Testing for TEFL Teachers

Introduction

Tests are everywhere in language teaching. Students often do a “placement” test before starting a course, and may do regular “progress” tests and a “mid-year” or “final” test. Some courses are aimed at preparing students for internationally-recognised tests, such as TOEFL, IELTS or the Cambridge ESOL suite of exams (KET, PET, FCE, CAE, CPE).

This session aims to interest you in some of the issues surrounding testing, to consider the various reasons for testing in the language classroom, and to give an overview of common testing techniques. When you have completed this session you should be in a position to decide whether and how to test your students, as well as to evaluate professionally-produced testing material or, if necessary, to design and produce your own tests.

Why testing is a “hot” issue

Motivating or de-motivating?

Do tests motivate or de-motivate? Think of your own experience of taking tests or exams (at school, on language courses, at university). Were your experiences positive or negative? Why? At Madrelingua School of English our clients expect professional language teaching (which probably includes elements of testing) but also want to be encouraged and motivated. Given that they pay a lot of money for their courses, their level of satisfaction and feeling of making progress is extremely important both to them and to us. There is therefore a delicate balance to be struck!

Too many or too few?

Do students benefit from more frequent tests? Or would it be better to have fewer? Or none at all? What would be the right number?

The “positive wash-back effect”

Does what’s in the test/exam have any effect on what gets taught in the course? Is it a positive or a negative effect?

Evaluation and certification

Who’s the test for? For the student, for the school, for an employer or potential employer? What does the test really mean anyway?

Test content, formats, standards and technologies

Have we changed what is taught and how?
Are tests getting easier, or are teachers getting better at training students to take them?
What about computers and the Internet in testing?

The Why, What and How of testing

Why?

First decide WHY you are testing. Here are some typical reasons:
  • It’s required by the institution you work in
  • Your students expect or want to be tested
  • It will have a motivating effect (encourage them to study or revise)
  • It will help students evaluate their progress
  • It will help you evaluate their progress
  • It’s the best or only practical way to organise students into coherent groups
  • It’s good practice for some more serious exam down the road
  • It brings “closure” to your course
If you can’t identify the purpose of your test, it may just be better to forget the whole idea!

What?

Having decided WHY, it should be easy to choose the WHAT. For example:
  • speaking (fluency, accuracy, pronunciation, communication strategies, what?)
  • listening (gist, detail, what?)
  • reading (speed, comprehension, what?)
  • writing (what format or formats, marked how? Accuracy? Content? Style?)
  • vocabulary
  • grammar
  • other
You’ll notice that I’ve put grammar near the bottom of the list. In fact, it tends to be the thing that gets tested most? Why do you think that is?
Another issue of importance: do you test just what you’ve taught, or do you test in a more general way (i.e. a skills-based test appropriate to the students’ level).

How?

The HOW, in turn, depends on the WHY and the WHAT.
Remember tests should be:
  • objective (as far as possible)
  • appropriate (to the reason for testing, and to the people being tested)
  • valid, that is to say, effective (testing what is intended, not something else)
  • efficient (not causing too much chaos, not requiring too many resources, quick to mark)

Overview of testing techniques

Remember that certain testing techniques may be more common in some countries than in others. Italian schools and universities often use oral testing, whereas in Anglo-Saxon countries essay-writing is much more common. You need not necessarily follow the normal practice of the country that you teach in, but you should bear it in mind. Italian students, for example, need a lot of help writing decent essays, but tend to come across well when interviewed. Choosing to test exclusively using either one of these methods may therefore produce deceptive results.

Common testing techniques

  • Essays
  • Interviews
  • Listening tests (question types vary)
  • Reading tests (ditto)
  • Multiple choice
  • Cloze (a gapped text)
  • Multiple choice cloze (each gap in a text has multiple answers given)
  • Sentence transformation (reconstruct sentence B so it means the same as sentence A)
  • Complete the sentence
  • Complete the dialogue
  • Identify/correct the errors
  • True or false?
  • Conjugate the verb (text where verbs are given in infinitive form and have to be put in the correct tense).
  • Matching – i.e. opposites, sentence halves
  • Ordering (stages in a process, words in a sentence, etc.)
  • Changing (active to passive, present to past, direct to indirect, formal to informal, etc.)
Which of the above techniques would be the most/least objective? Which would be fastest/slowest to mark? Which would be most suitable for testing beginners/advanced students?

Evaluating test material

You’ll find lots of test material in course books and teacher’s books. You could adapt material intended for classroom use, or for homework. You could use “practice test” material too (i.e. PET, FCE, CAE). One advantage of using the latter is that it is likely to be of good quality and to be appropriate for a certain specified level (course book tests are sometimes written by freelancers or junior members of the writing team, with consequent loss of quality!).

