Evaluation in language Education
Part IV:     TECHNIQUES OF SPECIAL INTEREST
1. Cloze Test and its Applications to Language Testing
2. Translation and its Significance in Language Testing

CLONZE TEST AND IT'S APPLICATIONS TO LANGUAGE TESTING:Cloze tests are based upon the Gestalt psychology. They generally refer to apparent ability of individuals to complete a pattern once they have grasped the structure of the pattern. In the past five or six years, scholars working in the area of language testing have evinced great interest in the use of Cloze Tests in Foreign Language Teaching. The concept of Cloze Test may be better explained with the following example:

A picture of a man's face is drawn on the black board leaving out an ear and a leg. The students are supposed to grasp what the picture is ment to be and when they are asked to complete the picture by drawing the ear and the leg. This is what is called clozure and the procedures in completing the picture are called the Cloze Procedures.

In the recent past, the British Council had reported in the TIMES EDUCATIONAL SUPPLEMENTS that "Cloze procedures have been shown to be highly useful, operational in assessing and grading performance and excited more interest than any other topic at the Seminar". The use of cloze test has been argued to be more useful in assessing the readability of text books. According to Pamela Riley, the cloze procedure may be defined as follows:

"Cloze procedure is a method of deleting words from a prose selection and evaluating the response a reader makes as he supplies the words deleted. Form the reader's score on the passage both readability or the reading difficulty of the passage and the reading achievement of the subject can be assessed. Words to be deleted (cloze items) may be at random or by a mechanical procedure every Nth (standard cloze), or according to a criteria such as word class, redundancy or predictability".21 The recent experiments with the cloze tests as opposed to the other types of tests and in particular the multiple choice comprehension questions, have shown that the former is as valid as any other type, and to add to that it has been found that the construction of such tests and their administrtion is much easier as opposed to the other types.

Cripwell has presented the following arguments supporting the use of cloze test as opposed to the multiple choice type:

Firstly, it is rare that two individuals will select the same points within a passage for questioning; this immediately introduces a subjective element. Secondly, it is equally rare that two individuals will give the same weightage to the questions they frame on any given point within the text. For example, if the text is : The cat is on the mat. One person could ask, "Where did the cat sit?" and the other, "Do you think cats should sit on mats?" Two different kinds of questions depending on subjectively in the construction of the distractions in the multiple choice battery. Only a great deal of committee work and test and evaluation can establish 'good' multiple-choice comprehension questions. Teachers rarely have the time nor the facilities to develop question banks for use in their classrooms. Cloze tests on the other hand are comparatively easy to construct. There is another factor that I think ought to be mentioned at this point. Comprehension questions tend to lead the reader by the nose; they organize him to look at the reading material from a certain point of view. In this way they can construct the students' reading. We have all seen how often the student turns to the questions before reading the passage. This is because the questions tell him what he should be looking for in the text. In the cloze test this is has been claimed that in this way the Cloze Test, unlike most other tests is a creative rather than a failing test.

The third point that has been observed is that students using Cloze Tests really do read the passage several times as instructed. This is often not the case with comprehension questions where attention to the text and re-reading is often sketchy as students often tend to work backwards from the questions to the text rather than the other way around.

A fourth factor is that readers are encouraged to examine the discourse of the total passage rather than individual items (lexis) or sentence (structure). This is an important element to stress, particularly for intermediate or advanced students. Finally, the test teaches the student to cope with redundancy by forcing him to concentrate on the meaning of the whole passage.

The uses of Cloze Tests can be modified. The first use of Cloze Test may be said to establish the levels of readability to a given text in terms of a given readership. This factor is more important in the case of a second or foreign language instruction. If the learners are able to read a given text with ease and fluency, the teaching methodology adopted can be said to be clear and the need for the re-explanation of the test does not exist. If the learners exhibit some difficulty in reading, it might be necessary to concentrate on the learner's more understanding of the text. If the learners find it extremely difficult and are found to be at the frustration level, it is desirable to discard that particular text. The reason may be either the language is too difficult or the content is far away from the experience of the learners.A second use of Cloze Test is to assess the achievement of the learner by using randomly selected passages from the texts and using them as the Cloze Test items. It has been proved that they provide a fair reflection of the learner's achievement in the class.

