The Approach to Assessing Listening 1

THE APPROACH TO ASSESSING LISTENING

Composed by:
Samsudin R. Ishak and Irvan S. Humonggio

A. INTRODUCTION
English teacher especially one in Indonesia always face problems regarding the teaching
of four language skills in terms of whether it should be taught in integrative or separately in
discrete point. At the same time, teacher should consider whether students‟ main objective
of learning English is to know the language or to use the language. Fortunately, the
Department of Education has provided guidance for teachers in these terms in curriculum.
However, the challenge is that how teachers can take their position in teaching English so
that students can both be successful in test as what has been expected by the government in
National Examination, and be successfully comprehend English in communicative ways.
In consideration of successful English teaching, it is important for teachers to put an
emphasis in process of teaching itself. Now, the question is „how can teacher determine the
success of their teaching?‟. In this point, Brown (2004) states that teachers should
understand three interrelated concepts namely testing, assessment and teaching. Teaching as
Brown says, is the broadest area in this interrelated concept where in teaching there should
be an assessment to measure the success or achievement students make after learning. There
many ways of holding an assessment, and test is one of those ways of measurement.
Assessment itself can be held by observing the process and the product through learning
activities in four skills. However, how to observe such result when both process and product
could not be directly observed? It exists in two of four skills that are receptive (reading and
listening) where it is difficult for teacher to determine whether or not students receive or
obtain the information.
This paper will focus on discussing how to assess listening skill rather than adding one
with reading skill because listening is an early step for students to get their self into
communicative English. There will be some issues regarding approaches to assessing
listening discussed in this paper namely: Discrete-point test, integrative testing, and
communicative testing. Those are taken from main readings of listening assessment by
Brown (2004) and Buck (2001).
 The Approach to Assessing Listening 2

B. DISCUSSION
This part is the summary of listening assessment taken from Brown (20014: pp. 1-18;
116-139) and Buck (2001: pp. 61-93). The discussion will be focused on the approaches to
assessing listening including the kinds of tests involved in such approaches including the
procedures of assessment. The application of the assessment in classroom will be discussed
in next task.
Brown (2004: pp. 7-11) elaborates the common approaches to language testing into four
terms, namely: (1) discrete point testing; (2) integrative testing; (3) Communicative testing;
and (4) performance based assessment. Especially for listening assessment discussed here,
the last term is omitted in consideration of the receptive characteristic of the skill.
Furthermore, Brown (2004: pp. 120) explains that there are at least four types of listening
performance listed: intensive, responsive, selective and extensive listening.
1. Discrete-point testing
“Discrete point test are constructed on assumption that language can be broken down
into its component parts and that those parts can be tested successfully” (Brown, 2004:
p. 8). It entails the involvement of four language skills (listening, speaking, reading and
writing) on the test with discrete linguistic points such as phonology, morphology,
lexicon, syntax and discourse. Below are three discrete point based test designs taken
from Buck (2001)
a. Phonemic discrimination test
In this design, test taker listen to one word spoken in isolation and have to identify
which word they have heard. The phonemic pair can be in consonants, vowel or in one-
word stimulus as directly quoted from Brown (2004: pp. 123-124) below:
Phonemic pair, consonants
 The Approach to Assessing Listening 3

Phonemic pair, vowels

One-word stimulus

Buck states that this kind of test is not fashionable now because it is not natural. In
real communication, students not only use phonetic information to determine what
was said, but also use the context.
b. Paraphrase recognition
Test takers listen to a statement as a stimulus, and then choose the correct paraphrase
form. This kind of test can use sentence or dialogue as the stimuli:
Sentence Paraphrase:
Test takers hear: John ran into a classmate on his way to the library
Test takers read:
(a) John exercised with his classmate
(b) John ran to the library
(c) John injured his classmate with his car
(d) John unexpectedly met a classmate

Dialogue Paraphrase:
 The Approach to Assessing Listening 4

c. Response evaluation
Test takers listen to question followed by four possible responses and choose most
appropriate response. Brown also suggests an open-ended response which in fact has
problem with its practicality because of spending more time for teacher to give
score.
Test taker hear: how much time did you spend in Boston?
They read:
(a) Yes, I did
(b) Almost $250
(c) Yes, I had to
(d) About four days

