Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 5

Name: Le Thuy Linh, Phan Thi Ngoc Linh

LANGUAGE ASSESSMENT REFLECTION

1. Assessment targets

It goes without saying that learning targets are the heart of assessment for learning, which
needed to make clear right from the planning stage. Based on the test matrix, we can
conclude the assessment targets of the test. First, by the end of the assessment, students are
expected to understand the vocabulary knowledge enough to be able to use that word in
appropriate context. Second, students can identify the specific grammar concepts to
distinguish between the use of past simple and past continuous tense, etc. Third, students are
able to apply the knowledge of comparison, reported speech, modal verbs, adverbial clause of
time, conditional sentences to transform or combine sentences. Fourth, students can
understand the key points of the provided reading passage, identify the writer's point of view,
use scanning skills to find specific information, and skimming skills to establish the main
ideas. Last, students are expected to identify how to pronounce words with -ed/-s/-es endings.

Comparing to the objectives of the Ministry of Education and Training, it seems to match
with the provided requirements on assessment for students such as understanding the
descriptions and meanings of the most important elements and events in different texts,
browsing the texts (stories, news, abstracts, messages, advertisements, etc.) to find relevant
information, etc. However, in terms of language use, the test matrix does not mention
vocabulary of which topics, what aspects of vocabulary such as form, meaning, use will be
assessed, which leads to the unclearness of the target of vocabulary assessment.

3. Assessment methods and question types

The assessment methods applied in this standardized test is selected-response assessments,


especially multiple-choice questions, which require students to select the best answer from
among a set of four options. This kind of method more or less matches the appropriateness
and feasibility, due to a wide range of knowledge included in the MCQ test. It seems to be
beneficial for teachers to assess the whole knowledge they have taught, such as phrasal verbs,
conditional type 3, double comparison, reported speech, passive voice, etc. Moreover, by
taking an MCQ test, students need to consider four options seriously before coming to an
inclusion, which leads to the assessment of the ability to analyse the options.

On the other hand, this type of test has some negative aspects. First, it is difficult to assess the
real ability of students due to the lucky/ unlucky factor. Many students are not as good at
English as others; however, they random the options and fortunately they are correct. This is
unfair for others! Besides, in MCQ tests, teachers cannot assess students' writing skills -
which is very essential in learning English. From my perspective, instead of the task “Choose
the sentence that is closest in meaning to the one given”, we can ask students to rewrite the
sentences without changing the meaning. Based on this task, teachers have an opportunity to
assess their ability to produce language and apply the structures. Regarding the task “choose
the underlined part that needs correcting”, it is another disadvantage of MCQ test. This task
does not require students to correct the mistakes, which cannot avoid the guessing factor.

4. The quality of test task components (instruction, input, expected outcomes)

Considering the validity, we find that this final test does not have high content validity. In
terms of content validity, the test can cover most of the content that students have learnt in
2nd semester of English 12 in 7- year programme such as pronunciation of -ed/ -s, passive
voice, reported speech, conditional sentence. But there are some language use knowledge that
are assessed without being taught for students in this semester of grade 12. Gerund and
infinitive following ‘remember’, ‘forget’ are examples. Although having learnt about them in
English 10 and English 11, students do not access to a new aspect of those grammatical items
or have a revision but they are still assessed, which can reduce the validity of the test.
Another example is brand- new phrasal verbs which might overwhelm students as they do not
appear in the textbook. When choices in a MCQ include two, even four new words, students
are not able to deduce the meaning of those words, so it makes room for guessing factor. That
not only leads to low validity of the test but also causes the disadvantage of assessing
instrument.

In addition, this test is not highly valid in terms of construct. It can be used to assess two
language competences: grammatical and textual competences, according to Bachman’s
Communicative Language Ability Framework (1990). Students are encouraged to develop
these competences in their learning process, and they are tested the same competences, which
seems to create high construct validity for the test. However, this quality of the test is lowered
by indirect testing way in questions about language function ( question 4) and writing. While
the assessing target in question 4 is to use appropriate language to respond to other’s opinion,
the assessing method used is selected response, which allows high guessing factor and cannot
measure fluency and accuracy of learners precisely. Similarly, MCQ is used in writing task,
to measure students’ ability to transform and combine sentences. Because of the objectivity
of MCQ, it cannot reflect how accurately students produce the language in writing. Instead,
constructed response might be the better choice to raise the quality of construct validity in
this writing task. Students might be required to rewrite the sentences. Hints and words
limitation can be provided, and students have to write down their answer, to directly perform
the task.