A third use of Cloze Test is said to be the teaching itself. After the administering of Cloze Tests and their scoring arguments could arise as to why some other words should not be used as alternatives for the expected words to be used to fill-in blanks in the passage. Such arguments through discussion in the classroom could be examined and solutions found out to the satisfaction of the teachers and the taught. Thus teaching can also be supplemented by the use of cloze test materials. Another use of cloze tests seems to be that a high correlation has been established between the listening and reading skills as measured by cloze tests.

As already mentioned above the concept of cloze test is very recent and it seems to offer a wide range of possibilities for experimentation to the class room teacher. It is very much within the reach of any language teacher to construct cloze tests and score them without much difficulty.



TRANSLATION AND ITS SIGNIFICANCE IN LANGAUGE TESTING
In the recent past, scholars working in the area of language teaching and language testing have questioned the appropriateness and suitability of translation as a device in language testing. The tendency seems to be that the classical tests like translation and composition are very likely to be retained to some extent within the present system of testing and examinations.

It is necessary to clearly devise the teaching and testing units that are geared to the needs of the language learners and for this purpose a classifications of the situations in which the languages are to be used by the learners has to be undertaken. One of the basis for describing the language needs in the light of situations in which the target language has to be made use of by the learners is the socio-professional and socio-cultural aspects. For example, a scientist will naturally be expected to know enough scientific items in the target language and to be able to understand, speak, read and even write in his field of specialization in the target language. However, certain situations necessitate only a passive knowledge of the language. The socio-cultural ambition that a learner may possess is likely to motivate him to acquire a certain quantum of generally frequent vocabulary items of the target language so that the learner will be adequately equipped to discuss topics relating to various subjects may be outside the field of his specialization. Normally people use two or even more languages in discussion wherein the listeners are in a position to understand these languages. Such situations where the speakers use two or more languages demand a special kind of testing strategy which may be called the bilingual type.

It is in this context that the use of translation as a testing strategy becomes important. Translations were originally a part of the Translation method a language teaching. But scholars have questioned the appropriateness of translation on the ground that "translation is a highly complex skill requiring special talent and special training". 22 It is not only translation that is a specialized art but writing o essays and composition is also an art requiring special talent as they involve the organization of thoughts and ideas in a sequentially arranged manner, in addition to the appropriate use of language. According to Nickel, translation has greater value from a diagnostic point of view than some other tests because translation is a complex activity.23

Experiments have shown that a very close correlation exists between the marks given for written translation and oral performance in the target language than what is generally thought of. The use of translation is argued to be more useful in the testing of higher order skills. Translations will have to be given in defined cultural context pertaining to the target language culture, the reason being that they should correspond to real life situations. The translation tests are more reliable, valid and objective if the context is as simple as possible.

With the developments in the field of language teaching, the newer methods have not restricted the use of target language only in the course of instruction but have included the use of the mothertongue of the learners or may be a language commonly known both to the instructors as well as learners during the course of instruction. It has been established that the use of mother tongue elements which includes the presentations of the target language rules does not seem to interfere with the acquisition of the target language. Thus a strong argument against the use of mother tongue as a meta language in second/foreign language teaching has lost its importance. However, this argument does not support the active and more intensive use of mother tongue in a second/foreign language teaching.

From the contrastive point of view, the need to use the mother tongue in acquiring a second language may reinforce the learner's knowledge of rules of the target language. One of the objectives of language testing is to test the amount of interference that may take place between the mother tongue and the target language. Needless to say that the degree of interference between the source language learners but it will differ from one group to the other if not from one individual to the other. The technique of translation as a testing device involving to the other. The technique of translation as a testing device involving the translation of sentence in the mother tongue into the target language has certain advantages over other pure target language tests involving completion, transformation etc.

In spite of the arguments in favour of translations being used as a testing technique, scholars like Lado have questioned the appropriateness and validity of translation in the context of language testing. Lado points out that if translation is used as a test of speaking, the problem of face validity arises as translation is something different from speaking. He also raises the difference between the immediate translation and delayed translation.Oller has pointed out that research has shown that the kinds of errors people make in translating from the native language into the target language are precisely analogues to the kinds of errors they make in spontaneous speech in the language, and also in imitating long sequences of information in the target language.

In he light of the above discussion, we may conclude that the use of translation as a technique in language testing is still controversial. If this is used in the context of language testing, the examiner must ensure the assignment of appropriate weightages to various factors ranging from grammar, lexicon and the fluency of style etc., depending upon the level of achievement that the learners are expected to attain at the end of instruction. Perhaps it is needless to say tat priority should be given to acceptability over communicability at higher levels. As already mentioned, the question of using translation in language tests is still controversial and needs further thought and research.