Consider the test material according to the WHY, WHAT, HOW criteria discussed above, but also consider the TIME aspect. You test should not be too long (that it becomes boring, repetitive or wasteful) or too short (so few questions that the result may vary wildly – i.e. in a test with only 10 questions, one question right or wrong will represent a 10% difference in the final result, which may be too much to effectively differentiate the performance of different students).

Check also that the instructions are clear and that you have the answers handy (or can work them out for yourself). It’s helpful if the total number of marks is a nice round number, as it makes calculating percentages much easier!

Designing and producing your own tests

Bear in mind that professional test writers go through many stages of evaluating and testing their tests to make sure they are good enough for the purpose. Your material is not likely to be perfect at the first attempt. That said, no one knows your students and their needs like you do, so you may well be the best person to design or produce their test.

Here’s a suggested proceedure if you’re thinking of producing your own test:
  1. What’s the reason for the test?
  2. Are you testing just what you’ve taught, testing their communicative skills, or both? Make this decision first.
  3. How much time have you got to administer and mark the test?
  4. How much testing material do you need?
  5. What material is already available for you to re-use or adapt?
  6. What do you need to write, if anything?
  7. What format or formats would be most valid, effective, efficient, objective and appropriate?
  8. How many questions do you want in total?
  9. Sketch an outline of the sections, with the number and type of questions in each.
  10. Write your first draft of the test material.
  11. HAVE SOMEONE CHECK IT, or better still, DO IT.
  12. Revise your draft as necessary.
  13. Pilot your test with a smaller group if possible, preferably students whose level/ability is already known to you.
  14. Modify your material if it’s too easy or hard.
  15. Don’t forget to make a teacher’s key!

Final thoughts

Surprise tests

Do students need to know about a test in advance? Or not? Will stressing them out with advance notice be beneficial, or not? What effect will it have on the usefulness of the test?

Administering tests

How “heavy” do you want to be? Test conditions should match students’ expectations and the reasons for testing. A friendly group of adult professionals in an evening class may or may not want to be tested in the same way as a group of rowdy middle school children.

What to do if the results surprise you?

It’s always a good idea to consider in advance what will happen if your test throws up an unexpected result i.e. all your best students fail.. Are the test results binding? Can students retake the test? Does the test mark make up the whole of the overall grade, or is there also a subjective component, such as the teacher’s evaluation?

Other ideas

Could students mark their own tests? This is VERY motivating, and easy on the teacher’s social life!
Why not discuss testing techniques with your students and find out their views?
Why not let students write their own test?
Why not do a “project” instead of a test?
What about some form of “peer” or “self” evaluation?

Footnote: The Art of Placement Testing

Placement tests are used to assign new students to an appropriate group or, in the case of individual students, to evaluate a student’s level in advance of the first lesson in order that the teacher can choose appropriate materials.

Test format and function

The actual test is likely to include a written and spoken component, perhaps a short multiple choice test and a brief interview with a qualified person (a trained teacher, or the Director of Studies).

Given that placement tests may be done BEFORE a course has been sold in many cases (in order to advise a potential client which day/time/level combinations are available, for example), the “testing” proceedure is also an opportunity to find out about objectives/preferences/previous experiences. Relevant information might include the following:
  • what are the student’s objectives?
  • does the student need to do an exam or reach a particular level, and if so why and when?
  • when/where has the student previously studied the language?
  • does the student already know his/her level (perhaps he/she has just finished a course or done an exam elsewhere), or have expectations about what his/her level should be (people who claim to be beginners may often not be, and vice-versa!).
  • (if the student appears to be between levels) would the student prefer a more demanding or an easier class? Does he/she have the time and motivation to do a bit extra to catch up, or would he/she prefer to take it easy?

Who knows best?

Does the placement test function as a “gatekeeper” in the sense that getting a certain score “opens the door” to a particular level whereas failing to get a certain score “bars entry” to certain classes? Or does the placement test serve to “inform” the student in making his/her choice of classes?

Perhaps the answer is that it should be somewhere between the two: at the end of the placement testing processs, the student should be aware of which options are available, and this may well involve a choice between a higher or a lower level, or between classes of different types (conversation, general English, exam preparation, individual lessons, etc.). Sometimes, the tester can help make such a decision, but sometimes it may be a choice best left to the individual, especially given that personal circumstances, motivation and capactity for language learning are all factors which vary enormously and are likely to be unknown to the person administering the test.

Conclusion

And back in the real world, commercial imperatives must always be kept in mind too. Overall quality (and so customer satisfaction) is ensured in part by having coherent groups. However, turning students away because the school is unwilling or unable to meet their perceived needs implies a waste of the resources invested in marketing and publicity, and may generate poor word-of-mouth in the local market. Compromise may be required! Which brings us back to placement testing, which in order to be effective on these many different levels needs to be seen as a delicate art, rather than a science!

We hope that you find this material useful. As usual, please e-mail any comments you may have to:

info@madrelinguabologna.com