Sei sulla pagina 1di 112

Prof. Dr.

Dinay Kksal anakkale Onsekiz Mart University ELT Department 2011

LANGUAGETEACHINGLEARNINGANDTESTING

Heaton(1990:5)emphasizesthattherehasbeena tendencytoseparatetestingfromteachinginalarge numberofexaminationsinthepastandthatitis impossibletoworkineitherfieldwithoutbeingconstantly concernedwiththeotherbecausebothteachingand testingarecloselyinterrelated. Wemustcontructtestsasdevicestoreinforcelearning andmotivatethelearners.Wealsomustconstructthem asameansofassessingthelearnersperformanceinthe targetlanguage.Hestatesthatintheformertestingis gearedtoteachingwhileinthelatterthecaseisthe reverse.
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

LANGUAGE TEACHING LEARNING AND TESTING


Testing is an important part of the learning and teaching process. Madsen(1983:3-4), discussing how testing helps learners learning English, stresses that well-made English tests help learners at least in two ways: 1. create positive attitudes toward instruction and classroom experiences by giving them a sense of success and a feeling that teachers evaluation of them matches what they have been taught.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 3

LANGUAGE TEACHING LEARNING AND TESTING


When discussing prejudices and problems concerning

assessment, Haris and McCan (1994:2) stresses that assessment may cause many students to feel panic and confusion and that there is often great pressure on them to succeed: those who cannot succeed are they become branded as failures. Such a competition creates more losers than winners. Many of these negative attitudes towards assessment come from the generalised feeling of a divorce between learning and teaching on the one hand, and assessment on the other. The fundamental reason for this is that assessment often does not feed back into the learning and teaching process.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 4

LANGUAGE TEACHING LEARNING AND TESTING


2. Learn the language by requiring learners studying hard for

exams and raising a better awareness of course objectives and personal language needs.

Some teachers may feel that tests can be used as a way of motivating students to work harder and virtually all of us feel insecure and uncomfortable when we have to pass or fail students. However, tests are not the only way to motivate students.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 5

Reasons for testing


Let us look in detail at why we should test our students. Firstly,

whenever a test is administered, there is a decision to be made: Haris (1969:2-3) enlists the chief objectives of language testing as follows: 1. To determine readiness for instructional programmes. 2. To classify or place individuals in appropriate language classes. 3. To diagnise the individuals specific strengths and weaknesses 4. To measure aptitude for learning. 5. To measure the extent of student achievement of the instructional goals 6. To evaluate the effectiveness of instruction.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 6

Reasons for testing


Haris and McCan (1994:26), giving the reasons for testing, states

that we, language teachers, may want: to find out about a candidate's suitability to follow a course of study to compare a student's performance with that of other students. to find out how much a student has learned during the course or academic year ie compare what students can do at the end of the course compared with the beginning of the course. to find out how a student is progressing during a course of study and possibly identify problem areas before a course ends.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 7

Reasons for testing


In all these cases, we need to make decisions about the

students. We must have a clear idea of the kind of decision to be made about our students so that we can identify the most appropriate kind of test for you. Haris and McCan (1994:26) maintains that there is a common misconception, held by students and teachers alike, that a test is something which is done to or at people rather than something which is done by them and for them. We should view the decisions which are made about students as decisions which are taken for them. Formal testing should be seen as a complement to other forms of assessment, eg self-assessment and informal assessment.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 8

Reasons for testing


The basic differences are that if we have a well designed,

reliable and valid test, then the test will measure students' ability in a more objective way than more subjective forms of assessment such as informal observation and selfassessment. This is not to say that all formal testing is objective. Nor should we take the view that subjective is necessarily bad and that objective is necessarily good. For example all oral tests are subjectively marked and all multiple choice type tests are objectively marked.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 9

Within a practical, comprehensive philosophy of language

instruction and testing, every test becomes a natural rung in the ladder toward the instructional goalthat is, toward some desired degree of proficiency in the target language and every instructional activity in which students participate becomes a language-testing activity. In such a comprehensive theory, tests express the essence of the instructional process as well as, or perhaps better than, any other activity. In other words, teaching itself is a testing procedure as much as it is an instructional one.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 10

Tests in any classroom setting have a variety of functions

that we must understand before we choose and administer any test. Language tests in the classroom may serve the following purposes: Institutional. Good tests instruct students and enable them to improve their proficiency in the target language. Managerial. Such tests provide feedback to both teachers and students and help them manage instruction and studies practicesfor instance, by providing a sensible basis for grading. Motivational. The tests serve as rewards or as goals, urging students and teachers toward higher achievement relative to well-defined goals.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 11

Diagnostic. The tests help teachers and students identify

specific instructional problems. Curricular. Good tests define the curriculum as a whole.

Inappropriate language testing may result in, reflect, or even constitute ineffective language teaching. From such a pragmatic perspective, it might be argued that language testing is language instruction and, conversely, that language instruction is language testing.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 12

DEFINITION OF BASIC TERMS


Here it will be useful to define the confusing terms, which are often used synonymously (Bachman 1995:18): measurement, test and evaluation. MEASUREMENT He defines measurement as the process of quantifying the characteristics of persons according to explicit procedures and rules and states that this definition consists of three distinguishing features:

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 13

DEFINITION OF BASIC TERMS


(i)Quantification It involves the assigning of numbers, which distin guishes measures from qualitative descriptions such as verbal accounts or nonverbal, visual representations. Non-numerical categories or rankings such as letter grades ('A, B, C . ..'), or labels (for example, 'excellent, good, average . . .') may have the characteristics of measurement However, when we actually use categories or rankings such as these, we frequently assign numbers to them in order to analyze and interpret them, and technically, it is not until we do this that they constitute measurement.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 14

