Kazakhstan-American free university the problem of testing in grammar assessment
К содержанию номера журнала: Вестник КАСУ №2 - 2010
Автор: Елаков В.В.
The study of grammar has had a long and
important role in history of foreign language teaching. Among all words used in
a classroom there is the only word that usually makes the students shudder:
“test”. There is hardly a person who would claim that a student likes tests and
finds them very motivating. Many scientists consider that the role of tests is
very useful and important, especially in language learning. It is a means to
show both the students and the teacher how much the learners have learned
during a course.
Language testing today reflects current
interest in teaching genuine communication, but it also reflects earlier
concerns for scientifically sound tests. Testing during the last century was
basically intuitive, or subjective and dependent on the personal impression of
teachers. Grammar skills can be measured by means of grammatical tests. Today
testing becomes popular in checking knowledge in different spheres. Many
scientists have been studying tests and the process of testing for a long time.
For example, Harold S. Madsen studied testing techniques, James Purpura studied
grammar assessment, and J. Alderson studied test construction. (Madsen, 1999;
Purpura, 2002; Alderson, 2001)
Test is any standardized procedure for
measuring sensitivity or memory or intelligence or aptitude or personality etc;
"the test was standardized on a large sample of students". Testing in
education is a systematic way of studying the ability of an individual or a
group to solve problems or perform task under controlled conditions. It is an
attempt to measure a person’s knowledge, intelligence or other characteristics
in a systematic way. Teacher gives the test to see how well students have
learned a particular subject or grammar. Tests imply the use of any training or
supervising exercises resulted in sections of training to aspects of language
and kinds of speech activity. To pick up corresponding exercises, it is
necessary to correlate carefully the purposes and objects of testing, its
criteria and parameters.
Productive competences testing, like
speaking exams, requires active creative answers, while receptive competences,
like multiplechoice reading tests, tend to rely on recognition, with students
simply choosing the letter of the best answers. Tests of language subskills
measure the separate components of English, such as vocabulary, grammar and
pronunciation. Communication skills tests, on the other hand, show how well
students can use the language in actually exchanging ideas and information.
Another set of contrasting tests is that of
norm-referenced and criterion-referenced exams. Norm-referenced tests compare
each student with his classmates. But criterion-referenced exams rate students
against certain standards, regardless of how other students do. Still another
pair of categories is that of discrete-point and integrative tests. In discrete-point
tests, each item tests something very specific such as a preposition or a
vocabulary item. Integrative tests are those that combine various language
skills.
Tests cannot be avoided completely, for
they are inevitable elements of learning process. They are included in
curriculum at schools and are to check the students’ level of knowledge and
what they are able to do; they could be accomplished at the beginning of the
academic year and at the end of it; the students could be tested after working
on new topics and acquiring new vocabulary. Moreover, the students are to face
the tests in order to enter any foreign university or define the level of their
English language skills.
It is often conventionally assumed that
tests are mostly used for assessment. The test gives a score which is assumed
to define the level of knowledge of the tester. This may be in order to decide
whether the student is suitable for a certain job or admission to an institution,
has passed a course, etc. But in fact testing and assessment overlap only
partially: there are other ways of students’ assessment and there are certainly
other reasons for testing. Test may be used as a means to:
- give the teacher information about where
the students are at the moment, to help decide what to teach next;
- give the students information about what
they know, so that they also have an awareness of what they need to learn or review;
- motivate students to learn or review
specific material;
- get a noisy class to keep quiet and
concentrated;
- get students to make an effort, which is
likely to lead to better results and a feeling of satisfaction;
- give students tasks which themselves may
actually provide useful review or practice, as well as listening;
- provide students with a sense of achievement
and progress in their learning (Ur, 2003).
However, very often the test itself can
provoke the failure of the students to complete it. With the respect to the
linguists, we are able to state that there are two main causes of the test
being inaccurate:
- Test content and techniques;
- Lack of reliability.
