000028779 READING COMPREHENSION TESTS FOR INTERMEDIATE STUDENTS AT HANOI UNIVERSITY OF FOREIGN STUDIES BÀI KIỂM TRA ĐỌC HIỂU DÀNH CHO SINH VIÊN TRUNG CẤP TẠI TRƯỜNG ĐẠI HỌC NGOẠI NGỮ HÀ NỘI

000028779 READING COMPREHENSION TESTS FOR INTERMEDIATE STUDENTS AT HANOI UNIVERSITY OF FOREIGN STUDIES BÀI KIỂM TRA ĐỌC HIỂU DÀNH CHO SINH VIÊN TRUNG CẤP TẠI TRƯỜNG ĐẠI HỌC NGOẠI NGỮ HÀ NỘI

Certificate of Originality of Study Project Report

I certify my authorship of the study project report submitted today - entitle:

Reading Comprehension Tests for Intermediate students at Hanoi University of Foreign Studies.

In terms of the Statement of Requirements for thesis and field study reports in Masters’ programmes issued by the Higher Degree Committee.

(signature of author)August, 1999

I am particularly indebted to my supervisor Mr Le Huy Truong, M.A for his valuable instructions, comments, criticism and tor his kind encouragement during the development of this field study report

W ithout his help this field study report would have been impossible

I also acknowledge my gratitude to Mr Tran Quang Huy, M.A of the Post-Graduate Department of HUFS (the Hanoi University of Foreign Studies), Mr Dang Dinh Thien, M.E.D and Mrs Luu Phuong Lan, G Dip for their helpful comments and suggestions

My appreciation and gratitude are also extended to Mrs Nguyen Thi Minh Thuan, lecturer of HUFS, who did all of the type-setting of the manuscript

I also would like to thank my collègues in the English Department of

H UFS for the support and encouragement

Finally, my thank goes to the group of intermediate students of classes FA 5-98, FA6-98 and FA12-98 of the English Department for their cooperation and test-trialling

A BSTRA CT

At present, in Vietnam, especially in the English Department of the Hanoi University of Foreign Studies, a well-designed test of reading comprehension is an urgent demand Little attention has been given to this problem This field study report is aimed at providing some guidance

on the design of reading comprehension tests for intermediate level students at the Hanoi University of Foreign Studies This field study report consists of five chapters as follows:

Chapter one is a brief introduction and an overview of language testing in Hanoi University of Foreign Studies The aims of this field study report are also given in this chapter

Chapter two covers questions such as role of testing, main characteristics of a good test and types of language tests It also presents

a formula to calculate the reliability and validity of a test

Chapter three begins with the characteristics of reading tests, the role of reading skills in Headway Upper-Intermediate (a textbook used for second year students of the English Department) and is followed by the discussion of the designing and trialling of test items for reading comprehension tests

Chapter four deals with test improvements, some tips of designing a good test are also given The chapter ends with formats of reading comprehension test for intermediate students at the English Department

o f Hanoi University of Foreign Studies

Chapter five presents the conclusion of the field study report

C H A PTER III TESTIN G TECH N IQ UES FO R READING

COM PREHENSION AND HOW T O IDENTIFY

3.4.1 Identification of Test Needs 38

C H A PTER IV PROPOSED IM PROVEM EN TS OF CU RREN T

READING T E STS FO R TH E ENGLISH

APPEN DIX STUDEN TS AND TH EIR T E S T SCORES

BIBLIOGRAPH Y

Chapter One

IN TRO D U C TIO N

Testing is an important part of teaching - learning process We know very well that a well-made test can motivate the students’ learning and improve the teachers’ teaching as well T o obtain the above mentioned goal, of course I think, it is necessary for the English Department of Hanoi University of Foreign Studies to solve a number of problems concerning testing, especially testing for the students at the intermediate level in terms of reading comprehension such as:

First, there is a gap between what is taught and what is tested For example, tests designed for proficiency level are given to students of intermediate level Such tests are not valid and reliable at all They are

so difficult that only few students can accomplish

Second, there is a tendency of using tests exclusively for grading, there is no feedback about the tests

Third, there is no discarding of bad tests or bad items Some items are so difficult that most of the testees give the wrong answers or items are too easy so that all testees can obtain the correct answers Such items should be discarded or replaced

1

Fourth, there is the problems of using imported tests which may lack face validity For example, a reading comprehension test containing such words as ‘buck’, ‘chook’ and ‘prince of darkness’ will obviously not be valid in Vietnam no matter how valid and useful a test it has proved in Australia That is because of the different culture.

And the last problem is that some of the testees may lack reliability because they are not pre-tested anywhere else for the sake of confidentiality Ironically, for the sake of ‘confidentiality’ test designers are often informed to design the test at short notice, just a day or two before it is administered The test designers do not even know who the testees will be, what level of knowledge and competence they have In such circumstances who can say for sure that the required standards, criteria will be met by the testees

Because of the problems already mentioned, I would like to focus my dissertation on the research of testing, offering a number of suggestions improving the reading test for the students of the English Department at the intermediate level in terms of theory and practice as well

The students for whom the reading test is intended are studying English in the first year and second year at the English Department of the Hanoi University of Foreign Studies They are provided with the

2

four integrated skills of general English: listening, speaking, writing and reading.

In the middle of each academic year the students have to take a test, the mid-term test and at the end of the academic year they have to take the achievement test These tests often include speaking test, listening test, reading test and writing test

In the reading test, the candidate should be able to

scan the text to locate specific information;

skim the text to obtain the gist;

identify stages of an argument;

guess vocabulary from the context text;

distinguish between relevant and irrelevant information;

predict information based on given information;

complete ideas based on previous background;

identify functions of cohesion and coherence;

identify key words;

form hypothesis about the text or message

Presently in Vietnam, general language testing has the following characteristics:

It has not been given adequate attention and careful study

Its role in teaching and learning nas not been fully recognised

Almost language teachers hold the idea that anyone who can teach should be responsible for testing, because testing is the only part of teaching and learning process However, they are not professional test designers.

There is no cooperation among teachers about tests

Using ready-made tests rather than self-made tests because they are very convenient and time-saving for teachers The disadvantage is that they are not relevant to the purposes of our course

Test contents and test formats are different from institution to institution because they are decided by the Examination Board belonging to each institution Test content is sometimes found

to be unrelated to the objectives of the course and very often items included in the test have not been covered in class

In the next chapter I will discuss the various solutions to the above mentioned problems by setting the theoretical context with an overview

of English language testing and characteristics

Chapter Two

ASPECTS OF TESTING

2.1 W H A T IS TESTING?

T o define the term “testing”, I would like to deal with its meaning in

a number of looks Madsen (1983) and Harrison (1983) give the definition of “testing” They also describe different kinds of tests in their books Le Huy Truong (1988) also denotes: “Testing could be understood as the use of any means of tasks that are designed to focus on

a particular area of learning.” In “Modern Language Testing”, Valette (1977) deals with such questions as preparing the classroom test, giving the classroom test, evaluating classroom test results Furthermore,

“testing” could be understood as a process of assessment consisting of five stages

2.2 W H A T IS A TEST?

A t FAAPI seminar, Mardel plata, Argentina, October 1995, Allan,

D paper wrote:

“Tests are instruments of evaluation, one of the ways in which we try

to measure participant performance They have a physical existence and operate within specific time frame, seeking to make accurate predictions

on the basic of relatively small samples of performance in the case of such an enormously complex thing as language.”

There is no the so called ‘the best test’ or ‘the best testing techniques’ for the purposes of testing

I will give a further discussion of test characteristics in the next part

“ \

of this chapter

2.3 ROLE OF TESTING

2.3 1 The Teaching - Testing Link

It is clear that teaching and testing should be linked Heaton

(1985 : 5) points out: “A large number of examinations in the past have

encouraged a tendency to separate testing from teaching Both testing and teaching are also closely interrelated that it is virtually impossible

to work in either field without being constantly concerned with the other.”