DEFINITION OF BASIC TERMS


(ii) Characteristics We can observe physical attributes such as height and

weight directly. In testing, however, we are almost always interested in quantifying mental attributes and abilities, sometimes called traits or constructs, which can only be observed indirectly. These mental attributes include characteristics such as aptitude, intelligence, motivation, field dependence/independence, attitude, native language, fluency in speaking, and achievement in reading comprehension. Whatever attributes or abilities we measure, it is important to understand that it is these attributes or abilities and not the persons themselves that we are measuring. That is, we must be aware of the fact that there is no a single measure or even a battery of measures that can adequately characterize individual human beings in all their complexity.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 15

DEFINITION OF BASIC TERMS


(iii) Rules and procedures Quantification must be done according to explicit rules and pro cedures. That is, the 'blind' or haphazard assignment of numbers to characteristics of individuals cannot be regarded as measurement. In order to be considered a measure, an observation of an attribute must be replicable, for other observers, in other contexts and with other individuals. Practically anyone can rate another person's speaking ability, for example. But while one rater may focus on pronunciation accuracy, another may find vocabulary to be the most salient feature. Or one rater may assign a rating as a percentage, while another might rate on a scale from zero to five. Ratings such as these can hardly be considered anything more than numerical summaries of the raters' personal conceptualizations of the individual's speaking ability. This is because the different raters in this case did not follow the same criteria or procedures for arriving at their ratings.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 16

DEFINITION OF BASIC TERMS DEFINITION OF BASIC TERMS


Measures, then,

are distinguished from such 'pseudomeasures' by the explicit procedures and rules upon which they are based. There are many different types of measures in the social sciences, including rankings, rating scales, and tests.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 17

DEFINITION OF BASIC TERMS


TEST Carroll provides the following definition of a test: a psychological or educational test is a procedure designed to elicit certain behavior from which one can make inferences about certain characteristics of an individual (1968:46).
From this definition, it follows that a test is a measurement

instrument designed to elicit a specific sample of an individual's behavior. As one type of measurement, a test necessarily quantifies characteristics of individuals according to explicit procedures. What distinguishes a test from other types of measurement is that it is designed to obtain a specific sample of behavior.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department

Testing and Evaluation

18

DEFINITION OF BASIC TERMS

EVALUATION Evaluation can be defined as the systematic gathering of information for the purpose of making decisions (Weiss 1972).The probability of making the correct decision in any given situation is a function not only of the ability of the decision maker, but also of the quality of the information upon which the decision is based. Everything else being equal, the more reliable and relevant the information, the better the likelihood of making the correct decision. Few of us, for example, would base educational decisions on hearsay or rumor, since we would not generally consider these to be reliable sources of information. Similarly, we frequently attempt to screen out information, such as sex and ethnicity, that we believe to be Prof. Dr. Dinay Kksal irrelevant to a particular decision.
anakkale Onsekiz Mart University ELT Department 19

Testing and Evaluation

DEFINITION OF BASIC TERMS


One aspect of evaluation, therefore, is the collection of

reliable and relevant information. This information need not be, indeed seldom is, exclusively quantitative. Verbal descriptions, ranging from performance profiles to letters of reference, as well as overall impressions, can provide important information for evaluating individuals, as can measures, such as ratings and test scores.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 20

DEFINITION OF BASIC TERMS


Evaluation, therefore, does not necessarily entail testing. By the

same token, tests in and of themselves are not evaluative. Tests are often used for pedagogical purposes, either as a means of motivating students to study, or as a means of reviewing material taught, in which case no evaluative decision is made on the basis of the test results. Tests may also be used for purely descriptive purposes.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 21

Basic terms
Assessment This term is used to refer to a vatiety of ways

of collecting information on learners language ability or achievement. Testing and assessment are often used interchangeably, however the latter is an umbrella term which emcompasses measurement instruments administered on a one-off basis such as tests as well as qualitative methods of monitoring and recording student learning such as observation, simulations, or project work.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department

Testing and Evaluation

22

Basic terms
Assessment is also distinguished from evaluation concerned with the overall language programme and not just what individual learners have learnt. Here it will be useful to mention about the kinds of assessment: Profeciency assessment vs achievement assessment: Proficiency assessment refers to the assessment of learners general language abilities independent of a course of study and such assessment is often done by administering standardised language proficiency tests while achievement assessment is frequently carried out by the teacher based either on the specific course content or on the course objectives to establish what a learner has learnt in relation to a particular course.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 23

Basic terms
Formative assessment vs summative

assessment Formative asseement is carried out by teachers during the learning process

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 24

APPROACHES TO LANGUAGE TESTING

When we have a quick glance at the history of language testing, we see that the shifts in emphasis in language teaching have inevitably had consequences for language testing. Testing techniques and theories, however, have been rather more resistant to change than theories about methodology and course design. Changes in approaches to language teaching inevitably resulted in attempts to develop testing techniques appropriate to the new pedagogy. Such fundamental questions as 'What makes a test a good test?' and 'How, should we go about constructing a test?' will receive quite different answers from adherents to different schools. Procedures acceptable to one approach may be anathema to another and so on.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department

Testing and Evaluation

25

The approaches to language testing can be categorized into five main groups

Era

Pychology Learning Theory

Linguistic s Theory of language

Methodology Testing / Aprroach Language teaching EssayTranslation Discretepoint

Pre-scientific Faculty

Traditional GrammarTranslation

Scientific Modern

Behaviouristi Structuralis Audio-lingual c m

Cognitivistic Generativis Cognitive-code Integrative m learning theory

Communicati Humanistic Semanticis Communicativ Pragmatic ve m e Approach Prof. Dr. Dinay Kksal Inovative
Testing and Evaluation

Psycholinguis PragmatismFunctionalanakkale Onsekiz Mart tic University ELT Department notional

Functional
26

PRESCIENTIFIC/TRADITIONAL APPROACH
Essay-translation approach Essay-translation approach, commonly refered to as prescientific stage of language testing, requires no special skill or expertise in testing , best characterized by the use of tests such as essay-writing, translation, grammatical analysis.

