Jumat, April 06, 2012

Tests of Grammar and Usage

Posted by Unknown 18.07, under | No comments



The following are some of the most common types of objective items used to test awareness of the grammatical features of the language. Each type of item will be treated in greater detail in this chapter.
-          multiple-choice items
-          error-recognition items
-          rearrangement items
-          completion items
-          transformation items
-          items involving the changing of words
-          ‘broken sentence’ items
-          combination items
-          addition items

It should always be remembered that such items as the above test the ability to recognize or produce correct form of language rather than the ability to use language to express meaning, attitude, emotions, etc. nevertheless, it is essential that students master the grammatical system of the language they are learning. Thus, classroom tests of grammar and usage can play a useful part in a language program.


MULTIPLE – CHOICE GRAMMAR ITEMS, ITEM TYPES

The type of multiple-choice item favoured by many constructors of grammar tests is the incomplete statement type, with a choice of four or five options. This items may be written in any of the following ways:

Type 1     Tom ought not to ……… (A. tell                  B. having told                C. be  telling      D. have told)    me your secret, but he did.

Type 2     Tom ought not to ……… me your secret, but he did.
A.      tell
B.      having told
C.      be telling
D.     have told

                                                A. tell
Type 3     Tom ought not to      B. having told           me your secret, but he did.
                                                   C. be telling
                                                   D. have told.

Type 4     Tom ought not to have told me your secret, but he did.
A.      No change
B.      Tell
C.      Having told
D.    Be telling                    

 Item types 2 and 3 are preferable to 1 because the option do not interrupt the flow of meaning in the sentences: these items present the entire sentence so that it can be read at a glance. Unfortunately, type 1 confuses the reader because of the long parenthesis (i.e. the four options occurring between ought not to and me). Item type 4 shows the correct (or an incorrect) form as part of the sentence in such a way  that it cannot be compared on equal terms with the other options: a correct option, for instance, is generally easier to recognize when it appears in the framework of the sentence than as part of a list of distractors.
Another item type appears below, but it is not recommended since it requires the testees to spend time on unnecessary reading. Not only is it uneconomical but it does not present the ‘problem’ (i.e. the options) as clearly as item type 2 does.

Type 5   A. Tom ought not to tell me your secret, but he did.
            B. Tom ought not to having told me your secret, but he did.
            C. Tom ought not to be telling me your secret, but he did.
            D. Tom ought not to have told me your secret, but he did.

The following method is useful for testing short answers and responses:
Type 6   ‘Tom ought not to have told anyone the secret.’
               A. ‘So ought you’                    C. ‘Neither you oughtn’t.’
               B. ‘Neither ought you.’           D. ‘So oughtn’t you.’

   In type 7 requires the students to select the alternative which is true according to the information conveyed in each sentence. Such an item may be included either in a test of reading comprehension or in a test of understanding of the sentence.

 Type 7   ‘Tom ought not to have told me.’
A.      Tom did not tell me but he should.
B.      Perhaps Tom may not tell me.
C.      Tom told me but it was wrong of him.
D.    It was necessary for Tom not to tell me.

   It may be argued that an understanding of syntactical patterning is just as necessary for the following item:

            ‘ …… was Robert late last week?’
            ‘Three times.’
            A. How much                           C. How often
            B. How may                              D. How long

Items which appear in a test of grammar and structure should be made to sound as natural as possible. The following  mechanical test item:

            This book belongs to Peter. It is …….
            A. Mary’s book                         C. the book of Peter
            B. the book to Mary                  D. the book of Peter’s

can be rewritten as follows

            This book belongs to Peter, but that is ……
            A. Mary’s book                         C. the book of Mary
            B. the book to Mary                  D. the book of Mary’s

     Note that distractors should generally be correct both in written and in speech. The following item proved unsuccessful when it was included in a test because many of the more able students selected option D. the reason being that they pronounced used to quite correctly as use to/ju:stÓ™/:

            I …… to go to my uncle’s farm every weekend.
            A. am used                                C. was used
            B. used                                     D. use

Note that sample items discussed in this section have so far taken the form of short decontextualised items. In practice, however, such items would all form part of a paragraph or series of paragraphs of descriptive, narrative or expository prose. The provision of a detailed context in this way, however, often  limits the range of grammatical features being tested. It is usually impossible, for example, to test the past continuous tense in a narrative set in the past  (unless direct speech is used). Similarly, a paragraph describing a simple manufacturing process may not provide  the test writer with the opportunity to test all the verb forms and tenses he or she may wish to test. This is the price to be paid for including more  natural, contextualized items. This is real language used for a particular purpose. furthermore the provision of context helps to ensure tat there is only one correct option in each case. Short decontextualised sentences can lead to ambiguity as they are usually open to several interpretations when used as  stems for multiple choice items. For example, option D in the following decontextualised item might be correct (as well as option B) if the student happens to assist with research and deliberately catch a cold so as to be able to test various cures.

