Speaking test and some application to vietnamese high school pupil = kiểm tra kỹ năng nói và một số ứng dụng đối với học sinh PTTH ở việt nam

Student-examiner Information Gap CHAPTER 3: SOME APPLICATIONS TO TESTING SPEAKING SKILL OF 10 TH FORM STUDENTS 3.1.. Sample Test for Testing Speaking Skills of the 10 th Form Students P

vinh university department of foreign languages

DƯƠNG THị HƯƠNG

speaking test and some applications

to Vietnamese High School pupils (KIểM TRA Kĩ NĂNG NóI Và MộT Số ứng dụng

đối với học sinh ptth ở Việt Nam )

I would like to thank all those who have contributed indifferent ways to my thesis

First of all, I would like to express my deepest thank to all of

my teachers of English at Foreign Language Department,especially, Mrs Le Thi Khanh Khang, my supervisor, who gave me

so much help during my studying

I also acknowledge all of my friends who were very kind togive a lot of useful ideas and lend me some precise documents,which were used in my thesis

Finally, i indeed thank my family for their encouragementand help

Without these help and enthusiastic encouragement, I couldnot have finished my graduation thesis

Vinh, May, 2005

D¬ng ThÞ H¬ng

Table of contents

1 Acknowledgement

Part A: introduction

1 Reasons For Choosing the Topic

2 Aims of The Study

3 Methods of the Study

4 Scope of the Study

5 Design of the Study

1.1.1.2 Useful Features of C.L.T Applied in Speaking Testing

1.1.2 Testing Speaking Skills

1.1 2.1: What is a Test?

1.1.2.2: What is a Good Test?

1.1.2.3 What is a Speaking Test?

1.1.2 4: Why do We Test?

1.1 25: Characteristics of a Goof Speaking Test?

CHAPTER 2: TECHNIQUES FOR SPAKING TESTS

2.1 What do we test?

2.1.1 Testing Low Level Students

2.1 2 Testing Higher Level Students

2.2 Rating Scales of Oral Testing Criteria

2.3 Procedures of Building a Speaking Test

2.3.1: Some Suggestions For Building a Speaking Test

2.3.2: Stages of Test Construction

` 2.3.2.1: Statement of the Problem

2.3.2.2 Providing a Solution to the Problem

2.4 Suggested Techniques

2.4.1 Role-play

2.4.2 Oral Presentation

2.4.3 Oral Interview

2.4.4 Using Visual Material

2.2.4.5 Student-student Interaction

2.4.6 Student-examiner Information Gap

CHAPTER 3: SOME APPLICATIONS TO TESTING SPEAKING SKILL

OF 10 TH FORM STUDENTS 3.1 Problems of Testing Speaking in Vietnam

3.2.Types of speaking test Applied in Vietnamese High Schools

3.2 Sample Test for Testing Speaking Skills of the 10 th Form Students

PART C: CONCLUSION

REFERENCES

Part A:

Introduction

1 reasons for choosing the topic:

Most of us would agree that the oral skills are the most importantones, especially speaking skill, of all four skills (listening; writing;reading and speaking) People who know a language are considered as

“speakers" of that language So, speaking has always been considered

as key-skill for language learner And speaking test is obviously anintrinsic part of teaching and learning foreign language in general andspeaking skill in particular

However, speaking skills are the most difficult to test It is easyenough to get an impression, and probably a fairly accurate one fromjust listening to students’ speaking, but it is very hard to make accurateassessment

In English classes in Vietnam, speaking skills have not beenmainly focused So people have not attached enough special importance

to speaking tests

As teachers of English language in the future, we allacknowledge that, speaking test has been part and parcel of languageteaching all along and will doubtless continue to be so It is importantand necessary to help my students develop their speaking skills andtesting their speaking ability Therefore, I decide to choose the topic

“speaking test and some applications to Vietnamese High School pupils” as my study.

2 Aims of the study:

The purposes of the study are:

- To show the importance of test in general and speaking tests inparticular

-To familiarize teachers with various techniques of testingspeaking skills and enable them to judge their effectiveness

-To enable them to choose suitable techniques to test theirstudents’ speaking skills effectively

3 Methods of the study:

In order to do the research effectively, I have used the followingmethods:

Statistic analysis: Collection and analization of data

*Observation: observation and pointing out the fact

*Explanation

*Synthesis and analysis: collection information from referencesand other sources

4 Scope of the study:

Testing is a wide-range field in language teaching and learning

In my study, I will not cover all spheres of testing I focus on testingspeaking skills, however, I will not study testing speaking in particular,

and I mostly focus on speaking test in Vietnam and its applications toVietnamese High School pupils.