Based on the language teaching philosophy in grammartranslation method, learners are required to write open-ended written compositions which consists of passsages for translation from and into the target language with a heavy literary and cultural bias, free compositions in it and selected items of grammatical, textual or cultural interest. One must rely completely on the subjective judgement of the experienced teacher. It is assumed that a person who can teach can also judge the proficiency of his learners. There is a lack of concern for statistical matters or for such notions as objectivity and reliability. Prof. Dr. Dinay Kksal
anakkale Onsekiz Mart University ELT Department 27

Testing and Evaluation

STRUCTURALIST APPROACH
Discrete point approach/ pyschometric-structuralist trend)

This approach, commonly referred to as psychometric-

structuralist trend, or as scientific stage of language testing, tries to identify and test the test-takers, or learners mastery of separate elements of the target langauge (phonology, voacabulary, and grammar), completely divorced words and sentences from context on the gounds that it makes it possible to cover a larger sample of language forms in a comparatively short time. Communicative skills (listening, speaking, reading and writing) are also tested separately because it is of great importance to test one thing at a time. to provide objective measures using various statistical techniques to ensure reliability and certain kinds of validity.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 28

Unlike essay-translation approach, the key concern here is

Testing and Evaluation

STRUCTURALIST APPROACH
In this period of language testing, on the theoretical side, it

is agreed that language learning is chiefly considered as the systematic acquisition of a set of habits; on the practical side, testers wanted and structuralists knew how to deliver long lists of small items that could be tested objectively. testers notion of discrete skills to be measured, we witness the flourishing of standardised tests with emphasis on what is called discrete-point item. Such tests aimed to serve the following goals: 1. diagnosing learner strengths and weaknesses; 2. prescribing curricula at particular skills; and 3. developing specific teaching strategies to help learners overcome particular weaknesses.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 29

Based on the structural linguists view of language and

Testing and Evaluation

STRUCTURALIST APPROACH
4. planning for remedial instructions 5. helping the educators to design effective programms 6. facilitating test administration and scoring procedures They are considered disadventagous because they have outdated and weak psychological and linguistic backgrounds are not efficient in measuring the learners proficiency with little or no relevance with the actual use of language in real life situations do not give any idea about the contribution of every single language element to the overall langauge use.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 30

STRUCTURALIST APPROACH
The assumption that language components and skills cannot

operate independently and that language elements should be tested as a whole since they all funtion together to make language operate as a means of communication gave way to the holistic approach. The need for assessing the practical language skills of foreign students who wanted to study in the UK and USA in paralel with the need within the communicative movement in teaching for tests which measured productive language skills led a demand for language tests that required language users to have an integrated performance.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 31

INTEGRATIVE APPROACH
Integrative/Holistic Approach This approach, primarily concerned with the total communicative effect of discourse and meaning, involves the testing of language in context. Integrative tests , best characterized by the use of cloze tests, dictation, oral interviews, translation and essay writing, are designed to test the test-takers ability to use two or more skills simultaneously. The arguments for the tests based on this approach are: 1. They have stronger linguistic and pyschological backgrounds in comparison with the discrete point-tests. 2. They pertain to the contextualised form of language. Prof. Dr. Dinay Kksal
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 32

INTEGRATIVE APPROACH
The arguments against such tests are as follows:

1. They are time-consuming to administer and score. 2. Reliability and valididity of these tests are a matter of discussion. 3. The items in dictation and or cloze tests are interdependent. 4. Scores on dictation tests depend on understanding the spoken form of the language and the ability to speed write. 5. They required trained raters
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 33

COMMUNICATIVE APPROACH
Communicative Approach (functional tests)

Communicative tests, primarily concerned with how langauge is used in communication, aim to include tasks that approximate to those facing the learners in real life situations. Hymes theory of Communicative Competence has changed the scope of language testing. Hymes (1970)maintains that knowing a language requires more than knowing its rules .There are culturally specific rules of use, which relates the language used, to features of the communicative context.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department

Testing and Evaluation

34

COMMUNICATIVE APPROACH
The model of Communicative Competence ,developed by

Canale and Swain (1980) involves four types of competence ; Grammatical Competence : the kind of knowledge of systematic features of grammar ,lexis and phonology . (discrete-point testing) Socio-cultural Competence : Knowledge of rules of language use in terms of what is appropriate to different types of interlocutors , to different settings ,in different topics. Strategic Competence: the ability to compensate in performance for incomplete and imperfect linguistic resources in a second language Discourse Competence : the ability to deal with extended use of language in context
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 35

Testing and Evaluation

COMMUNICATIVE APPROACH
In communicative language tests , language has been used in a manner appropriate to its context .That is the language introduced in tests should be useful and authentic. There are two important features of the communicative language tests; These are performance tests,requiring assessment to be carried out ,when the learner lor the candidate is engaged in the act of communication ,either receptive, or productive, or both. They pay attention to the social roles candidates are likely to assume in real world settings and offer a means of specifying the demands of such rules in detail.
The British ELTS Test (International Language TestingSystem ) and The American TOEFL (Test of English as a Foreign Language )are good examples of tests widely used in English .