       I couldn’t take the test last week because I ………. a cold.
       A. have caught                                            B. would catch
       B. had caught                                             D. was catching

Much better for testing purposes is the following item. The passage is taken straight from a newspaper article and thus the language is authentic background knowledge and details to avoid ambiguity and alternative interpretations, and the newspaper report itself is very interesting. Does it really matter if it will not allow us the opportunity to test every point of grammar which we may want to test. Students taking this test are being given a real feel for the language they are learning.

         A long way from home
A 72-year-old Samoan  who (1) ……… no English at all spent  thirteen days (2) ……. on buses in the San Frasisco area he  had become separated (3) …..  his family, police said. (4) ….. said that Faitua Logo, (5) …….. moved to the United States two-years ago, left his son and daughter-in-law, etc………

(1)  A. is speaking             B. speaks           C. has spoken               D. was speaking
(2)  A. to ride                     B. was riding     C. ride                          D. riding
(3)  A. with                       B. from              C. by                            D. off
(4)  A. he                           B. they              C. one                          D. it
(5)  A. which                    B. that               C. who                         S. what
etc. …..


CONSTRUCTING MULTIPLE – CHOICE ITEMS

Although it is not always possible to use samples of students’ own written work to provide the basis for the test items, it should  not be too difficult for constructors of classroom test and school achievement tests to take advantage of the types of errors made by students in their free compositions and open-ended answers to questions.

Item 1
Let us ignore the error in the first sentence for the time being  and concentrate on the error of tense after hope.

Step   1: The first step is to reduce the length of the sentence and to correct the error (and any other errors in the original sentence). Thus,

         I hope that you wouldn’t mind on such a long  period between my last letter and  this one.

becomes   I hope you won’t mind waiting for so long.

Step  2: Next we write out the sentence, substituting a blank for the area being tested. We write in the correct option and the distractor which the student has provided for us. However, we have to add a sentence because in certain (rare) contexts, wouldn’t may be correct.

       I hope you ……. mind waiting for so long. I promise to reply sooner in the future.
       A. won’t                 B. wouldn’t

Step  3: We now add another two distractors. Again, we go to the written work of our students to provide these distractors. But if we cannot locate any suitable errors without too much difficulty, we use our own experience and knowledge of the target and native languages. Thus, two useful distractors which would also balance the existing two options might be shouldn’t  and  shan’t.

Step  4: One suggestion may be that we replace shan’t with can’t. if students from a particular language background make such mistakes as can’t mindcan’t should be used as a distractor, and possibly shouldn’t changed to couldn’t. as can be seen at this early stage, the actual process of item writing is extremely subjective.

       I hope you ……. mind waiting for so long. I promise to reply sooner in future.
       A. won’t                 B. wouldn’t                   C. couldn’t                    D. can’t

An alternative suggestion for a fourth option might be don’t  or  didn’t.

       I hope you don’t mind waiting for so long.
       I hope you didn’t mind waiting for so long.

 Unfortunately, both don’t and didn’t are correct. However, in the following context, didn’t  is not acceptable.

       ‘How long are you going to be?’
      ‘About half an hour. I hope you ……. mind waiting for so long.
      A. won’t                  B. wouldn’t                   C. shouldn’t                  D. didn’t

It may be argued that didn’t stands out too much. If so-and if it is equally useful to test the use of don’t (instead of won’t) after hope – the item could be rewritten as:

       ‘How long will you be?’
       ‘About half an hour. I hope you …….mind waiting for so long .’
       A. don’t                 B. wouldn’t                   C. shouldn’t                  D. didn’t

       Obviously, there are varying degrees of refinement in the construction of multiple-choice items. Furthermore, some items are much more difficult to construct than others. The following two items based on errors in the student’s letter are fairly simple to write.

Item  2
Error: ……. and enjoy looking the children playing.
Item:   Old Mr. Jones enjoys …….the children playing.
           A. looking                      C. looking on
           B. looking at                   D. looking to

Some test constructors might be tempted to use for as a distractor. It can be argued, however, that looking for is correct: old Mr. Jones might enjoy looking for the children playing (i.e. he might enjoy walking through the park, chatting to his friends, etc. while he is in the process of looking for his grandchildren, who are playing).
Note that the correct option is now in the third position. C. it is important to vary its position. Note also that the word looking appears in each option: in some tests the  item might appear as follows:

      Old Mr. Jones enjoys looking ……the children playing.
            A. -                   B. on                C. at                 D. to

However, when this format includes a dash (-), it is unnatural and not recommended since the insertion of a dash in the stem would not be normal practice in real life.