5 Design of the study:

Apart from acknowledgements, table of contents, and reference,the thesis consists of three main parts

Part A: Introduction

1 Reasons for choosing the topic

2 Aims of the study

3 Methods of the study

4 Scope of the study

5 Design of the study

Part B: Content

Chapter 1: Theoretical background

Chapter 2: Techniques for speaking tests Chapter 3:Some applications to testing speaking skill of 10th formstudents

Part C: Conclusion

References

1.1.1 Communicative Language Teaching.

1 1.1.1: Features of Communicative Language Teaching

Recently, teaching methods has a tendency to followCommunicative Language Teaching This has the reason that, C.L.T hasmany useful features for teaching and learning foreign language.

Firstly, the goal of language learning is communicativecompetence It focuses on the ability of communicating of students,how they communicate in real-life-situations rather than theirknowledge of grammar or spelling

Secondly, the teacher has to develop tasks in which language andcommunication are independent

Thirstly, the teacher organizes dialogues and role-plays Theseactivities are considered as the most effective and popular way ofdeveloping the students’ communicative competence

The other typical feature of this method is that the teacherorganizes skits, debates (for advanced students) and they includeconversation and discussion section in the lessons They also use audio-visual aids in order to have their students compare sets of pictures, notesimilarities and differences, work out a likely sequence of events in aset of pictures missing features in a map or picture, or shape thefollowing instructions of a classmate, give instructions, followinstructions, and, in general, solve problems from shared clues

Also in this method, correction of errors is infrequent in order not

to hinder communication This avoids the stress and not to interrupt theflow of thinking of students

The last but not least important characteristic of C L T isthat ,the main role of the teacher in the class is to facilitate thecommunication between all participants in the classroom For the abovefeatures, the Communicative Language Teaching has been praised timeand again

1.1.1.2: Useful Features of C.L.T Applied in Speaking Testing.

C.L.T emphasizes that language is for communication So that,the testing and assessment has to develop students' ability to useEnglish in real settings and contexts In other words, a communicativeapproach to language testing emphasizes the ability to functioneffectively in particular settings and contexts It means that, a testshould involve linguistic activity of kinds and under conditionsapproximating real life According to Weir, it is important to apply the

following communicative characteristics in designing a communicativetest.

* Realistic context: The test tasks should be regarded asappropriate to the candidate's situation

* Relevant information gap: The candidates should have toprocess new information as they might in a real life situation

* Inter-subjectivity: The tasks should involve candidates both aslanguage receivers and language producers

* Scope for development of activity by the candidates: The tasksshould give candidates the chance to assert their communicativeindependence and allowance should also be made for the creativeunpredictability of communication in the tasks set and the markingschemes that are applied

* Allowance for self - monitoring by candidate The tasks shouldallow candidates to use discourse processing strategies to evaluatetheir communicative effectiveness and make any necessary adjustment

in the course of an event

* Processing appropriately - sized input: The size and scope ofthe activities should be such that candidates are processing the kind ofinput that they would normally be expected to process

* Normal time constraints operative: The task should beaccomplished under normal time constrain

The format of a “verbal essay" or a" talk on a given topic enables

a wide rage of criteria, including fluency, to be applied to the output.However, the topic specified may not be interest to some students, and

it may not be something we are normally asked to do extempore in reallife

Weir [8,74] also suggests other formats, that teachers may want

to try, for example, oral presentation, free interview, controlledinterview, role- play, and student - examiner information gap e.t.c

Applying the above-suggested features into communicativetesting, teacher can conduct a successful oral test

1.1.2: testing speaking skills

1.1.2.1: What is a Test?

