Communicative Testing

Communicative Testing

Language Performance Evaluation

Introduction

• A systematic testing component is an essential part of every language program and is used to measure:

• language proficiency • placement • diagnosis• progress • and achievement.

• A systematic testing component also provides:• feedback for the program evaluator(s), • washback information for teachers and students, • and motivational implications for all concerned.

Traditional Assessment (1)

• A test is "a systematic method of eliciting performance which is intended to be the basis for some sort of decision making" (Skehan, 1998, p. 153).

• There is a tendency of testers to place an emphasis on "standardization in assessment in the belief that such methods of examining performance will have more to contribute to reliable measurement than assessment by people who may be very familiar with particular language users" (Skehan, 1998, p. 153).


• According to this view, language can be learned by studying its parts in isolation, acquisition of these parts can be tested and will successfully predict performance levels, and the learner will somehow reconstruct the parts in meaningful situations when necessary.

• The easily quantifiable, reliable, and efficient data obtained from discrete (and cloze) testing implies that proficiency is neatly quantifiable in such a fashion (Oller, 1979, p. 212).


• However, Kelly (1978, pp. 245-246) pointed out that itis possible to develop proficiency in the integrativetest itself, and that discrete point tests cannotdiagnose specific areas of difficulty in relation to theauthentic task. Such tests can only supply informationon a candidate's linguistic competence, and havenothing to offer in terms of performance ability (Weir,1998).

Traditional Testing (4)

• In other words, knowledge of the elements of alanguage in fact counts for nothing unless theuser is able to combine them in new andappropriate ways to meet the linguistic demandsof the situation in which he wishes to use thelanguage (Morrow, 1979, p. 145).

• These facts led to a perception that the ability toperform should be tested in a specified socio-linguistic setting.

Emergence of Communicative Assessment (1)

• Based on work by Hymes (1972), Canale & Swain(1980), and Morrow (1979), the emphasis shifted fromlinguistic accuracy to the ability to function effectivelythrough language in particular contexts of situation (ademonstration of competence and of the ability to usethis competence).

• Communicative assessment was adopted as a meansof assessing language acquisition (though with somelack of initial agreement or direction, cf. McClean1995, p. 137; Benson, 1991).

Emergence of Communicative Assessment (2)

• What we need is a theory which guides and predicts:

– how an underlying communicative competence ismanifested in actual performance;

– how situations are related to one another,

– how competence can be assessed by examples ofperformance on actual tests;

– what components communicative competenceactually has;

– and how these interrelate.

• Since such definitive theories do not exist, testers haveto do the best they can with such theories as areavailable. (Skehan, 1988, cited in Weir, 1998, p. 7)

Traditional vs. Communicative Testing

Traditional Testing Communicative Testing

Testing and instruction are two separate activities Testing is an integral part of instruction

Students are conceived in a uniform way Each learner is seen as a unique person

Decisions are based on test scores Tests are one of many sources of data

Emphasis on weakness/failure (what students

cannot do)

Emphasis on strength/progress (what students can

do)

One-shot test Ongoing assessment

Cultural/socio-economic status bias Intercultural approach. More culture-fair

Focus on one “right answer” Possibility of several perspectives as in real life

Judgment without suggestions or opportunity for

improvement

Immediate feedback with useful information for

improving/guiding learning

Teaching is adapted to tests Tests are adapted to teaching

Focus on linguistic competence (language

components)

Focus on communicative competence (language

skills)

Promotes individual learning and comparison

between students (norm-referencing)

Encourages collaborative learning and compares

students to their own performance and the aims

Promotes extrinsic motivation for a passing grade Promotes intrinsic motivation for the student’s own

sake

Communicative Testing (1)

• As can be seen from this list, communicative testing isa learning tool, providing evaluative information toboth learners and teachers.

• Its focus on student-centered and student-managedongoing assessment also reflects educational thoughtin other areas of language acquisition:

– collaborative learning (Vygotsky, 1978);

– individual learning styles and preferences (Bickley,1989; Keefe, [Ed.], 1979; Reid, 1987);

– the importance of affect (Arnold, [Ed.], 1999);

– and the process syllabus (Breen, 1984).


• Communicative testing encourages a cycle ofintention, action and reflection, facilitated bycontextualized situations, which appeal to thestudent’s reality, thus becoming meaningful andself-motivational.

• In addition, an institutional respect for thelearner (which must be implicit in this approach),sees him/her as an active and sociallyresponsible agent, fully capable of needs analysis,

goal setting, and assessment of achievement.


• Psychologists and educators still know little about howlanguage learning occurs, and why and how someindividuals are more competent than others, so that itis wrong to test discrete symptoms of the process.

• Observable factors that appear to be associated withlearning include:

– construction of meaning,

– sharing of experiences,

– identification of needs and purposes,

– critical evaluation of performance strategies, andawareness of this process (Harri-Augstein &Thomas, 1991, p. 7).


• These factors can be satisfactorily examined(from the point of view of both teacher andstudents) using reflective, authentic,communicative and interactive testingmethods in appropriate learner-centeredclassroom activities.

• Integrated into the entire curriculum,assessment can become both a means and anend, and considerations of validity, reliabilityand efficiency remain as a major issue in theongoing reflective examination of languageperformance.

Conclusion (1)

• It would be ideal if test designers could followall these principles. But in reality, it is verydifficult to achieve this. First of all, “testingauthentic use of language” is extremelydifficult, though desirable. According to thisprinciple, to test a learner’s oral skills, it is bestto see how he or she performs in a realcommunication situation, for example, askingfor directions on the street.

Conclusion (2)

• What teachers usually do is to give the studenta situation and ask what he or she should sayin that situation. For example, What would yousay if you want to ask the policeman how to goto the railway station? Still, this is not realcommunication. That the student knows whatto say does not mean he or she can really say itin real situation.

Conclusion (3)

• In any case, assessment can be done in manyways, and testing is only one of them.

• When tests have to be used in assessment, theymust always follow a set of principles whichguarantee assessment validity (real-likecommunication) and reliability.

• Varying test formats according to theparticular assessment purposes and contextshelps to make testing fairer and more reliableand authentic.

Documents

Communicative Testing