Item  3            
Error:   ‘I suppose that you were not angry to me.’
Item:   I do hope you were not angry …..me.
            A. to                 B. with             C. on                D. about

Note that at is also incorrect and may be used as a possible distractor. On the other hand, it may be felt that a number of native English speakers do say angry at a person. The decision whether or not to include at in the list of incorrect option is again a very subjective one.


CONSTRUCTING ERROR RECOGNITION MULTIPLE – CHOICE ITEMS

The sentence ’Sun is shining, trees become green and ….. The error caused by the omission of the article may be tested as follows, using a multiple-choice item:

       …….is shining brightly today.
     A. Sun         B. The sun        C. A sun           D. Some sun

     It may be argued, however, that the choice here is strictly between option A and B at certain levels where students have learned to avoid using ‘a’ and ‘some’ with ‘sun’. in such instances, one useful device (still using the multiple-choice format) is the error-recognition type of item.

Type 1
Each sentence contains four words or phrases underlined, marked A, B, C and D. Select the underlined word or phrase which is incorrect or unacceptable.

       1. I do hope you wouldn’t mind waiting for such a long time.
              A                        B                  C          D
       2. I’m worried that you’ll  be angry   to me
                      A                           B       C        D
      3. I didn’t see Bill since he went into hospital last month.
                  A                           B     C                  D
      4. My car had broken down, so I went there by foot.
                          A                   B      C           D
Type  2
There is a mistake in grammar in each of the following sentences. Write the letter of that part of the sentence in which it occurs.
            A          B                      C            D
      1. Sun/is shining/brightly today/, isn’t it?
            A                      B                      C                                  D
      2. Old Mr. Jones/enjoys/looking the children/playing in the park.
            A                      B                                  C          D
      3. Tony’s father/would not let him/to stay out/late at night.
              A            B                              C                                              D
      4. Didn’t/Susan tell you/she wouldn’t mind to come/with us on the picnic?

Item type 2 allows the test writer to test errors caused by omission: e.g. Sun is shining and looking the children. This type of error cannot be tested by the first item of the error-recognition type. However, there are different ways of correcting many sentences. For example, students may write B or C to denote the incorrect part of the third sentence above, according to which of these correct versions is in their mind:

       Tony’s father would not permit him to stay out late. (=B)
       Tony’s father would not let him stay out late. (=C)

For this reason, the test writer is strongly advised to avoid items of the second type.


CONSTRUCTING REARRANGEMENT ITEMS

Rearrangement items can take several forms, the first of which to consider here will be the multiple-choice type.
    Here are two of the errors he made:

       ‘You know how is it.’
       ‘I wonder did you grow more fatter since summer.’

     If we attempt to test the first error by means of an ordinary multiple-choice item, we are faced with the problem of being restricted to only two options: the correct option and the distractor (i.e the error).

       You know how …….
       A. it is                    B. is it

     As the item stands here, we cannot possibly construct other options. It becomes necessary, therefore, is lengthen the original statement to: You know how warm it is today. The item would then read:

       ‘Won’t I need a coat?’
       ‘Well, you know how……..’
A.      warm is it today
B.      today is it warm
C.      is it warm today
D.     warm it is today
E.      today is it warm

     There seems to be a danger here of confusing the testees by presenting them with the problem in such a way that a certain amount of mental juggling becomes necessary on their part. A preferable item type is the following word-order item:

Complete each sentence by putting the words below it in the right order.
Put in boxes only the letters of the words.

‘Won’t I need a coat?’
‘Well, you know how …….’
A. it         B. today            C. warm     E. is                       

I wonder if ……since summer.
A. grown     B. fatter   C. you        E. have

Word order items are useful for testing other structures and features involving inversion:

Everyone’s forgotten …..
A. cup                 B. he                 C. which           D. used
Not only ………but he took me to his house.
A. me                  B. he                 C. did               D. meet
However…….you’ll never pass that test    
A. you                 B. try                C. hard             D. may

   The order of adjectives and the position of adverbs can be tested in this way, as indeed can several other grammatical areas

The police are looking for ……..
A. big                  B. two               C. cars              D. black
Would you like to read David Brown’s …..?
A. short               B. new              C. story D. exciting
Tom said …………..cleaning his car.
A. had                 B. finished        C. he                D. just