6

It a test is regarded as important, then preparation for it can come to dominate all teaching and learning activities Davies (1968 : 5) has said that: “The good test is an obedient servant since it follows and apes the teaching.”

I find it difficult to agree The proper relationship between teaching and testing is surely that of partnership Heaton (1988 : 5) goes on with his idea : “Tests may be constructed primarily as devices to reinforce learning and to motivate the student’s performance in the language.”The test is geared to the teaching that has taken place, whereas teaching is often geared largely to the test The T E FL News letter (1977 : 15) suggests: “W hen a test is administered to new students, the results can indicate the general level of proficiency of a class and the initial level at which a teaching programme should aim A test at the end of a teaching programme can indicate the extent to which the student has mastered what has been taught.”

The role of teaching - testing can be summed up as follows: Teaching

helps teachers to locate the precise area of difficulty encountered

by the whole class or by individual student;

helps teachers to diagnose students’ weaknesses, difficulties andstrengths;

7

enables teachers to increase their own effectiveness by making adjustment in their teaching;

gives teachers an opportunity to identify the errors and reasons why a student makes them, so that they can provide further assistance;

enables teachers to recognise which parts of the language programme have been found difficult by the class;

helps teachers to evaluate the effectiveness of the syllabus, methods and materials they are using

2.3.2 Testing - Learning Link

In regard to students and testing, Heaton (1988 : 7) also points out that “A well-constructed classroom test will provide the students with

an opportunity to show their ability to perform certain tasks in the language.”

and properly administered test can reinforce the students’ learning processes.”

Thus, we can see that the students can benefit from the test because: They are motivated to display what they have learned

The feedback they get from the results may add to their motivation

They are made aware of their errors and reasons for making them, so they will be able to avoid them in the future.

2 4 TYPES OF LANGUAGE TESTS

Test classification varies slightly from writer to writer For example, Valette (1977 : 5) divides tests into four basic types: aptitude, progress, achievement and proficiency Heaton (1988 : 171, 172, 173) class progress tests, achievement tests, proficiency tests, aptitude tests, diagnosis tests

Language tests have different purposes, that is, the information obtained from test is used for different types of decisions Some results help decision makers determine whether the student will be accepted to

a certain language programme, others help determine if the student will need special help, others if a student will be allowed'to continue with his language studies - and yet others - to examine if the student has mastered the material - so the teacher can start teaching the new material The type of information, which the tester needs for making such decisions, will determine to a large extent the type of test, which the tester will use

Since the test is a sample of the language knowledge, the language knowledge varies according to the purpose of the test

9

Madsen (1983 : 8) divides tests into seven contrasting pairs as

follows:

Knowledge tests and performance tests;

Subjective tests and objective tests;

Productive tests and receptive tests;

Language sub-skill tests and communicative tests;

Norm-referenced tests and criterion-referenced tests;

Discrete-point tests and integrative tests;

Each type of test has its own characteristics and I will now give a brief description of each type

2.4.1 Progress Tests

Progress tests are used to measure how much the students have learned in a specific course and to give classroom teachers a chance to

unique and can only be evaluated fully by the class teacher in the light

of his or her knowledge of the students, the programme which they have been following, and the class teacher’s own particular aims and goals

measure the extent to which the students have mastered the material

taught in the classroom It is based on the language programme which the class has been following and is just as important as an assessment of the teacher’s own work as the student’s own learning.”

So it can clearly be seen that the formats of progress tests often reflect the various components of the curriculum Progress tests are usually given to provide student motivation They also help teachers find out areas of weakness or difficulty

2.4 2 Achievement Tests

Achievement tests, as their name suggests are designed to establish how successful individual student, groups of students, or the courses themselves have been in achieving objectives Final achievement tests are administered at the end of a course of study They may be written and administered by Ministry of Education, official examining boards or

by members o f teaching institutions

Harrison (1983 : 7) defines achievement tests as follows:

“An achievement test (also called an attainment or summative test) looks back over a longer period of learning than the diagnosis test, for example, a year’s work, or the whole course, or even a variety of different courses.”