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 36

KINDS OF FOREIGN LANGUAGE TEST AND TESTING PROGNOSTIC TEST PLACEMENT/ENTRY TEST SELECTION TEST DIAGNOSTIC TEST PROGRESS TEST ACHIEVMENT TEST PROFICIENCY TEST LANGUAGE APTITUDE TEST DIRECT AND INDIRECT TESTS NORM-REFERENCED AND CRITERION-REFERENCED TESTS INTEGRATIVE AND PRAGMATIC TESTS COMMUNICATIVE TESTS TEACHER-MADE TESTS VS.STANDARDIZED TESTS
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 37

PROGNOSTIC TEST

It refers to tests used to predict the future course of action about the test-takers and make decisions about the most appropriate channel of educational or occupational career for the test-takers and their future success on the basis of their present capabilities. These tests are categorized into two main groups: (i) Selection test and (ii) placement test.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 38

PLACEMENT/ENTRY TEST

The term Placement test refers to the measurement of the test-takerscapability in pursuing a certain path of langauge learning. They are administered to learners at the beginning of a course or the academic year to identify their levels. This type of test will indicate at which level a learner will learn most effectively in a situation where there are different levels or streams. The aim is to produce groups which are homogeneous in level that will use institutional and teacher time most effectively. The larger the groups of learners to be designed, the more homogeneous the groups will need to be and therefore the more reliable the entry or placement test will need to be. This type of test is less useful where students are grouped alphabetically or according to age rather than ability. Entry or placement tests are not very common in state run institutions.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 39

Testing and Evaluation

SELECTION TEST

These tests are used to provide information concerning the test-takers acceptance into a particular programme. For instance, according to the scores the test-takers get from TOEFL, they are accepted into academic programme in the universities in USA.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 40

DIAGNOSTIC TEST

As the name suggests, this type of test is used to find out problem areas. Where other types of tests are based on success, diagnostic tests are based on failure. We want to know in which areas a student or group of students are having problems, which parts of a course or learning objectives those students cannot cope with. One way of looking at this type of test is to consider it as a technique based on eliciting errors rather than correct answers or language. Diagnostic information is vital for teachers in order to design further course activities and work out remedial activities. The information can also be useful for learners, as they can analyze their own problems. Diagnostic testing is present in many progress tests for the simple reason that the progress tests identify problem areas. However, a reliable diagnostic test is difficult to design.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 41

Testing and Evaluation

PROGRESS TEST

Progress tests, administed during courses, may be used after certain blocks of study, eg after x number of units, at the end of each week, each term etc during courses, aiming to find out information about how well classes as a whole and individual students have grasped the learning objectives, how well the course content is functioning within the specified aims and objectives and future course design.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 42

PROGRESS TEST
These tests help teachers identify easily how well students

are progressing in a very short period of time, eg a progress test of half an hour can give a great deal of information about the class if the test is well designed and if the test samples widely from the course content. Progress tests can also provide important feedback to the learners. When linked with self-assessment, feedback can help learners to identify their own problems and to set their own goals for the future learning.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 43

ACHIEVEMENT TEST

This type of test is designed to measure how much a learner has learnt rfom a particulr course or sylllabus. For instance, an achievement test may be a reading comprehension test based on the reading passages in a coursebook. It helps the teacher to judge the success of his/her teaching and determine the learners weaknesses. A proficiency test may use similar test items but as mentioned before they are not linked to any particular textbook or language syllabus. The achievement test serves two purposes: determining progress and diagnosing learner weaknesses.
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

44

PROFICIENCY TEST

This type of test is designed to measure the test-takers general level of langauge mastery . It is not linked to a particular course of instruction. Some proficiency tests have been standardized for world-wide use, such as the American TOEFL used to test the English langauge proficiency of foreign students who wish to study in the United States.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 45

PROFICIENCY TEST
This type of test aims to describe what students are

capable of doing in a foreign language and are usually set by external bodies such as examination boards. Proficiency tests enable students to have some proof of their ability in a language. They also provide potential employers with some guarantee of proficiency in a language because examination boards are seen as bodies which set standards in an impartial way and boards' examinations are generally considered to be reliable and valid. Some proficiency tests, while claiming to be communicative, often have a large language component such as grammar or vocabulary. This can have a negative washback effect on teaching in terms of examination preparation.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 46

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 47

DIRECT AND INDIRECT TESTS


Test formats and procedures that involve the use of

language for communication in real life situations are referred to as direct tests. For intance, oral interviews. Testing is said to be direct when it requires the test-takers to perform precisely the skill we intend to measure. That is, if we want to know how well they speak, we need to get them to speak and if we want to know how well they write compositions, we need to get them to write compositions. The tasks and texts shoul be as authentic and realistic as possible

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 48

DIRECT AND INDIRECT TESTS


The advantages of direct testing 1. Providing that we are clear about what abilities/skills we wish to assess, it is relatively simple to create the conditions which will elicit the behaviour on which we can base our judgements. 2. In the case of the productive skills, the assessment and interpretation of the test-takers performance is also quite straightforward. 3. Direct testing may have positive washback effect since practice for the test involves practice of the skills we want to develop. 4. Direct tests are usually constrasted with indirect tests
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 49

DIRECT AND INDIRECT TESTS


When a test attempts to measure the abilities underlying

the skills we are interested in, it is called indirect test. For instance, one section of the TOEFL in which the testtakers have to identify erroneous or inappropriate elements in formal standard English was developed as indirect measure of writing ability. Another example is testing pronunciation ability by a paper and pencil test in which test-takers have to identify pairs of words which rhyme with each other. Since the relationship between performance on these tests and performance of the skills in which we are usually more interested tends to be weak in strength and uncertain in nature, they doProf. Dr.give us feedback. not Dinay Kksal
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 50

DIRECT AND INDIRECT TESTS


Should you aim for direct or indirect testing? To help

with this decision you may find the following helpful: Indirect testing makes no attempt to measure the way language is used in real life, but proceeds by means of analogy. Some examples that you may have used are: Most, if not all, of the discrete point tests mentioned above. Cloze tests Dictation (unless on a specific office skills course) Indirect tests have the big advantage of being very testlike. They are popular with some teachers and most administrators because can be easily administered and scored, they also produce measurable results and have a high degree of reliability.
.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 51