CONSTRUCTING COMPLETION ITEMS

Carefully constructed completion items are a useful means of testing a student’s ability to produce acceptable and appropriate forms of language. They are frequently preferable to multiple-choice items since they measure production rather than recognition, testing the ability to insert the most appropriate words in selected blanks in sentences. The word selected for omission are grammatical or functional words (e.g. to, it, in, is, she): content words may be selected in a vocabulary or reading test.
   The error Sun is shining, although only one answer is possible here, this completion item would have to appear as:

            ………. Sun is shining today.
or as:    ………. Sun is shining today.
The former item suggests to the testees that no determiner is necessary (since Sun is written with capital letter) while the latter item suggests that a determiner is necessary (because sun is written without a capital).
            The main item can be simply be rewritten as a question to overcome its problem:

       Is …….sun shining today?

Here are two more examples of completion items based on the student’s letter:

Write the correct word in each blank. 
1. The old man enjoys looking ……..the children playing.
2. That car belongs ……….. Helen’s mother.
3. I hope you are not angry ……me.

Put a, the, or some in each blank only where necessary. If you think that  no word should be placed in the blank, put a cross (x) there.

1. Can you see ……sun shining through the clouds?
2. I saw your uncle …..day before yesterday.
3. What have you been doing since I saw you ……last summer?

Completion items cannot, of course, be machine-marked but they are very useful for inclusion in classroom tests and for exercise purposes. The following example indicates the wide range of possibilities for one completion item:

       I go to cinema regularly, but I …… to the theatre for months.

            The answer obviously required by the tester is haven’t been; however, possible answers are :
haven’t been                                            shan’t be going
hadn’t been                                              can’t go
(sometimes) don’t go                                haven’t been able to go
may not go                                              am not going
don’t know whether I’ve been                   didn’t go
shan’t go                                                 haven’t gone
won’t go                                                  haven’t been going

If the aim of this particular item is to force the use of the present perfect tense, there are three ways of restricting the choice available to the testees (although the first two ways depend heavily on reading comprehension):

(a) by providing a context:
Kim usually goes to the cinema about once a week but she …….four films already this month and it’s only the 20th today. (possible answers: has seen/will have seen/must have seen)

(b) by providing data:
I go to the cinema regularly, but it’s ages since I last saw a play.
I go to the cinema regularly, but I ……….to the theatre for months.
(Possible answers: haven’t been/haven’t gone/haven’t been going/haven’t been able to go)

(c) by using multiple-choice techniques:
I ……to the theater three times since I last saw you.
A. go                                 C. had gone
B. have been                     D. went

There are two major advantages in using a passage of continuous prose rather than separate sentences when giving a completion type test. Firstly, the use of context often avoids the kinds of ambiguity referred to in the previous paragraphs. Secondly, the students experience the use of grammar in context, being required to use all the context clues available in order to guess many of the missing words.


CONSTRUCTING TRANSFORMATION ITEMS

The transformation type of item is extremely useful for testing ability to produce structures in the target language and helps to provide  a balance when included in tests containing multiple-choice items. It is the one objective item type which comes closest to measuring some of the skills tested in composition writing, although transforming sentences is different from producing sentences.
   The following transformation items have been based on errors which occurred in the student’s letter.

Rewrite each of the following sentences in another way, beginning each new sentence with the words given. Make any changes that are necessary but do not change the general meaning of the sentence.

1. I haven’t written you for a long time.
    It’s a long time ………………………
2. In sunny weather I often go for a walk.
    When the weather ……………………
3. Old Mr. Jones likes to look at the children playing.
    Old Mr. Jones enjoys …………………………….

   Other transformation items giving some idea of the range of areas that can be covered in this way are:

1. It was impossible to work under those conditions.
    Working ………………………………………….
2. I don’t think it’s necessary for you to stay any longer.
    I don’t think you …………………………………..
3. I was able to leave the office early yesterday.
 ………………………………………………….

            The following examples illustrate how each of the sentences for transformation can be made to form part of a continuous sentence.

1. Changing sentences according to a given pattern
(a) Very few objective tests allow for choice.
     You have ………………………………………
(b) However, the instruction should be carefully checked.
     However, you ……………………………………….
(c) Different types of questions on the same paper will necessitate changes in the instructions.
     The instructions ……………………………………………………………..

2. Changing sentences by using selected words
(a) Remember that it is not necessary to answer the questions in the order set (NEED)
          …………………………………………………………………………………
(b) You are advised to check your answers carefully after each question. (ADVISABLE)
    ……………………………………………………………………………….
(C) Most teachers also recommend you to leave five minutes spare at the end of the examination in order to check your paper. (SUGGEST)
  ……..………………………………………………………………………….