The first one means that the test design
should response to what is being tested. First, the test must content the exact
material that is to be tested. Second, the activities, or techniques, used in
the test should be adequate and relevant to what is being tested. They should
not frustrate the learners, but, on the contrary, facilitate and help the
students take a test successfully.
The next one implies that one and the same
test given at a different time must score the same points. The results should
not be different because of the shift in time. For example, the test cannot be
called reliable if the score received after the test completed for the first
time differs from that completed for the second time. Furthermore, reliability
can fail due to the improper design of a test (unclear instructions and
questions, etc.) and due to the ways it is scored.
The tests can facilitate the students’ acquisition
process and function as a tool to increase their motivation; however, too much of
testing could be disastrous changing entirely the students’ attitude towards
learning the language, especially if the results are usually dissatisfying.
Assessment is an important aspect for the teacher and the students, and it
should be correlated in order to make it “go hand in hand”.
The test should be valid and reliable. They
should test what was taught, taking the learner’s individual pace into account.
Moreover, the instructions of the test should be unambiguous. Validity deals
with what is tested and degree to which a test measures what is supposed to
measure. Reliability shows that the results of the test will be similar and
will not change if one and the same test will be given on various days. The
test should be practical, or in other words, efficient. It should be easily
understood by the examinee, easily scored and carried out. It should not last
for eternity, for both examiner and examinee could become tired, for instance
during a five-hour non-stop testing process. Moreover, while testing the
students the teachers should be aware of the fact that as well as checking
their knowledge the test can influence the students negatively. Therefore,
teachers ought to design such a test that could encourage students. The test
should be a friend, not an enemy. Thus, the issue of validity and reliability
is very essential in creating a good test. The test should measure what it is
supposed to measure, but not the knowledge beyond the students’ abilities.
Moreover, the test will be a true indicator whether the learning process and
the teacher’s work is effective (Alderson, 1995).
Language consists of its vocabulary
structure and its grammar structure. The vocabulary doesn’t constitute the
language, but acquire the tremendous significance when it comes to be governed
by the grammar of any language. Grammar, the structural glue, the “code” of
language, is arguably at the heart of language use, whether this involves
speaking, listening, reading or writing (Azar, 1998).
For the past fifty years, grammar competence
has been defined in many cases as morphosyntactic form and tested in either a
discrete-point, selected-response format – a practice initiated by several
large language testing firms and emulated by classroom teachers, or in a
discrete - point, limited-production format, typically by means of the cloze or
some other gap-filling tasks. These tests have typically been scored right /
wrong with grammar accuracy as the sole criterion for correctness. Tests of
this kind are appropriate for certain purposes and make sense, for example, in
situation where individual grammatical forms are emphasized, such as in
form-focused instruction. However, we must recognize that separate tests of
explicit grammatical knowledge provide only a partial measure of grammar
competence, and scores from these tests might be related to those produced from
more comprehensive measures of grammar competence.
In recent years, the assessment of grammar
competence has taken an interesting turn in certain situations. Grammar has
been assessed in the context of language use under the rubric of testing
speaking or writing. This has led, in some cases, to examinations in which
grammatical knowledge is no longer included as a separate and explicit
component of language in the form of separate subtest. In other words, only the
students’ implicit knowledge of grammar as well as other components of
communicative language ability (e.g. topic, organization, register) is measured
(Purpura, 2002). Even with the sudden increase of research since the middle
1980s on grammar teaching and learning, there still remains a surprising lack
of consensus on:
- Grammatical knowledge;
- What type of assessment task might best
allow teachers ad testers to infer that grammatical knowledge has been
acquired;
- How to design task that elicit students’
grammatical knowledge for some specific assessment purpose, while at the same
time proving reliable and valid measures of performance (Purpura, 2002).
Tests are an attempt to test what is called
expectancy grammar. The context is now at the paragraph or discourse level and
meaning comes into play. In spite of some claims to the contrary, cloze
techniques do not test the ability to use the language. There have been other
suggestions for how to test communicative grammar but none of them have gained
wide acceptance. This may be because it is impossible to measure communicative
grammar directly. Rea Dickens says that in order to measure communicative
grammar a test must have five characteristics.