2. Integrative testing
Supporters of this approach argue that language skills cannot be tested separately as
what has been formulated in discrete-point testing because language competence is a unified
set to be used in communicative interaction. It puts the emphasis on assessing the processing
of language as opposed to assessing knowledge about the elements of language. There are
seven designs of integrative testing elaborated by Buck (2001) as summarized below:
a. Noise tests
In this kind of test, test takers will hear a recording that is mutilated by the addition
of background noise that is called white noise. It is aimed to create a natural setting of
real life communication. Unfortunately, many test takers dislike it because the noise is
more being annoying rather than a naturalistic background noise.
b. Listening cloze
This is the most common design used in our classroom. It is suggested that in
listening cloze, text is selected and words are replaced by blanks in a systematic way,
usually every 5th, 7th or 10th words. Buck (2001) quoted the study of Templeton (1977)
who proved that this kind of test is highly reliable because of having little administration
time and the items is easy to write. In his study, Templeton made recordings of material
in two ways. First, the text is recorded with a beep sound in every 15th word in order to
mask the word and create a blank. In the second recording, he made a pause in every
syntactic boundary after the beep sound. The pauses can be a chance for test takers to fill
the blanks with the appropriate/exact words they have already heard. The scoring of this
 The Approach to Assessing Listening 5

test is also simple by summarizing the correct answers test takers got and counting the
average value of it to get a grade.
c. Information transfer
Brown (2004) suggests information transfer as the way of assessing listening
performance in “selective listening”; in which test takers must look for specific
information from limited text they hear. This technique involves listening cloze (as
previously explained), multiple picture-cued selection, single picture-cued verbal-
multiple choice, and chart filling.
In multiple picture-cued selections, test takers have to observe multiple pictures
provided at their sheet and then choose the correct picture based on the monologue they
hear. In contrary, single picture-cued verbal-multiple choice test takers will observe
single picture and then hear a question and its multiple choice to choose the correct
answer. The scoring of both tests is same to multiple choice scoring.
In chart-filling test, test takers hear the information and then complete the partially
completed chart provided in their test book. The scoring may depends on the total
number of information test takers should fill in the chart. Nation and Newton (2009)
suggest that information transfer technique may include family tree completion, in
which test takers will hear a monologue telling about family and then complete the
family tree provided.
d. Gap-filling
This kind of design is similar to cloze test but the blanks that are created in the text is
put randomly based on assessment need as opposed to orderly created blanks in Cloze
test. Buck (2001) claims that the problem of this test is the test takers will treat the test
as a common reading test because in fact the blanks is possible to be filled even without
playing recording.
Another design of gap-filling that is suggested is gap-filling on summaries. The test
takers will read a summary of the passage on their test sheet, in which some important
words have been replaced by blanks. Then, they hear a recording of the original passage,
note taking the important information and writing down the appropriate word on the
blanks based on their understanding of the text.
It is claimed by Lewkowicz (1991) as cited by Buck (2001) that there will be two
dangers in using gap filling in summary: (1) test takers could fill in the blanks without
 The Approach to Assessing Listening 6

understanding the original passage on the recording; and (2) it is difficult to determine
what information should appropriately placed to the blanks. It is suggested that this
technique can be firstly pre-tested before it is used at classroom.
e. Dictation
Dictations have been widely used for decades. It is an integrative test because it
entails students‟ understanding to meaning rather than to specific aspect of language.
Buck (2001) suggests that in dictation, students will hear a passage twice: first at normal
speed; then, with a pause within a segment during which time students write what they
have heard. For intermediate level, Brown (2004) adds that it is important to recite the
passage at normal speed after the second recite to give a chance for students to confirm
their answer.
Dictation can be scored by the total available score based on total number of words
in each segment. However, Buck (2001) suggests that “the best way of scoring dictation
is to delete marks, starting with the number of words for each segment; one mark is
subtracted for each mistake. Intrusions count as one mistake each, as do omission”
This way of scoring, in fact, is difficult to be applied for low-ability test takers
because they make many mistakes. Hence, Hughes in Buck (2001) suggested using
partial dictation as presented in example below:

f. Sentence repetition tasks

This test is similar to dictation, except that students repeat the text orally during the
pause between each section. Students hear each sentence once and repeat it back
 The Approach to Assessing Listening 7

immediately after they have heard it. The responses are usually tape-recorded and then
scored later.
g. Statement evaluation
Instead of asking students to repeat the sentence, listening skill can be assessed by
doing statement evaluation. The procedure is that (1) students are given a statement and
then evaluate whether the statement is true or not; or (2) they are given two statements
and then decide whether or not they mean the same thing.
Example:
Test takers hear:
The sun rises in the west
Or
Test takers hear:
a. The snow delayed the flight
b. The plane arrived on time despite the snowstorm.

h. Translation
The procedure is same to dictation, except students write down what they have heard
in their first language. The students listen to the recording, and during the pause they
write the translation of the sentence they have already heard.
3. Communicative testing
a. Communicative stimulus-response tasks
This kind of the test is commonly used in National Examination Listening test, in
which “test takers are given a stimulus monologue or conversation and then is asked to
respond to a set of comprehension questions” (Brown, 2004: p133). This includes
dialogue and multiple-choice comprehension items, and dialogue and authentic
questions on details.
In dialogue and multiple-choice comprehension items, test takers will hear a
dialogue and then read the question and multiple-choice on their test book and choose
the correct answer based on the dialogue. The second technique is similar, except for the
questions that require test takers complete understanding toward the dialogue. It requires
an inference ability of test takers.
 The Approach to Assessing Listening 8

b. Authentic listening tasks

A recent issue in listening assessment is that whether or not the text or information
used in listening test can help test takers (students) in real life. It is assumed that in some
way listening material/recording is intentionally created in order to measure specific
language feature mastered by test takers. However, it seems artificial contrast to what we
find in real life context. By this consideration, Field (2008: p.281) suggests the use of
authentic recording “it means that authentic recordings should be taken to cover a wide
spectrum of styles from formal (an interview) and even „read aloud‟ (the
radio news) to very informal (a pub conversation).” There are four techniques suggested
by Brown (2004) related to authentic listening tasks:
1. Note taking: test takers are given a stimulus in form of lecture recording and then
are required to take a note from the lecture. Brown proposes the rubric as
follows:

It is important for teachers that use this technique to consider the practicality of
note taking. The scoring is time consuming and may be affected by subjectivity
of teachers. Multiple rater assessment is needed.
2. Editing: test takers will read a written stimulus and then listen to a spoken one to
do a probe for discrepancy. Brown (2004) explains:
 The Approach to Assessing Listening 9

3. Interpretive tasks: this task requires test takers‟ interpretation toward the
material. It may include song lyrics, recited poetry, news report, and an oral
account of an experience. Test takers then are required to answer the
comprehension questions in open-ended form.
4. Retelling: test takers listen to recording of a story or news events and then retell
or summarize it either orally (on an audiotape) or in writing. It has less
practicality as note-taking because of its time consuming scoring process.

C. APPLICATION (This part will be presented separately from the paper)

D. CONCLUSION
The listening assessment technique having been summarized in this paper will help
teacher in designing test. Teachers may consider which kind of test will meet their need at
school. It might be possible for them to use Discrete-point test to separately measure
language aspect they have been teaching, integrative testing for confirming the meaning
rather than language features, or communicative testing to build authentic assessment
connected to real life practice. However, the issues of practicality and authenticity as
explained above should be carefully considered, adjusted to teachers need at school.

References:
Brown, D. (2004). Language Assessment: Principle and Classroom Practices. California:
Longman
Buck, G. (2001). Assessing Listening. Cambridge: Cambridge University Press
Field, J. (2008). Listening in the Language Classroom. Cambridge: Cambridge University
Press
Nation, I.S.P & J. Newton. (2009). Teaching ESL/EFL Listening and Speaking. New York
& London: Routledge