According to Oxford dictionary (1995, p.1233): “a test is a shortexamination of knowledge or ability, consisting of questions that must

be carried out.” A test is also viewed as an activity whose main purpose

is to convey (usually to the tester) how well the testee knows can dosomething

Conventionally, tests are assumed to be used for measuring ourachievement against our personal goals, for proving our worth in theacademic stakes, for improving our career prospects They also have thepotential to change educational practices through their powerful “washback effect” on teaching and learning

In high schools, tests are really a testing instrument, by whichteachers assess learners’ studying results in mark

The purpose of this thesis is to introduce the idea of testing as ateaching act, which is divided by the teacher himself for a very specificpurpose

1.1.2.2: What is a Good Test?

When writing a good test, there are three factors to consider:

 Validity

 Reliability

 Practicality Here we are dealing with what these are and how they must beconsidered

1.1.2.2.1Validity

1.1.2.2.2.2Definition of Validity :

According to Carroll and Hall (1985): Validity is that we areactually measuring what we are intending to measure A listening testwith multiple choice options may lack validity if the printed choice is sodifficult to read that the exam actually measuring readingcomprehension as much as it does listening comprehension Validitymay be, in other way, defined as the degree to which a test actuallytests what it is intended to test

This definition has two very important aspects The first is thatvalidity is a matter of degree Tests are not either valid or invalid Thereare degrees of validity, and some tests are more valid than other Asecond important aspect of this definition is that tests are only valid orinvalid in term of their intended use If a test is intended to test reading,but it also tests writing, then it may not be valid for testing reading

1.1 2.1.2: Types of Validity:

Validity is divided into different types It is broadly discussedunder the headings: Face; content; response; concurrent; and predictive.

1.1.2.2.1.2.1 Face Validity:

A test is accepted to have face validity if it looks like it will testwhat it is intended to test It is the opinion of non-experts about what atest is really getting at While their opinion is not expert, it can beimportant, because it is the kind of response that you can get from thepeople who are taking the test If a test does not appear to be valid to thetestees, they may not do their best, so the perceptions of non-experts areuseful

1.1.2.2.1.2.4 Concurrent Validity

It is the degree to which a test correlates with other tests testing

at the same time In other words, if a test is valid it should give thesimilar results to other measures that are valid for the same purpose

There are several matters we should consider when judgingconcurrent validity

First, the measures that are being used for comparison of the test

in questions must be valid If the measure is not valid, there is no point

in testing another test' validity against it

For example, teacher's ranking might be used to test validity butthe teacher' ranking may be affected by a number of factors that are notrelated to the students 'actual proficiency So the possible solution forthis problem is to average the rankings of several teachers tocompensate for this

Second, the measure must be valid for the same purpose as thetest, whose validity is being considered A reading test can not be used

to test the concurrent validity of a grammar test Moreover, if the testbeing considered is a grammar test, then the teacher should be asked torank the students according to their grammar proficiency, not theiroverall English language ability

1.1.2.2.1.2.5 Predictive Validity

A test considered to have predictive validity if it can makepredictions about future performance For example, a test of Englishability has predictive validity if it accurately predicts how well studentswill get along in a university in an English -speaking country

1.1.2.2 3 Reliability

1.1.2.2.3.1 Definition of Reliability

Every test should be reliable In other words, a test shouldmeasure precisely whatever it is supposed to measure If a group ofstudents were to take the same test on two occasions their results should

be roughly the same, provided that nothing has happened in the interval.Thus, if the students' results are very different e.g the top studentscoring low marks the second time, the test can not be described asreliable

1.1.2.2 4 Practicality

The main question of practicality is administrative A test must

be carefully organized well in advance These following questions arepractical since they help ensure the success of test and testing:

-How long will the test take?

-What special arrangements have to be made? (For example: what happened to the rest of the class while individual speaking test take place?)

-Is any equipment needed?

-How is marking the works handled?

- How are test stored between sittings of tests?

1.1.2.3: What is a speaking test?

Most of us would agree that the oral skills are the most importantone However, they are most difficult to test As teachers of English wehave all come across fluent speakers who communicate effectivelywithout ever shifting tense, or worrying about agreement We alsocome across students who pay so much intention on accuracy that theirspeech is full of hesitation, false start and self correction, how do wedeal with them? The problem simply does not arise in their writtenwork as we place the student's compositions side by side on the desk.That’s why, besides using written work to test students' ability, teachersneed another instrument - oral tests

Oral test is viewed as an activity whose main purpose is to convey howwell the testees speak

Almost without exception, these tests will involve the teacher in aone -to-one situation with the student taking the test

1.1.2.4: Why do We Test Speaking Skills?