- The test must provide more context than
only a single sentence.
- The test taker should understand what the
communicative purpose of the task is.
- He or she should also know who the
intended audience is.
- He or she must have to focus on meaning
and not only form to answer correctly.
- Recognition is not sufficient. The test
taker must be able “to produce grammatical responses” (Trasher, 2000).
The term assessment is generally used to
refer to all activities teachers use to help students learn and to gauge
student progress. Though the notion of assessment is generally more complicated
than the following categories suggest, assessment is often divided for the sake
of convenience using the following distinctions:
- formative and summative;
- objective and subjective;
- referencing (criterion-referenced,
norm-referenced);
- informal and formal (Alderson, 2001).
Grammar tests are designed to measure
student proficiency in matter ranging from inflection to syntax. Syntax
involves the relationship of words in a sentence, including matters such as
word order, connectives. There are several reasons for testing grammar. Much
English Language teaching has been based on grammar; and unlike various
measures of communicative skills, there is general agreement on what to test.
Assessing the grammatical accuracy of a piece or written or spoken discourse
cannot be done fairly using a “count the number of mistakes” approach. But this
does not mean that it cannot be done. The test takers who opted to sacrifice
fluency for accuracy would be rated highly on the accuracy side but would be
given a low rating for the difficulty of the task that they attempted.
The test taker who took the opposite course
(sacrificed accuracy for fluency) would get a lower rating in accuracy but
would be rated highly on the difficulty side. Taking both fluency and accuracy
into consideration will be fairer than looking accuracy alone, but how can we
fairly assess grammatical accuracy? Should we merely count the number of
errors? Or should we consider both the number and the severity of the errors?
Assessing grammar by examining the accuracy
of what the test taker produces in a test of writing or speaking is not the
only approach that can be taken. Multiple-choice items can be used to measure
the ability to decide whether a grammar structure is correct or not. Such items
have been criticized because they do not measure the ability to produce
grammatically correct structures and therefore are claimed to be inauthentic language
tasks.
The ability to recognize mistakes in
grammar is a skill we utilize even when speaking our native language, so
multiple-choice items that tap this skill cannot be called inauthentic tasks.
Developing the same skill is one of our tasks in learning a second or a foreign
language. The skill allows us to monitor our production of the spoken language
or proofread what we write and know when to make appropriate repairs or
restatements. It is also possible to test grammar using fill in the blank type
items. In this sort of item the test taker must write in the best word to fill
the blank and put it in the appropriate grammatical form (Trasher, 2000).
Grammar items, such as auxiliary are spotted and counted. As for vocabulary
exams, either passive or active skills can be checked. Also grammar can be
tailored to beginners or advanced learners in testing grammar. We can do a good
job of measuring progress in a grammar class, and we can diagnose student needs
in this area (Madsen, 1999).
In developing grammar assessment, teachers
first articulate the purposes of test, consider the constructs and identify the
situational domain in which they would like to make inferences about the
test-takers’ grammatical ability. As the goal of grammar assessment is to
provide a measurement as useful as possible of students’ grammatical ability,
teachers need to design test task in which the variability of students’ scores
is attributed to the differences in their grammatical ability, and not to
uncontrolled or irrelevant variability resulting from the types of tasks or the
quality of the tasks that teachers have put on our tests. As all teachers know,
the kinds of tasks they use in test and their quality can greatly influence how
students will perform. In order words, specifically designed tasks will work to
produce the types of variability in test scores that can be attributed to the
underlying constructs giving the contexts in which they are measured.
It is always easier to correct someone
else’s written work than one’s own. Using this generalization as a starting
point, we can propose three peer correction strategies that are useful when the
teacher wants students to focus on correcting specific grammatical errors
during the editing process.