DIRECT AND INDIRECT TESTS


Direct tests, on the other hand, try to introduce authentic tasks, which model the students real life future use of language. Such tests include: Role-playing. Information gap tasks. Reading authentic texts, listening to authentic texts. Writing letters, reports, form filling and note taking. Summarising. Direct tests are task oriented rather than test oriented, they require the ability to use language in real situations, and they therefore should have a good formative effect on your future teaching methods and help you with curricula writing. However, they do call for skill and judgment on the part of Prof. Dr. Dinay Kksal the teacher.
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 52

DISCREET POINT AND INTEGRATIVE TESTING


You may see or hear these words when being asked to assess or

even write a test so lets see what they mean. Discrete Point tests are based on an analytical view of language. This is where language is divided up so that components of it may be tested. Discrete point tests aim to achieve a high reliability factor by testing a large number of discrete items. From these separated parts, you can form an opinion is which is then applied to language as an entity. You may recognise some of the following Discrete Point tests: 1. Phoneme recognition. 2. Yes/No, True/ False answers. 3. Spelling. 4. Word completion. 5. Grammar items. 6. Most multiple choice tests. Such tests have a down side in that they take language out of Prof. Dr. Dinay relationship to the concept or use of context and usually bear no Kksal whole language.
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 53

DISCREET POINT AND INTEGRATIVE TESTING


Integrative tests In order to overcome the above defect, you should

consider Integrative tests. Such tests usually require the testees to demonstrate simultaneous control over several aspects of language, just as they would in real language use situations. Examples of Integrative tests that you may be familiar with include: 1. Cloze tests 2. Dictation 3. Translation 4. Essays and other coherent writing tasks 5. Oral interviews and conversation 6. Reading, or other extended samples of real text
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 54

COMMUNICATIVE TESTS
These tests require test-takers to perform communicatively

like that of real-life situations. Communicative tests are concerned not only with these different aspects of knowledge but on the test-takers ability to demonstrate them in actual situations. So, how should you go about setting a Communicative test? Firstly, you should attempt to replicate real life situations. Within these situations communicative ability can be tested as representatively as possible. There is a strong emphasis on the purpose of the test. The importance of context is recognised. There should be both authenticity of task and genuiness of texts. Tasks ought to be as direct as possible.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 55

COMMUNICATIVE TESTS
When engaged in oral assessment we should attempt to

reflect the interactive nature of normal speech and also assess pragmatic skills being used.

Communicative tests are both direct and integrative. They

attempt to focus on the expression and understanding of the functional use of language rather than on the more limited mastery of language form found in discreet point tests. subject to criticism in some quarters, yet as language teachers see the positive benefits accruing from such testing, they are becoming more and more acceptable.
Prof. Dr. Dinay Kksal

The theoretical status of communicative testing is still

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

56

COMMUNICATIVE TESTS
They will not only help you to develop communicative

classroom competence but also to bridge the gap between teaching, testing and real life. They are useful tools in the areas of curriculum development and in the assessment of future needs, as they aim to reflect real life situations. For participating teachers and students this can only be beneficial.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 57

COMMUNICATIVE TESTS
While preparing a communicative language test the following criteria should be taken into consideration: Be precise about the skills and performance conditions Identify the most important components of language use in particular contexts. Ensure that the sample of communicative language ability in our tests is as representative as possible. Ensure that tests meet the performance conditions of a context as fully as possible. The important role of context should be stressed, because language without a context cannot be meaningful. Use authentic materials and genuine texts, because using inauthentic materials may cause some problems. Tests of communicative language ability should be as direct as possible and attempt
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 58

COMMUNICATIVE TESTS
to reflect the real life situation. Unsimplified language should be used as input. The tasks should be conducted under normal time

constrains. Minimize any detrimental effect that may cause disturbance for the candidate. The holistic and qualitative assessment of productive skills should be utilized.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 59

TEACHER-MADE TESTS VS.STANDARDIZED TESTS


The differences between teacher-made tests and

standardised tests can be listed as follows: 1. As the name suggests, teachers determine the content of the test whereas the content of the standardised tests are determined by the curriculum, that is, what to teach and what to test is defined in the syllabus. 2. The directions in teacher-made tests may be understood by the students of the teacher who gives those directions only in his classes but may not mean anything to the students from other schools, while the ones in standardised tests can be understood by students from all schools and nationalities because they follow a uniform procedure and culture free.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 60

TEACHER-MADE TESTS VS.STANDARDIZED TESTS


3. The norms in teacher-made tests are local and may be interpreted differently by different teachers, while in standardised tests there are norms described by the specialists to be neither low nor too high. If the standards are too high, the learners will be discouraged and if they are too low, the learners will find the tests very easy and will not put forth their maximum efforts. The standards must be uniform, well balanced and clearly defined according to the the expectations, objectives and the level of proficiency to be attained by the learners. Tecaher-made tests that are generally hurriedly constructed by a single teacher are dependent on the teachers intuition, knowledge and experience while standardised tests that are constructed by a group of experts are pretested after the test items are prepared. Then upon the item analysis the test items are revised before the test is administered.

4.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 61

TEACHER-MADE TESTS VS.STANDARDIZED TESTS


5. In teacher-made tests reliability is usually unknown. It can be high if they are carefully constructed while in standardized tests it is high, frequently above 0.90.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 62

NORM-REFERENCED(NRTs) AND CRITERION-REFERENCED TESTS(CRTs)


Norm Referenced Tests

Norm Referenced tests answer such questions as, how student A compares with student B. Attainment or Achievement tests should be specifically designed to answer this question. Criterion Referenced tests answer such questions as, How much has student Y learnt? Or, How much does student X know? Proficiency tests should be designed to answer such questions. NRTs thus differ from CRTs in focus, timing, purpose and theoretical motivation and reflect different perspectives and goals. Brown (1995) classifies CRTs and NRTs according to their test characteristics and logistical dimensions
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department

Criterion Referenced Tests

Testing and Evaluation

63

NORM-REFERENCED(NRTs) AND CRITERION-REFERENCED TESTS(CRTs)


Differences between NRTS and CRTS (Brown, 1995, p.12).