Testing has long been part and parcel of teaching all along andwill doubtless to be so However there have been controversial ideasboth for and against test The first tendency considers tests as notnecessary for it takes so much time and effort of both teachers andstudents Moreover tests also cause tense and can not reflect as exactly

as expected what learners have achieved The second and also the morepopular tendency agree to view test as an important instrument oflanguage teaching for these following reasons:

a To give teacher information about what level of knowledge thestudents are at, from that they know what to teach next

b To check and measure how progressive the students are, andhow successful the teaching has been

c To find out where students need to follows to improve theirspeaking

d To decide what area teacher should teach in the future anddefine goals for students to aim towards in learning speaking They willregard speaking as less " serious" than the other skills.

1.1.2.5: Characteristics of a Good Speaking Test.

To achieve a good speaking test, beside general elements that areneeded for ensuring for a good test, a speaking test should have specificcharacteristics as the followings:

1.1.1.5.1: Length of the Oral Test.

A good oral test should be as long as feasible Much reliableinformation can not be included in less than about 15 minutes.However, 30 minutes probably provides all information necessary formost purposes As part of a placement test, a five or ten minute- -interview should be sufficient to prevent gross errors in assigningstudents to classes

1.1.2.5.2: Wide Range Sample of Specified Content.

A wide-range sample of specified content in the time available isanother characteristic of a good oral test Teachers select what regarded

as a representative sample of the specified content and then plan how toelicit the necessary behaviors

1.1.2.5.3: Careful Planning.

Careful planning is an important characteristic of a test in generaland especially an oral test, while one of the advantages of individualoral testing is the way in which procedures can be adapted in response

to a candidate's performance For example, it is a mistake to begin aninterview with no more than a general idea of the course that it mighttake

1.1.2.5.4: Fresh Starts.

During a good oral test, testers should give candidate as many

"fresh starts" as possible It should consist of three criteria Firstly morethan one format should be used if possible and appropriate Secondly,candidate is desirably to interact with more than one tester Thirdly,within a format, there should be as many separate "items" as possible.Moreover, if a candidate gets into difficulty, not too much time should

be spent on one particular function or topic In addition, tester shouldcreate chance for candidate to attempt to express what they want to say

1.1.2.5.5: Sympathetic and Flexible Interviewers.

A good oral test needs having a sympathetic and flexibleinterviewer A successful interviewing is by no means easy.Interviewers need to be sympathetic and flexible So they can lessen thetestees’ nervousness and create chances for them to show their abilities.

1.1.2.5.6: Quiet Room With Good Acoustics:

A good speaking test can not be carried out in a chaotic room, itwill distract testees' attention and flow of thinking It will also maketester difficult to listen to their voice

1.1.2.5.7: Relevant Information.

A good oral test needs to have relevant Information If the testintents to determine whether a candidate can perform at a certainpredetermined level, then, after an initial easy introduction, the testshould be carried out at that level If a candidate is obviously very weakand can't reach the criterionlevel, interviewers should bring gently theinterview to a close, because it is no use to subject them to a longerordeal On the other hand, the test intents to see what level thecandidate is at

1.1.2.5.8: Fair Talking.

During a good interview, the interviewers don't talk too much butgive enough talking time to candidates They should avoid thetemptation to make lengthy or repeated explanations of something thatthe candidate has misunderstood

1.1.2.5.9: Unstressful Candidate

The candidate is put at their ease Individual oral tests willalways be particularly stressful for candidates For this reason, duringthe interview, it is important to make them feel easy by being pleasantand reassuring throughout So at the initial stages, it is speciallyimportant to make the initial stages well within the capacities of allreasonable candidates For example, interviews can begin withstraightforward requests for personal (but not too personal) details,remarks about the weather, and so on Moreover, testers should notconstantly remind candidates that they are being tested In addition, theinterview should be ended at a level at which the candidate clearly feelscomfortable so that, leaving him or her with a sense of accomplishment

Beside these above criteria, a good oral test also needs to haveother conditions For example, it should have a second tester forinterviews As the fact that, it is difficult to conduct an interview andkeep track of the candidate's performance so that, it needs a secondtester, who can not only give more attention to how candidate isperforming but also elicit necessary performance for a reliablejudgment.