CONSTRUCTING ITEMS INVOLVING THE CHANGING OF WORDS

This type of item is useful for testing the student’s ability to use correct tenses and verb forms. It is a traditional type of test but the layout is improved in this particular case by providing blanks on the right of the text for completion. The continuity of the text is not impaired more than necessary by having both blanks and underlined words inserted in the sentences. Thus the risk of obscuring the meaning of the text is reduced.

1.   Verbs; tenses; etc.
Researchers (1) to convince that a drug                                                 (1) ……………
they (2) to test can improve the memory and that                           (2) ……………
it (3) to be the forerunner of other drugs which                                     (3) ……………
eventually (4) to improve mental ability.                                               (4) ……………

2.   Word building
Students who were given the drug for a fortnight did
considerably (1. well) in tests than others. The tests                  (1) ……………
included the (2. memorize) of lists of words as well                  (2) ……………
as of (3. inform) from two messages transmitted at the             (3) ……………
same time. During the first week there as no (4. notice)            (4) ……………
difference between the two groups, but after a fortnight
the group of the drug was found to have increased
   its (5. able) to learn by almost twenty per cent.                       (5) ……………


CONSTRUCTING ‘BROKEN SENTENCE’ ITEMS

This type of item tests the student’s ability to write full sentences from a series of words and phrases, and thus not allow the test writer to concentrate exclusively on testing those particular grammatical features which may have just been practiced in class. It is nevertheless a useful device for testing grammar provided that the tester is aware that several other areas of the language are being tested in addition to those on which he or she wishes to focus attention.
     In this type of test item, students should be instructed to make whatever changes are necessary to form good sentences, adding articles, prepositions, etc. where required and putting verbs in their correct tense.


Take / drug and stimulants / keep awake / while revise examination / often be very harmful / it be far better / lead / balanced life / and get enough sleep / every night. / There / be / limit / degree and span / concentration / which you be capable/ exert. / Brain / need rest / as much body. / Indeed, / it be quality / than quantity work / that be important.


CONSTRUCTING PAIRING AND MATCHING ITEMS

This type of item usually consists of a short conversation: e.g. a stimulus in the form of a statement or question followed by a response often in the form of a statement. It is used to test the ability to select appropriate responses to stimuli which would be presented orally in normal everyday situations. The item is more useful for testing students’ sensitivity to appropriacy and their awareness of the functions of language rather than their knowledge of grammar (although grammatical clues may prove important in completing this item satisfactorily). To perform the task required, students are simply required to write the letter of the correct response in the space provided.

Column 1                                                         Letter Column 2
Doing to see a film tonight?                               …….    A. No, I didn’t.
How was the film?                                             …….    B. Most are, I think.
I can’t stand war film, can you?                           …….    C. It’s one of the reasons.
So you went to the cinema.                                 …….    D. I had a lot of work to do.
Don’t you find war films too violent?                 …….    E. Actually, I quite like them.
Have you ever seen a Japanese war film?             …….    F. Yes, I probably will.
I like war films.                                                  …….    G. No, I haven’t.
Is everyone going to see the film?                       …….    H. What a good idea! I prefer them to war
     films.
What about going to see a cowboy film instead?  …….    I.  So do I.
Why didn’t you come with us to see the film?      …….    J. All right. Nothing special.
Is that why you don’t like war films?                   …….    K. Not really, I quite like them.


It should be remembered, of course, that this is not authentic discourse. However, although the language and situation here are inevitably artificial, the item does serve to help students to associate the language they are learning with real-life situations, albeit to a limited extent.


CONSTRUCTING COMBINATION AND ADDITIONAL ITEMS

These objective-type items have long been used in past tests. They should be used sparingly, however, as they involve largely mechanical responses on the part of the student. Note that although the separate sentences are linked to one another by theme,   the items can hardly be described as being contextualized in any real way.

1.   Combination items
(Students are instructed to join each pair of sentences using the word in brackets)
(a)    You finish the paper then check your answers carefully. (AFTER)
(b)   Some questions may be very difficult. They should be left until later. (WHICH)
(c)    You should usually write answers in complete  sentences for all the questions on your paper. However, write notes for those questions which you do not have time to answer. (ALTHOUGH)

2.   Addition items
(Students are instructed to insert the word in capitals in the most appropriate place in each sentence.)
(a) YET                            Have you answered all the questions?
(b) STILL                         Some students had not mastered the correct techniques  for answering examination questions.
(c) OCCASIONALLY    There may be little choice of questions.

Add to Google Reader or Homepage

0 komentar:

Posting Komentar

Related Posts Plugin for WordPress, Blogger...

Tags

Blog Archive