T o some extent, achievement tests are similar to the class progress tests because they are generally based on a syllabus and they are

designed to measure what has been taught and learned However, achievement tests look back over a longer period of learning than the progress tests.

In Vietnam, annual school examinations, school leaving tests and University Graduation Examinations are tests of this kind

2.4.3 Diagnostic Tests

According to Arthur Hughes (1988 : 13) “Diagnostic tests are used

to identify students’ strengths and weaknesses They are intended primarily to ascertain what further teaching is necessary.”

From this definition we can see that the main purposes of this kind

of tests are to identify the test-taker’s strengths and weaknesses in the language, as well as to attempt to explain why certain problems occur, and what “treatment” can be assigned

Commenting on this test Heaton (1988 : 173) states: “Although the term diagnostic test is widely used, few tests are constructed solely

as diagnostic tests Achievement and proficiency tests, however, are frequently used for diagnostic purposes: areas of difficulty are diagnosed in such tests so that appropriate remedial action can be taken later.”

We should note that this type of test is frequently carried out for groups of students rather than for individuals If only one or two

students make a particular error, the teacher will not pay too much attention However, if several students in the group make a certain error, the teacher will note the error and plan appropriate remedial teaching.

2 4 4 Aptitude Tests

Valette (1977 : 5) gives the definition of this type of test as follows:

“The aptitude test is conceived as a prognostic measure that indicates whether a student is likely to learn a second language readily It is generally given before the student begins language study, and may be used to select students for language course or to place students in

From Valette’s point of view it can be seen clearly that aptitude tests are designed to measure the students’ probable performance in learning

a foreign language, indicating whether the student has any special aptitude for learning a new language

Heaton (1988 : 73) also denotes: “Language learning aptitude is

a complex matter, consisting of such factors as intelligence, age, motivation, memory, phonological sensitivity and sensitivity to grammatical patterning.”

A t the Hanoi University of Foreign Studies, this type of test is given before students begin language study, and could be used to select

students for a certain language course It is also used to place students in classes appropriate to their ability In recent years, several aptitude tests have been developed to select talented candidates for its entrance exams

to various specialised courses such as music, painting, industrial design, architecture and theatre

2.4.5 Proficiency Tests

Heaton (1988 : 172) remarks on proficiency tests as follows:

“The proficiency test looks forward, defining a student’s language proficiency with reference to a particular task which he or she will be required to perform Proficiency tests are in no way related to any syllabus or teaching programme; indeed, many proficiency tests are intended for students from several different schools, countries and even language backgrounds The proficiency test is concerned simply with measuring a student’s control of the language in the light of what he or she will be expected to do with it in the future performance of a particular task.” Other theorists such as Harris (1983), Carroll and Hall (1985), Harrison (1986) also agree that the aim of a proficiency test is

to assess the student’s ability to apply in actual situation what he has learned For example, does a student know enough English to follow a particular course given in the medium of English; or to carry out a specific task requiring the use of English? This type of test is not related

to any particular course or syllabus because it is concerned with the student’s current standing in relation to his potential performance and future needs.

In Vietnam now, this type of test has been used to grant certificate level C for scientific research workers, engineers, teachers, and architects of various branches who pass this test This test is given by the Ministry of Education and Training every quarter of the year

2.5 W H A T MAKE A GOOD TEST?

According to Harrison (1983) and other testing experts, such as Valette (1977), Carroll and Hall (1985), Heaton (1988) and Arthur Hughes (1989), all good tests possess four main characteristics:

Reliability is a necessary characteristic of any good test It is said that

no test is valid if it is unreliable Magoon and Garrison (1976 : 371)

15

point out that reliability is essential to validity A test can be reliable without being valid but it can only be as valid as it is reliable.