CRTs
Foster learning Diagnosis, progress, achievement Classroom specific Know content to expect

NRTs
Classify/group students Aptitude, proficiency, placement Overall, global Do not know content Percentile

Test Characteristics Underlying Purposes Types of Decisions Levels of Generality Students' Expectations

Score Interpretations Percent Score Strategies ReportTests and students answers

toOnly scores go to students

Logistical Dimensions Group Size Range of Abilities Test Length Time Allocated Relatively small group Relatively homogeneous Relatively few questions Large group Wide range of abilities Large number of questions Long (2-4 hours) administration

Prof. Dr. Dinay Kksal Relatively short time

Testing and Evaluation

Cost

anakkale Onsekiz Mart Teacher time & duplication Test booklets, tapes, proctor University ELT Department

64

PRAGMATIC TESTS

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 65

BASIC CONSIDERATION IN TEST DESIGN:CHARACTERISTICS OF GOOD LANGUAGE TESTS


The concept of validity:testing the test 1. Construct validity 2. Content validity 3. Face validity The concept of reliability Practicality

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 66

The concept of validity


The purpose of a test is to measure a learners knowledge

or ability, but it can only be considered valid if it actually measures what it is intended to measure. The concept of validity is significant for all aspects of the design of all types of test, whatever the approach or methods used for learning. A consideration of various types of validity will include the following: Construct validity Content validity Face validity Washback validity Criterion-related validity
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department

67

The concept of validity:testing the test


Construct Validity Mousavi (1997) defines construct validity as a form of

validity which is based on the degree to which the items in a test reflect the essential aspects of the theory on which the test is based (construct). Thus, if a test is based on a different construct from that used during the learning process, it cannot be said to have construct validity. Weir (1990) tells us that some researchers consider construct validity to encompass the other types of validity (Anastasi, 1982; Cronbach, 1971). A test is like a definition of a construct and must therefore adequately reflect the construct in order to be valid.
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

68

The concept of validity:testing the test


We can discover what a test measures by relating test

scores to other external data. Weir (1990) views such data as a necessary condition for establishing test validity, but not of itself sufficient. Test designers should attempt to establish a priori construct validity for a test, even if there is no adequate framework for the construction of tests, as is the case with the communicative approach. Following the test, statisitical procedures can be applied to determine the extent to which the test is valid.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 69

The concept of validity:testing the test


Convergent and discriminatory validity are two processes

used to investigate the statistical relationship between a test and the construct being measured. Convergent validity means that the same trait is tested in different ways and there is a high level of correlation between the results. Discriminant validity requires that the results of two different methods of testing the same construct have a high correlation, whereas the same method of testing applied to different constructs should have a lower correlation value.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 70

The concept of validity:testing the test


Statistical a posteriori validation thus occurs if a test

correlates closely with the behaviour it is intended to measure. However, another school of thought emphasises the need for a priori construct validation of test design. Since validity is based on the construct, and the construct is in turn based on a theory, the theory must be developed as fully as possible, so that clear statements can be made about what should be measured. Following the test, statistical investigations can then be used to confirm validity.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 71

The concept of validity:testing the test


Content Validity Content validity refers to the extent to which a test contains representative samples of and measures those aspects of behaviour it is designed to measure. Some scholars consider that construct and content validity overlap to a great extent, especially in the case of tests of general proficiency. Communicative tests are often intended to pinpoint areas of deficiency and content validity is very useful in this case, as it determines the extent to which the tasks used in the test reflect the universe of tasks being tested. Anastasi (in Weir, 1990) gives guidelines for establishing content validity:
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 72

The concept of validity:testing the test


1.

2.

3.

the behaviour domain to be tested must be systematically analysed to make certain that all major aspects are covered by the test items and in the correct proportions the domain under consideration should be fully described in advance, rather than being defined after the test has been prepared content validity depends on the relevance of the individuals test responses to the behaviour area under consideration, rather than on the apparent relevance of item content
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

73

The concept of validity:testing the test


Defining the area of language to be tested is thus of

paramount importance in establishing content validity, but this can sometimes be very difficult, especially if the area to be covered is very wide or imprecise. Testers should, however, aim to design a test which is as relevant as possible. It is very important to specify clearly what exactly is being tested and to construct items accordingly. One way of investigating whether this has been achieved is to carry out an introspective study with some of the test candidates. This can help establish whether the processes used by candidates actually conform to the test specifications.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 74

The concept of validity:testing the test


Face Validity If, to an observer, a test appears to measure what it

purports to measure, then it is said to have face validity. This is a subjective evaluation which thus has no place in objective assessments of validity and is therefore rejected as being of no relevance by some scholars. However, the question of face validity may be of significance to test takers, administrators and other interested parties. If there is no perceived validity, and the test seems unacceptable to candidates, their performance may be affected. Test designers must not overemphasise the importance of face validity to the detriment of objective validity, but it seems it should be taken into consideration for practical reasons.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 75

The concept of validity:testing the test


Washback Validity Washback validity is defined by Mousavi (1997) as the
degree to which a test satisfies students, teachers and future users of test results. If tests are designed to provide feedback for teachers and students and thus identify areas of weakness or strength, then the tests possess washback validity. They contribute to learning and can also have an effect on the curriculum. In the case of such tests, there is obviously a close relationship between the test and the teaching which preceeded it.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 76

The concept of validity:testing the test


Criterion-related Validity Whereas the types of validity discussed so far are closely

related to what the test measures, in the case of criterionrelated validity this relationship is not so crucial. Criterionrelated validity is concerned with the correlation of test scores to external performance criteria and can be divided into concurrent and predictive validity. Cohen (in CelceMurcia, 1991) explains them as follows:

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 77

The concept of validity:testing the test


1.

