroom as regards reading/literacy development. One thing we need to bear in mind is washback. A valid test needs to have good washback effects. This means that we need to teach for the test, yes and we also need to make sure our test covers only the things we have covered in the class. In this class we have been advocating the use of more open-ended, task-based activities in relation to the text we expect our students to read. We hold fast to the belief that there are many ways of interpreting a text and this makes testing with closed-ended answers hard to do. It does not make sense, it is in fact highly detrimental, to test in a fundamentally different way than the way we teach. This is a dilemma we need to face, because it really does affect test validity.

2. What are the different purposes of testing for reading skill?

Nuttall mentions five basic purposes for testing reading, one of which is, strangely enough, void of purpose. This reflects the reality of many teaching situations where parents and society at large desire us to test so that they know how their children are doing related to other children. This is not really a valid reason for testing. Testing just for the purposes of grading is not a valid purpose from the teacher`s point of view. It is this type of testing which falls under the `no purpose` category.

Placement tests - should be various, more general skills

Diagnostic tests - should focus on different skill, strategy areas specifically

Progress tests - should be highly focused on what was actually taught in the class

Achievement tests - These are similar to placement tests but a little more general

No purpose - These are tests given only because a teacher needs to grade her students and we need numbers to do so.

As a teacher you need to have a clear idea of why you are testing and then, based on that, which purpose/type of test you want to develop. Just remember reading and literacy are skills based and testing using the same texts you taught with has little or no validity. If you are using the same texts then you can`t really test skills, you are merely testing for memory. Yikes!

3. What are some different aspects of reading skills that we can and want to test?

Well, we need to make a clear distinction between tests that focus on top-down or bottom up skills. Right between both these concerns, occupying a central position, is of course vocabulary. We need vocabulary to recognize or activate certain schemas, both formal and regular as well as trying to use vocabulary to get meaning in a bottom up way. In dealing with this, we concede that vocabulary is an extremely necessary aspect of literacy skill. Unfortunately, however, most of our tests related to vocabulary are generally recognition tests. Recognition tests are not going to really give us a clear indicator of how students are able to use vocabulary for reading purposes either as a top down or bottom-up tool. We need to test vocabulary not outside of context but in context and specifically in the context all of literacy.

4. What are some different methods for testing these reading skills?

Well, I think this is all well known to you. Nuttall lists several different kinds of methods after, of course, commenting that there is no one perfect or most useful method.

Discrete point

            Cloze tests and gap fillers

            Multiple-choice

            Short answer

Integrated approaches

            Information transfer

            *C-test

            Cloze-elide test

            Free-recall test

            Summary

            Gapped summary

I think the most important thing to remember about all this is that there are alternatives, real alternatives, to the multiple choice hegemony in language testing, particularly the testing of reading. We can create equally valid, and often even more valid, tests for our students by using other methods or various methods (yes, in the same test). Once ore, there should also be better washback effects for it is certainly true and not to be forgotten that any task we create for our students in the classroom can also be used as a test (under the right conditions of course).

OK integrated approach is the one that I think is the most interesting and useful is the C-text. What I like about the C-text is that while it is an integrated approach to testing it is also easy to assess, unlike so many of the other integrated approaches mentioned above. Since there is a specifically correct answer for each blank on a C-test a teacher can grade them quickly and accurately without worrying about being to subjective. I see test can be extremely objective, depending on how strict the teacher may be in the grading curve.

5. Why do reading teachers need to read?

This should be extremely clear to us by now. We spoke at length in this class about how a proper literacy environment at home is extremely important in getting a child ready for literacy at school. Well, obviously the same true as some carries over to the classroom itself. If a teacher is going to get her students interested in reading, and in English reading, then she also needs to read. This means that she needs to show the students that she`s actually reading things in English. She can talk about the things she has read, or bring them in and show the students, or simply make sure that she also reads all the students or reading during the short extensive reading period of the class.