Reliability:
- Reading text is too short for 12th grade students
- Level of the reading text (A1) is lower than required level (A2)
- There are no distractions among 4 options ( q10,15, 39)
- cannot discriminate students at different level
- Dang bai pho bien
- Correcting mistake: under level

Reliability is another quality of the test that cannot be neglected. In our opinion, 40 questions
of the test are designed for students to perform in 45 minutes, which is reasonable. The tasks
are familiar and student- friendly. Beside those good points, there exists some weaknesses in
reliability. Overall, in our opinion, this test can hardly discriminate students at different
levels. Although the level of students who take this test is not mentioned and the assessing
purpose is achievement assessment, they are students in grade 12 and expected outcome for
them is level A2. The highest level tested is level 3- understanding, in addition to the format
of MCQ, this test is not challenging enough to discriminate students, to exactly evaluate those
in top of class. Considering reading task, we find out that the reading text has 207 words, and
it is too short in comparison with the length of a standard text for 12th grade students of the
7- year programme, which is 280- 320 words. Furthermore, while the aimed level that needs
assessing is A2, this text is only at A1 level. It cannot meet the requirement about level of
difficulty, so it is not challenging and reliable enough to assess reading skills. Another
question that has reliability problem is question 35 in the mistake correction task. This item is
used for assessing knowledge of adverb clause of time. In the matrix, this knowledge is
expected to be assessed in level 3- applying, but in the test, it is included in a task that only
demands students to identify the incorrect part in a sentence. In this situation, level of the task
is lower than expected level of assessment.
In addition, there is a mistake in scoring. The assessing instrument MCQ chose for the test
permits objective scoring. However, in the matrix and the list of assessing content, the score
allocation is not precise and consistent. In the matrix, score for 13 questions about language
use in level 1 is 2.75 and in the content list, score for 14 questions of part 3 is 3.25. They
should be 3.25 and 3.5 so that score for each question in each level or each part is equal, and
total score is standardized of 10. This mistake seems not much considerable, but it might
confuse scorers, and minimize the scoring reliability of the test as well.

5. Item writing techniques


- Order of tasks in the test: writing-reading- language use- writing- reading...
- Reading: sequence of questions does not follow the order of information appearing in
the text
- Consistency of 4 options: one option is much longer than others (q19, 26); include
both nouns and compound nouns (q24)
- Wording of instruction: sentence combination.
- không đánh dấu thứ tự các task
- Instruction k đồng đều
There is one more factor that contributes to the quality of a test. It is item writing technique.
Evaluating this test, we consider some features problematic. First of all, it can be seen
obviously that tasks are not numbered or marked in order. Although instructions are bold, the
test is not easy to follow. In some tasks, the instructions are not given successfully. The
instructions are lengthy and not on point, as the one of the first writing task. It can be shorter
and simpler. Task demands are not consistent, either. We mean the demand of correction task
uses the cognitive verb ‘show’ while others use the verb ‘indicate’. Students may be not
patient enough to read and analyze all of these instructions as they are familiar with those
kinds of tasks. But we are evaluating the test in every detail, we cannot ignore this drawback.
There is a lack of organization as well. The tasks that belong to one knowledge or skill are
separated and not sequenced by any logical orders. Between two writing tasks are reading
and vocabulary ones. Then another reading task appears, and after it is three questions of
mistake corrections and two of phonology which should go with the very first task in the
task. This disorder can cause interruption and distraction to students. Disorder problem also
rises in passage comprehension task.The order of questions do not follow the sequence of
information in the text. It is hard to follow and time- consuming for students when doing this
task.

6. Authenticity of test tasks


Authenticity is one of the important factors that enables teachers to assess more accurately
how students solve the language problem in real-life context. However, the real-life situations
are not highly evaluated in this test. Particularly, the writing task shows the low authenticity,
because it focuses on assessing students’ ability to combine the sentences and rewrite the
sentence without changing the meaning, while the writing objective of the Ministry of
Education and Training is that students have the ability to write a paragraph of about 130-150
words about the content related to the topics students have learned or are familiar to.
Moreover, sentence 4 in the test, which is about language function knowledge, is also not
authentic, as the content does not match what students are taught. Students just learnt some
easy expressions of agreement such as “I agree with you completely”, “Absolutely! That’s
right!”, etc, but the phrases “There’s no doubt about it!”.

In conclusion, this test mostly has disadvantages in the abuse of multiple-choice questions,
the structure of reading task, the low authenticity and validity of test tasks, problematic item
writing techniques, etc. Therefore, we believe that it is necessary to make some appropriate
improvements for this test such as adding the writing task, editing the sequence of questions
of reading tasks, etc.
Contribution report

Name (Class 16E5) Covered criteria

Le Thuy Linh Assessment target

Assessing method and


instrument

Authenticity

Phan Thi Ngoc Linh Validity

Reliability

Item writing techniques

You might also like