We can suggest that paragraphs or short
papers by students be used whenever possible and that papers be selected
because they illustrate frequent errors types, such as substitution of the
present perfect for the simple past tense or overuse of the infinitive to +
verb after modals. To ensure maximum focus, the teacher may even correct all
other errors and ask students to do specific problem-solving correction
activities, such as, “Find two verbs that need an s to show the present tense”,
“Find four nouns that should take the definite article”, etc. We recommend using
explicit grammatical terminology with the students, especially in classes where
students already know it, although example errors and corrections or informal
circumlocutions can also be used. All exercises presuppose that grammar points
focused on the activity have already been covered.
An alternative of this use of students’
paragraph or essay as sources of peer correction materials is to prepare a
composite essay or test for group correction that illustrates similar errors
from several students’ written work. This avoids embarrassment by focusing on
common problems. We can use this procedure, for example, to practice correction
of tense and modal errors in conditional sentences with good results.
In large classes in which students write a
lot, the teacher cannot correct everything. Instead, they can take an
individualized approach by using a method called “the blue sheet”. In this
approach the teacher attaches a blue sheet to the paragraph, essay, and test,
lists two obvious structural errors made by the student, and refers the student
to pages and exercises in the class grammar text pertinent to the two errors.
Students do the assigned exercises when they get their blue sheets; the teacher
then corrects them before the students rewrite their passages. The same
approach could be used to identify and correct specific grammatical errors that
the teacher has detected with some frequency in the students’ work.
Even in smaller classes, not every error
needs to be corrected on every paper. In fact, such overkill tends to
discourage students and thus impedes progress in the long run. The best results
are achieved by focusing on one or two at a time, at least in the beginning.
An individualized checklist encourages
students to focus when they edit and correct their own work. Its use is the
next logical step in the grammar editing process and works best if, at first,
short pieces of writing are used. When the teacher returns the first draft,
grammar errors are underlined and major areas of difficulty are listed or checked
off on the attached checklist. In this way, each student is aware of the errors
they could correct as well as the location of the errors.
Students should consult with the teacher or
tutor if they do not understand what the errors are or how to correct them.
Each student should keep these checklists and all the drafts of each writing
assignment together in a notebook or folder. The teacher can have individual
conferences and, where necessary, refer students to additional exercises for
those grammar areas in which errors are most persistent. Student soon become
adept at making such targeted corrections. As they progress, they should be
asked to correct several structures already covered in class in passage in
which the errors are only minimally indicated; for example, by underlining.
Once the class (or the student) does not get very good at editing by this
means, neither the location nor the error type should be specified.
Some students respond better to a less
judgmental correction procedure in which the teacher or tutor merely specifies
rewordings for sentences or phrases that contain grammatical errors on an
attached sheet. The best results are achieved if the teacher moves gradually
from much focused correction procedures to less focused ones. Whichever error
correction strategy is used, it is imperative that student incorporate the
corrections and become aware of major grammatical problems. Over a period of
time, these correction strategies, combined with systematic grammar instruction,
have a positive effect on the accuracy of the writing produced by ESL students
(Witbeck, 1976).
Finally, the teacher should not give the
tasks studied in the classroom for the test. It can be explained it by the
fact, that when testing we need to learn about the students’ progress, but not
to check what they remember. Designing a test is not so fearful and hard as
many teachers think. When working out grammar tests, the teacher has to make
decisions about such factors as grading, the degree of control, and the degree
of realism.
BIBLIOGRAPHY
1. Alderson J. The test of
English grammar. – Cambridge, 2001.
2. Alderson J., Clapham C.
Language Test Construction and Evaluation. – Cambridge, 1995.
3. Azar B. S. Understanding
and using English Grammar. – Oxford, 1998.
4. Madsen S.H. Techniques
in testing. – USA, 1999.
5. Purpura J. E. Assessing
Grammar. – Oxford, 2002.
6. Trasher R. Test theory
and test design. – International Christian University, 2000.
7. Ur P. A course of
language teaching. – Cambridge, 2003.
8. Witbeck J. Text
editing and grammar correction. – Cambridge, 1976.
К содержанию номера журнала: Вестник КАСУ №2 - 2010
|