Concurrent validity: test results are compared with results from another test given at about the same time. Predictive validity: test results are compared with results from another test or another type of measure obtained at a later date.

2.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 78

The concept of validity:testing the test


Although some researchers maintain that external validation

is always preferable to internal analysis of a test, it is possible that a test valid in this way may not possess construct validity. Even if it does, it may be difficult to find similar tests to be used as concurrent criteria in this method of validation, especially in communicative testing. In such cases, it may be possible to use other non-test criteria, although it would be extremely difficult to establish these.

Criterion-related validity measures can be very useful,

especially if the results of a test will have an effect on the candidates futures. Criterion-related validity can indicate the equivalence of one test with another.
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

79

The concept of reliability


Reliability refers to the consistency of test scores. It simply

means that a test would give similar results if it were given at another time. It is about producing precise and repeatable measurements on a clear scale of measurement units. Such tests give consistent results across a wide range of situations. This is achieved by carefully piloted trials of the test. Sometimes, several versions of the test may be used on a controlled population of testees. The outcomes of these trials are carefully analysed in order to establish consistency. When a consistent set of figures is achieved the test may be deemed reliable.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 80

The concept of reliability


Three important factors effect test reliability.

Test factors such as the formats and content of the questions and the length of the exam must be consistent. Administrative factors are also important for reliability. These include the classroom setting (lighting, seating arrangements, acoustics, lack of intrusive noise etc.) and how the teacher manages the exam administration. Affective factors in the response of individual students can also affect reliability. Test anxiety can be alleviated by coaching students in good test-taking strategies. Prof. Dr. Dinay Kksal
anakkale Onsekiz Mart University ELT Department 81

Testing and Evaluation

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 82

Social issues in language testing


Social issues in language testing 1. Ethics in language testing 2. Accountability 3. Washback 4. Test impact

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 83

Testing Ethics
The moral and social considerations involved

in the construction and use of tests.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 84

Washback
The influence of tests on teaching and learning is called

the washback or backwash effect. If your students have to do a test or maybe a public examination at the end of the course this will affect the syllabus. If we have a good test, this should affect teaching in a positive manner. If we have a bad test, this might affect teaching in a negative manner. What is a 'good' test and what is a 'bad' test? A test can have a positive influence if it contains authentic, real-life examples of the type of tasks which your learners will need to perform in the future.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 85

Washback
Tests can have a negative influence if they

contain artificial tasks not linked to real future needs. Teaching methods will probably reflect these tasks and the learning process could end up revolving around what we might term 'exam practice'. Your own tests will also have an effect on your students' learning. If you test mainly grammar, your students will assume that this is the most important thing to learn and may make less effort during other more communicative activities.
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

86

Washback
Hughes (2003:53-57) makes the following suggestions to

achieve beneficial or positive backwash. We should test the abilities whose development we want to encourage. If the teacher want to encourage oral ability he must test oral abilility. Teacher generally tend to test what is easiest to test rather than what is most important to test(p.53) because oral interview, a direct way of testing oral ability, requires subjective scoring, or marking. However, we know that it is possible to obtain high level of reliability through analytic rating.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 87

Washback
We should train teachers about the

innovations in language teaching and testing. Guidance and training has to be given to teachers when a new test is introduced. If these are not given, the test will not produce the intended effect. New tests require changing teaching since test techniques must be paralel to teaching techniques. Dr. Dinay Kksal Prof.
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 88

Washback
We should sample widely and

unpredictably The sample should be representative of the full scope of the specifications. It is necessary to include a wide range of tasks when testing a skill if we want backwash to be beneficial. Similarly, the content of a test shouldnt be predictable if we do not want teaching and learning to concentrate on what can be predicted(p.54). So tests must contain the full range of specifications.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 89

Washback
We should use direct testing If we want learners to learn to write a

scientific report, then we should get them to write a scientific report in the test. Direct testing refers to the testing of performance skills through the use of authentic texts and tasks.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 90

Washback
We should make testing criterion-

referenced Test specifications should make clear what test-takers can do, with what degree of success if we want learners to have a clear picture of what they are expected to achieve. Learners who can perfom the tasks at the criterial level are considered successful, regardless of how other learners perform. These things help motivate learners.
Prof. Dr. Dinay Kksal anakkale Onsekiz Mart University ELT Department 91

Testing and Evaluation

Washback
we should base achievement tests on

objectives To provide a better picture of what has been actually achieved in the learning and teaching process, it is necessary to base achievement tests on objectives, rather than on detailed teaching and coursebook content.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 92

Washback
We should ensure that the test should

be understood by students and teachers If we want to get beneficial washback effect, language teachers and learners need to understand and know what the test demands of them. The rationale and specifications and sample items should be made available to those responsible for test preparation. This kind of information will also increase test reliability. Prof. Dr. Dinay Kksal
anakkale Onsekiz Mart University ELT Department 93

Testing and Evaluation

Accountability
Accountability A final element is accountability. As professionals, teachers should be able to provide learners, parents, institutions and society in general, with clear indications of what progress has been made and if it has not, why that is so. We should be able to explain the rationale behind the way assessment takes place and how conclusions are drawn, rather than hiding behind a smoke screen of professional secrecy.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 94

Test impact
The wider effect of tests on the community as a whole

beyond the classroom, including the school, is referred to as test impact. For instance, such tests asTOEFL and IELTS which are used as gatekeeping mechanisms for international education and administered to huge number of candidates all over the world.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 95

Test impact
The impact of test use operates at two levels:
Test taking and Use of test scores Macro:Society,education System Micro:Individuals

Impact

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 96

Test impact
The impact of test use needs to be considered

within the values and goals of society and educational programme in which it takes place and according to the potential consequences of such use.