Last but not least, in a good oral testing, testers set only tasks andtopics that would be expected to cause no difficulty for candidates intheir own language

Besides these above characteristics, depending on the capacities

of both testers & testees, a successful test may need a tactful andflexible adjustment to accomplish an oral test successfully

Chapter 2: Techniques for speaking tests.

2.I What do we test?

2.1.1: testing low level students.

The lower level is characterized by the ability to talk simplyabout self and family members They can only ask, answer questionsand participate in simple conversations on topics familiar with theirlevel These topics do not need to use complex language knowledge andskills e.g leisure time activities, the daily routine, family-memberdescription, e.t.c

For the above characteristics, to test speaking ability, we focus onfive aspects: pronunciation; grammar; vocabulary; awareness oflanguage function and ability to participate in conversation.

2.1.2:Testing higher level students.

Higher level is characterized by the speaker's ability to handlesuccessfully a variety of rather complicated communicative task andtalk about complicated topics such as environment, education, andsocial evil Moreover, they can initiate, sustain and close in acomplicated way a long and difficult task

For these characteristics, to test their ability of speaking we focus

on three main aspects Firstly, we test their ability of communicating,how well they can communicate in different situations Secondly, wehave to consider their ability to negotiate We should measure how wellthey negotiate to gain their purposes in the communication The lastpoint testers should focus on is the students’ ability to collaborate Inother words, the teachers have to test the students’ ability ofcollaborating with their partners to control the conversation

2.2 Rating scales of oral testing criteria

One of the great difficulties in testing speaking is grading It isnecessary but also difficult to set up a system of assessment that can beapplied as objectively as possible, though we can't avoid somesubjective assessment

The scale set for assessment can be the general one for overallspeaking ability, or it can be set up for separate aspects of the skill ofspeaking, such as: pronunciation; grammar, organization

According to TPSL-L Electronic discussion Forum, thedescriptions for a general scale might be as follows (the numberindicates the level, and it is followed by a description of thecharacteristics of speaker at that level)

There are seven levels, which corresponds to seven numbers Thehighest number indicates the highest level, and the followings in turnindicate the rest levels, which are ranked from the highest to the lowestlevel

(7) Spoken communication is fluent, appropriate; andgrammatically correct, with few errors

(6) Communication is generally fluent and grammatically correctwith only occasional errors in grammar or pronunciation

(5) Student produces numerous grammatical errors andhesitations, and these do not interfere with communication Utterancesare long and connected.

(4) Students produces numerous grammatical errors andhesitations, and these occasionally interfere with communication.Utterances are short and connected

(3) Student's communication is limited to short utterance anddepends in part on previously memorized conversational elements.They have difficulty in dealing with unpredictable elements There aremany hesitations and grammatical errors Communication is onlypossible with sympathetic interlocutor

(2) Communication is limited to short utterances, almost entirelymemorized conversational elements Students are unable to deal withunpredictable elements

(1) No communication possible

Another way of grading oral skill based on the scale, which isdivided between several aspects of speaking skill such as vocabulary,grammar, pronunciation etc These criteria are ranked from the lowest

to the highest ones They are divided in to 2 types: accuracy andfluency

Besides the above form of rating scale, there are three other types, which are considered as the most common They are:

- The numerical scale

- Graphic scale

- Descriptive scale

In numerical rating scales, a number stands for a point on the rating scale Thus, in the example below, "1" corresponds to the pupils

"always "performing the behavior;"2" to the pupils "usually

"performing the behavior, "3" to the pupils "seldom" performing the behavior and "4"if the pupils "never" perform the behavior

Example of numerical rating scale

1 2 3 4

2.2.2: Graphic rating scale

Graphic rating scale requires the rater to mark a point on a line divided into sections based upon a scale The rater marks an "X" at the

point that best describes the pupils' performance

Example of graphic rating scale:

A: Student stands straight and faces audiences,

B: Student changes facial expression with changes in the tone of the presentation

Always Usually Seldom Never

2.2.3: descriptive rating scales

The descriptive rating scales require the rater o choose among different descriptions of actual performance In descriptive rating scalesdifferent levels of the performance make up the descriptions of pupils'

performance from which the teacher chooses

Example of descriptive rating scale:

A: Student stand straight and faces audiencestands straight , always

looks at audience

weaves; fidgets ;eyes roam from audience to ceiling

constant ; distracting movements; no eye contact with audience

B: Student changes facial expression with changes in the tone of the presentation

2.2.4: General rules of using rating scales

Regardless of the type of rating scales one chooses, a couple ofgeneral rules will improve their use.