Some language testing specialists (Harrison 1983, Harris 1988 and

H eaton 1988) point out that test reliability is affected by a number of factors:

The extent of the sample of material selected for testing whereas validity is concerned chiefly with the content of the sample, reliability is concerned with the size The larger the sample (i.e the more tasks the testees have to perform), the greater the probability that the test as a whole is reliable - hence the favouring of objective tests, which allow for a wide field to be covered

different groups under different conditions or at different time? Clearly, this is an important factor in deciding reliability,

comprehension To ascertain the reliability of a test, Carroll and Hall (1985 : 118) and Heaton (1988 : 163) suggest the three following methods: test-retest, internal consistency and split- half

In his Field Study Report, Le Huy Truong (1988 : 21) states: “This method aims at finding out the reliability index to see whether the testees are similarly ranked in two successive application of the same test.”

Harris (1969 : 15) also suggests: “The simplest technique would be

to retest the same individuals with the same test.”

From the above point of view it is clearly seen that the simplest method of measuring the reliability of a test is to re-administer the same test after a lapse of time It is assumed that all candidates have been treated in the same way in the interval - that they have either all been taught or that none of them have Provided that such assumptions (which are frequently hard to justify) can be made, comparison of the two results would then show how reliable the test has proved

Heaton (1988 : 163) comments that: “Clearly, this method is often impracticable and, in any case, a frequent use of it is not to be recommended, since certain students will benefit more than others by a familiarity with the type and format of the test.”

2.5.1.1 T est-R etest

17

An example of a test which has been given twice to a class of fifteen students, the test-retest reliability (the rank-order correlation) can be calculated as follows:

•

Students (first, etc.) (first, etc.) (d) (d2)

The formula for working out the Rank-Order correlation coefficientis:

^ = i _ 6 x Total d~

n (n2 - 1)where:

Note: W hen testees have the same rank, take an average of theranks concerned and give each of them that average, e.g with two atrank 3, add 3 and 4 and divide by 2, i.e 7 - ^ 2 = 3.5 each

Figure 6: Rank order: Carroll and Hall (1985 : 119)

According to Carroll and Hall (1985 : 118): “Broadly speaking areliability index of 0.8 or above suggests fair test reliability.”

2 5 1 2 Internal Consistency

Another means of estimating the reliability of a test is internal consistency T hat is the extent to which the test questions are related to one another, and measure the same trait when given the mean and standard deviation (appendix A), the test internal consistency can be calculated by the following formula:

19

I Q _ n x s d~ - M (n - M)

s d ’ x (n -1)Where:

Thus, with the test of 40 items, a mean of 25.7, and a standard deviation of 6.0, the internal consistency coefficient can be calculated as

In Carroll and Hall’s opinion, this coefficient of 0.764 shows moderate but not marked internal consistency

The split-half method is yet another means of measuring test reliability This method estimates a different kind of reliability from that estimated by test, retest procedures The split-half is based on the principle that, if an accurate measuring instrument were broken into two equal parts, the measurements obtained with one part would be

follows:

I C = 40 x 62 - 25.7 ( 4 0 - 25.7)

62 x (4 0 -1 )

1 4 4 0 - 3 6 7 51 1404

^ 4 9 = a ? 6 4 1404

2 5 1 3 Split-Half

20

corresponding exactly to those obtained with the other The test is divided into two and the corresponding scores obtained, the extent to which they correlate with each other governing the reliability of the test

as a whole One procedure widely used is to ascertain the correlation between the scores on the odd numbered items and those on the even numbered items A more accurate procedure is to balance the items as follows:

However, it would be better to allow chance to decide which items

go into one half and which go into the other

The reliability of the whole test can be estimated by using the following formula:

m (N - m)NX2Where:

(Note that in this formula, X has to be squared.) For example, a test

is administered to 26 students on a particular test consisting of 40 items The mean is found to be 27 and the standard deviation 4.077 using these figures with the above formula, we obtain:

According to Fleaton (1988 : 164) a correlation of 0.484 indicates a strong relationship between the two halves of a test and therefore good test reliability

Briefly, the validity of a test is the extent to which it measures what

it is supposed to measure and nothing else In other words validity of a test is the question of whether the test meets the requirements of testing

or not There are different kinds of validity such as face validity, content validity, construct validity and empirical validity, but among those only two are vital for any test: face validity and content validity

2.5.2.1 Face Validity

Heaton (1988 : 159) points out that “If a test item looks right to other testers, teachers, moderators and testees, it can be described as having at least face validity.”

40 x 16.662

27 x 13

rn = 0.484

2.5.2 Validity

Hughes (1989 : 27) states: “A test is said to have face validity if it looks as if it measures what is supposed to measure.”

For example, a test which pretended to measure pronunciation ability but which did not require the student to speak might be thought

to lack face validity

2 5 2 2 Content validity

Harrison (1983 : 11) gives a definition of content validity as follows:

“Content validity is concerned with what goes into the test The content

of a test should be described by considering the purposes of the assessment, and then drawing up a list known as a content specification.”

Hughes (1989 : 22) gives a similar definition: “A test is said to have content validity if its content constitutes a representative sample of the language skills, structures, etc with which it is meant to be concerned.” From the above definitions we obviously see that the contents of a test are the main factors in achieving its validity Heaton (1988 : 160) gives an advice: “W hen embarking on the construction of a test, the test writer should first draw up a table of test specifications, describing in very clear and precise items the particular language skills and areas to be included in the test.”

23

A test must be practical, in other words, it must be fairly straightforward to administer It is only too easy to become so absorbed

in the actual construction of test items that the most obvious practical considerations concerning the test are overlooked Harrison (1983 : 12) states: “The main questions of practicality are administrative A test must be well organised in advance.”

It means the length of time spent on the test should be decided Equipment needed for the test, the financial expenditure for the test should be taken into account

In short, as far as practicality is concerned, tests should be as economical as possible in time and cost

Other practical considerations are:

The test paper should be printed or typewritten

Test instruction should be clear

Testees should write their answers on separate answer sheets, so that the test paper can be used several times

Markers should be given the answer sheets in order to maintain the speed of scoring

2.5.3 Practicality

24

In Harrison’s opinion (1983 : 11) “All language assessment is based

on comparison and discrimination.”

This is true for both teacher-made tests and standardised tests A good language test should be able to discriminate between a student and others taking the same test

Heaton (1988 : 165) also points out: “Sometimes an important feature of a test is its capacity to discriminate among the different candidates and to reflect the differences in the performances of the individuals in the group.”

According to Heaton (1988), 70% means nothing at all unless all the other scores obtained in the test are known Furthermore, tests on which almost all the candidates score 70% clearly fail to discriminate between the various students If the test is too easy or too difficult, it will not show any discrimination between students For example, with a reading test, student A is weak at reading skill but he gets 39 correct items out of 40 items of the test, then the test is evidently too easy It fails to discriminate between student A and others in his class Another example would be if in a reading test of 40 items, only 5 students out of

30 including the best ones, get just 4 items correct This test is said to be2.5.4 Discrimination

25

i

too difficult as it also fails to discriminate between the bright students and the slow ones.

Here we come to the question of “item difficulty” and “item discrimination”

2.5.4.1 Item difficulty

Difficulty level relates to how easy or difficult the item is from the point of view of the students who took the test It is important since test items which are too easy (that all students get right) can tell us nothing about differences within the test population

According to Heaton (1988 : 178) “The index of difficulty (or facility value) of an item simply sho.vs how easy or difficult the particular item proved in the test.”