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 97

Holistic vs. Analytic Scoring of Writing


Holistic Scoring:
Often referred to as "impressionistic" scoring Involves the assignment of a single score to a piece of writing on the basis of an overall impression of it. Individual features of a text, such as grammar, spelling, and organization, should not be considered as separate entities. Has the advantage of being very rapid (Hughes 1989: 86).

"Since, in holistic scoring, the entire written text is evaluated as a whole, it is important to establish the specific criteria upon which the evaluation is to be based prior to undertaking the evaluation. This does not mean establishing a catalogue of precise individual errors that might appear, but rather deciding what impact the errors that are present have on the overall tone, structure, and comprehensibility of the writing sample" (Terry 1989: 49).
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 98

Holistic vs. Analytic Scoring of Writing


Holistic Scoring Scale used by Educational Testing

Service for evaluating the Advanced Placement Examination in foreign languages:

works well and can be altered to fit the level of the

students and the focus of instruction. numerical scale that ranks performance at levels described as "superior," "competent," and "incompetent." for each level, the descriptions can be changed to reflect the kind of performance that teachers expect at a given level of language ability. reliability of this scoring method is considered good when the raters are trained to establish common standards based on practice with the kinds of writing samples that they will be evaluating (Cooper 1977).

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 99

Holistic vs. Analytic Scoring of Writing


Analytic Scoring:
A method of scoring that requires a separate score for each of a number of aspects of a task, such as grammatical accuracy, vocabulary, idiomatic expression, organization, relevance, coherence. Disposes of the problem of uneven development of subskills in individuals. Scorers are compelled to consider aspects of performance which they might otherwise ignore. The very fact that the scorer has to give a number of scores will tend to make the scoring more reliable. In some schemes, each of the components is given equal weight. In other schemes, the relative importance of the different aspects, as perceived by the tester (with or without statistical support), is reflected in weightings attached to the various components.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 100

Holistic vs. Analytic Scoring of Writing


Disadvantages: It takes more time than holistic scoring. Concentration on the different aspects may divert

attention from the overall effect of the piece of writing. Inasmuch as the whole is often greater than the sum of its parts, a composite score may be very reliable but not valid (Hughes 1989: 93-94) "Students should receive a copy of the analytic scoring criteria in advance. They will then know what is expected of them and how to interpret the evaluation of their written work. They can readily see where their strengths and weaknesses lie and can, over time, visualize their progress with subsequent evaluated samples of their writing" (Terry 1992: 247).
Prof. Dr. Dinay Kksal

Testing and Evaluation

anakkale Onsekiz Mart University ELT Department

101

REFERENCES
Harris David 1979. testing English as a Second Language.

New York: McGraw-Hill Book Company. Birjandi Parviz, Bagheridoust, Esmaeil, Rarviz Mossalanejad 1999. Language Testing: A Concise Collection for Graduate Applicants. An Alternative resolution to the Complications of Language Testing. Tehran: Shahid Mahdavi Publications. Brown, J. D. (1995). Differences between norm-referenced and criterion-referenced tests. In J. D. Brown, & S. O. Yamashita (Eds.), Language testing in Japan (pp. 12-19). Tokyo: The Japan Association for Language Teaching. Brown, James Dean. 1996. Testing in Language Programs. New Jersey: Prentice Hall. Bynom, Anthony. 2001.Testing: Basic Concepts: Basic Terminology originally published in ENGLISH TEACHING Prof. Dr. Dinay Kksal professional, Issue 20, July 2001.
Testing and Evaluation anakkale Onsekiz Mart University ELT Department 102

eir, Cyril J. 1990. Communicative Language Testing. New York: Prentice Hall. ughes, Arthur. 2003. Testing for Language Teachers.2nd edition. Cambridge: Cambridge University Press. aker David. 1989. Language Testing: A Critical Survey and Practical Guide. London: Edward Arnold. eaton, J.B. 1990. Writing English Language Tests. 3rd impression. London and New York: Longman.
Prof. Dr. Dinay Kksal

cNamara, Tim. 2000. Language Testing. Oxford: anakkale Onsekiz Mart Testing and Evaluation University ELT Department Oxford University Press.

103

Bachman, Lyle F. 1995. Fundemental

Considerations in Language Testing. Oxford: Oxford University Press. Bachman, Lyle F. And Palmer Adrian S. 2000. Language Testing in Practice. Oxford: Oxford University Press. Genesee, Fred and Upshur John A. 1998. Classroom-based Evaluation in Second Language Education. Third Impression. Cambridge: Cambridge University Press. Haris, Michael and McCann Paul. 1994. Assessment. Heinemann. Prof. Dr. Dinay Kksal Mousavi Abbas Seyyed. 1999. A Dictionary of anakkale Onsekiz Mart 104 Testing and Evaluation University ELT Department Language Testing. 2nd edition. Rahnama

Terry, Robert M. 1989. "Teaching and Evaluating Writing as

a Communicative Skill." Foreign Language Annals. 22 (1): 43-54. _____. 1992. "Improving Inter-rater Reliability in Scoring Tests in Multisection Courses." AAUSC Issues in Language Program Direction: Development and Supervision of Teaching Assistants in Foreign Languages. Boston:Heinle & Heinle Publishers. Cooper, Charles R. and Lee Odell, (Eds.). 1977. "Holistic Evaluation of Writing." In C.R. Cooper and L. Odell (Eds.), Evaluating Writing. Urbana, IL: National Council of Teachers of English, 3-31.
Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 105

John W. Oller, Jr. Foreign Language Testing, Part 1: Its

Breadth ADFL Bulletin 22, no. 3 (Spring 1991): 33-38

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 106

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 107

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 108

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 109

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 110

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 111

Prof. Dr. Dinay Kksal Testing and Evaluation anakkale Onsekiz Mart University ELT Department 112

Potrebbero piacerti anche