The first rule is to limit the number of rating categories There is

a tendency to think that the greater the number of rating categories to

choose from, the better the rating scale In practice, this is not the case

Few observers can make reliable discriminations in pupils' performance

across more than five rating categories and using more than five is

likely to make the rating unreliable

Second, teachers should strive to use the same rating scale foreach performance criterion This is not usually possible in descriptive

rating scales where the descriptions vary with performance criteria

However, for numerical and graphic scales, it is best to select a single

rating scale and to use it for all performance criteria Using many

different scales requires the observer to change focus frequently,

distracting attention from the performance and decreasing rating

accuracy

2.2.5: Standard rating scales

Of all three types of rating scales, numerical rating scale is themost straightforward and commonly-used approach to summarizing

performance on rating scales It assigns a point value to each category

in the scales and sums the points across the performance criteria one

assign In this type of rating sale,4 points to a rating of "always”3 points

to "usually”,2 points to "seldom”, and, none points to "never"

Thus, high scores indicate good performance There are 3 mainaspects to consider when rating oral presentation They are physical

expression, vocal expression and verbal expression

C: Student maintains eye contact with the audience

C: Student speaks loudly enough to be heard by the audience

D Student paces word in an even flow

4 3 2 1 E: Student enunciates each word.

E: Student summarizes main points at conclusion

2.3.Procedures of building a speaking test.

2.3.1 some suggestions for building a speaking test.

2 3.1.1: The Oral Test Should Be As Long As Feasible.

It is unlikely that much reliable information can be included in ashort time but longer time for example 30 minutes can probably provideall the information necessary for most purposes

2 3.1.2: There Should Be a Carefully-Planned Test.

The individual oral testing has an advantage that procedures can

be adapted in response to a candidate's performance

However, the tests nevertheless have some patterns to follow

2.3.1.3 Tasks and Topics Should Be Suitable.

These tasks and topics would be expected to cause testees no

difficulty in their own language Moreover, these tasks and topics

should be suitable with students’ age; living condition; and learningconditions

2.3.1.4: The Goal of the Test Should Be Kept In Mind

The goal of the test is what you want to measure It is necessarythat you only measure what you want is to measure and nothing else.For example, if you are trying to test communicative English you onlymeasure their communicative ability by combining testing only fluencyand accuracy

2.3.1.5 Teachers Should Consider How Much Weight Each Item, Each Section is Given.

It's obvious to say that you should give the greatest weight to themost important thing you are testing

2.3.2: Stages of test construction

The general procedure for test construction usually consists of 2main stages: stage the problem and provide solution to the problem

2.3.2.1: Statement of the Problem.

It is obvious that the essential first step in testing to make oneselfperfectly clear about what it is one wants to know and for what purpose.The following questions have to be answered.

- What kind of test is it to be? achievement; proficiencydiagnostic or placement?

- What is its precise purpose?

- What abilities are to be tested?

- How detailed must the results be?

- How important is backward?

- What constraints are set by unavailability of expertise,facilities, time (for construction; administration and scoring)?

2.3.2.2 Providing a Solution to the Problem

2.3.2.2.1 Writing Specifications for the Test.

The first form that the solution takes is a set of specifications forthe test This will include information on: content, format and timing,criteria level of performance and scoring procedure

2.3.2.2.1.1: Content

The way in which content is described will vary with its nature.The content of a grammar test may simply list all the relevantstructures The content of a test of a language skill, on the other hand,may be specified along number of dimensions The important thing isthat content should be as fully specified as possible

2.3.2.2.1.1.1: Operations:

The tasks that candidate has to be able to carry out

2.3.2.2.1.1.2: Types of Test

For a speaking test these might include:

Role-plays; oral presentation

Topics are selected according to suitability for the candidate and

the type of test

2.3.2.2.1.2: Format and Timing.

This should specify test structure (including time allocated tocomponents) and item types with examples it should state what