Heaton uses the following formula to calculate the index of difficulty(FV)

Where:

26

Thus, if 21 out of students tested obtained the correct answer for one of the items, the index of difficulty can be calculated as follows:

26Heaton, (1988 : 179) comments: “The particular items is fairly easy one since 77 per cent of the students taking the test answered it correctly Although an average facility value of 5 or 50 per cent may be desirable for many public achievement tests and for a few progress tests, the vacility value of a large number o f individual items will vary considerably between 0.3 and 0 7 ”

Heaton also gives a formula to calculate the discrimination index (D) as follows:

n

27

who got the item correct, and divide by the number of students in either

T t

group (not the total number of students but half the number)

The following item, which was taken from a test administered to 40 students, produced the results shown:

(the upper half contains 15 correct answers on a particular item, the lower: 6)

Commenting on D, Heaton (1988 : 180) states: “Such an item with

a discrimination index of 0.45 function fairly effectively, although clearly

it does not discriminate as well as an item with an index of 0.6 or 0.7.”

28

Discrimination indices can range from + 1 ( = an item which discriminates perfectly) though 0 ( = an item which does not discriminate in any way at all) to -1 ( = an item which discriminates in entirely the wrong way).

and in the next chapter I will mention a particular area in testing, that is reading comprehension test

29

is happening in the mind of the testee while the latter is comprehending

a text The tester taps comprehension indirectly This is in contrast to

testing the productive skills of writing and speaking where the tester can obtain an actual language sample from the testee by asking him to produce language through writing or speaking Traits which are evaluated indirectly put a heavier burden on the testing procedures which the tester decides to use and may have some effect on the score of

the testee - since we know that each testing method has some effect on the performance of the testee on the test, the tester, who is tapping comprehension attempts to use a number of testing methods.

In testing comprehension we generally impose an external factor on the text which the testee needs to comprehend This task is generally in

a form of questions or other tasks which the students are expected to respond basing on the text It is likely that this external factor has an effect and tends to “contaminate” the comprehension of the text

How do testing methods contaminate the measurement?

Most testing methods are imposed externally on the text and thus may add another task in addition to the comprehension of the text The four following types are the examples of this kind:

a CLOZE: most natural, no questions added but there is a question regarding the authenticity of the task

b M ULTIPLE CH O ICE: Task: Selection Questions and distracters

are an additional requirement to comprehension Comprehending the text and content of the question and the distracter The questions may not be a reflection of the text comprehension

Production The language and content of the questions may be

unfamiliar to testee The questions may be difficult to

31

understand It may be measuring writing and not only comprehension The questions selected may not be necessarily a reflection of the comprehension of the test.

d SUMMARY: Task: Comprehension and production of a whole text The method may impose an extra requirement, that is the

ability to write and produce full texts or for an oral report Ability

to express oneself in speaking

3.2 STRATEGIES OF TESTING READING COMPREHENSION

It is believed that comprehension is enhanced by using reading

comprehension strategies such as ability to:

• Recognise words and word groups, associate, sounds with their corresponding graphic symbols;

• Deduce the meaning of words by

( b ) contextual clues;

• Understand explicitly started information;

• Understand relations within the sentence, especially:

( a ) elements of sentence structure

( b ) negation

( c ) fronting the theme

( d ) complex embedding;

32

• Understand relations between parts of a text through both lexical devices (e.g.: repetition, synonyms, antithesis) and grammatical cohesive devices, especially anaphoric and cataphoric reference (e.g.: he, they, it, the former, the latter) and connectives;

Perceive temporal and spatial relationships, and also sequences of ideas;

Understand conceptual meaning, especially

( a ) quantity and amount

( b ) definiteness and indefiniteness

( c ) comparison and degree

( d ) means and instrument

( e ) cause, result, purpose, reason, condition, addition, contrast,

concession;

Anticipate and predict what will come next in the text;

Identify the main idea and other salient features in a text;

• Generalise and draw conclusions;

• Understand information not explicitly started by:

( a ) making inferences (i.e reading between the lines)

( b ) understanding figurative language;

Skim and scan (looking for specific information) ;

Read critically;

33