logo  ELT Concourse teacher training
Concourse 2

Testing in the classroom: an essential guide

test

You teach a child to read, and he or her will be able to pass a literacy test.
George W. Bush

Testing is a complex and technical area which is littered (some might say infested) with terminology.  There is a more complete guide to the area in the in-service section (most of which you don't need to know for the purposes of testing in the classroom).


Define

Defining terms

Which of following do you see as a test?  Click on the table when you have an answer.

testing task 1

Trick question?
No, not really.  The point is that any of these activities can be seen as a form of testing.  You are asking questions or setting tasks to find something out about the learners' abilities.
We can take them one at a time:

a public examination
This sort of test is instantly recognisable to most of us.  It is usually a very formal procedure with strict rules for its conduct.
The obvious examples are tests such as IELTS, Cambridge First Certificate (and the other levels in that suite) and so on.
doing that exercise
In this case, the exercise is only marginally a test because the only person who will look at the result is you.  However, in a classroom, you might set a task like this to find out what people already know.  In this case is was not a test but an awareness-raising activity to get you thinking.
asking someone if she understands
this is a kind of test but it's a pretty poor one because the answer won't tell you a great deal.  Learners from a range of cultural backgrounds will normally respond positively to a question like this because that's the least threatening thing to do.  A test to avoid.
using a multiple-choice exercise
You may be using such an exercise as a teaching tool, to alert people to the range of choices, for example.  However, most learners will see it as some kind of test and the outcomes will, presumably, be evaluated in some way.  So it's a test.
saying, "Can you pronounce this word?"
This is a test if you want to hear the learner pronounce it rather than being satisfied with an answer like "Yes, thanks" or a nod.
using a role play exercise at the end of a lesson
The purpose of activities like this is twofold:
a) they give the learners an opportunity to use the language in a freer way and make it a bit more personal to them but
b) they also allow you to discover if they can actually use it in the way you hoped.
So it's a test.
asking what a word means
This is clearly a test.  You want to know if the person knows the word or not.  Be aware, however, that giving an off-the-cuff definition of a word in a foreign language is very challenging (that's what keeps dictionary writers in business).  It's perfectly possible to know all you need to know about a word and still not be able to give a neat, clear definition of it.
asking a learner to tell you about yesterday
This may just be a bit of social chit-chat but in a classroom, many learners will assume they are being tested and their production evaluated.  Good teachers will be listening carefully to see how well the learner can do this.

thinking

Some terms to help us think

If almost everything we do can be seen as some sort of test, what are the differences?  The following is not meant to bamboozle you with terminology but to sort the wood from the trees and help us focus.  There are only four terms to grasp and they are pretty self-explanatory.

Can you make a stab at what any of these mean?

  1. formal testing
  2. informal testing
  3. formative testing
  4. summative testing

Click here when you have an answer.

informal

Informal tests

Good informal testing is something that comes with training and experience.  It's not rocket science and one of the best ways to do it is to listen carefully to your learners and ask questions which target what you need to know.
There's more on this in the guide to checking learning and the guide to asking good questions.
What follows concerns more formal, often written, testing.


formal

Formal testing

Validity

This is a term you will often hear applied to testing.  It means three things:

  1. Does the test measure what we say it measures?
    For example, if we set out to test someone's ability to participate in informal spoken transactions, do the test items we use actually test that ability or something else?
  2. Does the test contain a relevant and representative sample of what it is testing?
    For example, if we are testing someone's ability to write a formal email, are we getting them to deploy the sorts of language they actually need to do that?
  3. Does the test look like a proper test?
    You want your learners to take it seriously so presentation is important.

Simple.  Of course, we want our test to be valid so here's a short guide to how to write a good test.

The test can be formative (in the middle of a course or even a lesson) or it can be summative (coming at the end of a course or series of lessons).

The process:

making a test

Deciding what to test, prioritising and organising

This is where good record-keeping comes in handy.  Briefly, this means

Selecting test items and working out how to mark the test

Now you have your list, you can decide how to test the items.

Types of tests

true or false
This called an alternate-answer test.  You can expand it slightly to include True, False and Don't know (the answer isn't in the text).
For example:
In the list above there were three factors concerned with Validity:
True: O
False: O
multiple-choice
This is sometimes called a fixed-response test.  Typically, the correct answer must be chosen from three or four alternatives.  The 'wrong' items are called the distractors.
For example:
Choose the best answer:
Summative testing is designed mainly to:
A: help the learning process
B: plan the rest of a lesson
C: focus the learners on what they need to learn
D: find out how well something has been learnt
structured response
In tests of this sort, you give a structure in which to form the answer.  Skeleton sentence items of the sort which require the subject to expand a sentence are tests of this sort.
For example:
Use the words here to make a correct English sentence:
He / come/ my house / yesterday / 9 o'clock
______________________________________________________

(Answer: He came to my house at 9 o'clock yesterday)
free response
In these tests, no guidance is given other than the instructions (called the rubric, incidentally) and the learners are free to write or say what they like.  A hybrid form of this and a structured response item is one where the subject is given a list of things to include in the response.
For example (free response):
Write 200 words about a good holiday experience.
For example (hybrid):
Write 200 words about what method of transport you prefer
Say:
What it is and when you last used it
Why you like it
List three advantages
List two disadvantages
Finish with a recommendation

Have a look at this graphic and decide what goes in the columns on the right.  Then click on it to compare your answers.

test types


examples

Some more examples


  Data Task Comment
1 A text about someone's hobbies and interests.  For example,
I much enjoy walking and reading but have little time for sports and outdoor games which are just too energetic for me now that I'm getting on a bit.  I don't watch much television these days because I find it all rather dull and depressing.
The writer states that she used to do a lot of sports.
True or False?
The longer the text and the more complex the sentence structure, the harder it is.
The more the reader has to read between the lines, the harder it is.
2 The same as above. Choose the best answer:
a) She has never watched television much
b) She dislikes depressing television programmes
c) She never went in for outdoor activities
d) She has never played sports
The nearer the distractors are to the truth, the harder it is.
The more different the words in the question are from the text, the harder it is (see c)).
3 She / sports / now / she / getting Expand the sentences making the necessary changes You usually need a bit of context to get reliable responses.  With the text above, the response should be She doesn't play much sports now because she is getting on a bit.
4 She _______ play _______ many ________ as she _______ to. Fill the gaps with a single word. This is sometimes called a Cloze test although it isn't technically.  A better term is a gap-fill test.
5 Same text as above. Now write a similar paragraph about your hobbies and interests using the text as a guide.  Write no more than 150 words. This is a hybrid between free and structured response.


mark

Marking tests: reliability

Some of those test items require more judgement by the marker (usually you) than others.
The more subjective the marking is, the less reliable will be the results.  For example:

There are, of course, many more variants and types of test item than we can cover here.  Look at tests in coursebooks and be aware that most examination boards and ELT publishers have examples of tests and test types on their websites.  Go there for more ideas.



Related guides
testing the fuller and more technical guide to this area in the in-service training section
asking good questions for a guide to how to check learning and use formative, informal testing in the classroom
checking learning a related guide to informal formative and summative assessment


Of course, there's a test on all of this: some formal, summative evaluation for you.


Reference:
A very accessible and clear text is:
Hughes, A, 1989, Testing for Language Teachers. Cambridge: Cambridge University Press