Test purpose

According to the IELTS website (2013), there are actually two IELTS tests; an academic version and a general training version. The speaking test is the same for both versions.

The academic version is for students who want to enrol in universities or higher education institutions, or for professional such as doctors that want to work in English speaking counties. Different scores are required for different academic courses or professional placements (Cambridge English, 2013).

The general training version is used to assess people intending to do non-academic training, or for immigration to English speaking countries (Cambridge English, 2013). Different scores are required depending the reason for immigration.

IELTS is recognised by most academic institutions in Australia, New Zealand, United Kingdom, Canada, Ireland and South Africa, as well as about 3,000 in the United States. An acceptable score is also required for immigration to Australia, New Zealand and Canada. (Cambridge English, 2013).

Test description

The IELTS Speaking Test consists of three parts (excluding the introduction and conclusion), which is mostly a conversation between the interviewer and the candidate (Brown, 2003). The second part is not conversational, and the entire test lasts between 11-14 minutes (IELTS, 2013)

The first part is an introduction and interview, lasting 4-5 minutes. The interviewer and candidate make introductions, and confirm the candidate’s identity. The interviewer asks the candidate general questions about his/her personal life, including home, work, studies, family and interests. (Cambridge English, 2013)

The second part is an individual long turn (at talking), lasting 3-4 minutes. The candidate is given a topic card, which contains a topic and points the candidate can cover in a talk on that topic. The candidate is given one minute to prepare their talk, during which time they may make written notes to assist in the talk. The candidate then talks for 1-2 minutes on that topic. After the talk, the interviewer asks the candidate 1 or 2 questions about the topic, to test understanding (Cambridge English, 2013).

The third part is a two way discussion, lasting 4-5 minutes. This is generally considered the most difficult part. The interviewer asks more questions about the topic from part two. The candidate must answer each question, discussing abstract issues and ideas (Cambridge English, 2013).

During the test, many speaking skills are assessed. These include communicating ideas and facts about ordinary issues, experiences and situations. This is done by the interviewer judging the candidate’s ability use suitable language, organise ideas logically and express substantiated opinions. Analysis, discussion and speculation about the issues are also assessed (IELTS, 2013).

The candidate’s speaking performance is assessed by the interviewer, also known as the examiner. Scores are marked in whole and half bands, based on the rubrics in the Speaking Test Band Descriptors (Appendix 3) (Cambridge English, 2013). The four criteria assessed, as described by Cambridge English (2013), are:

Fluency and Coherence – the ability to talk using normal levels of continuity, rate and effort and to link language and ideas together. Fluency is assessed by speech rate and speech continuity. Coherence is assessed by logical sequencing of sentences, marking stages in a discussion clearly, narration/argument, and use of cohesive devices such as connectors, conjunctions and pronouns within and between sentences.

Lexical Resource – the range of vocabulary that is used, and the precision of meanings and attitudes expressed. It is assessed by the variety of words used, whether adequate and appropriate words are used and the ability get round a vocabulary gap by using other words (with or without noticeable hesitation).

Grammatical Range and Accuracy – the range, accuracy and appropriateness of the candidate’s grammatical resource. It is assessed by the length and complexity of spoken sentences, appropriate use of subordinate clauses, and range of sentence structures. The ability to move features around to enhance meaning is also considered. The key indicators are the number of grammatical errors and the communicative effect of error.

Pronunciation – the ability to produce understandable speech to during the interview. Consideration is given to the amount of strain caused to the listener, the amount of the speech which is not intelligible and any L1 influence.

Implementation and use of the test

According to the British Council (2011), the IELTS test is the most popular of the important language tests in he world, with 1.5 million tests having been completed by early 2011. The three largest users of the test are China, Australia and India, with North America, Philippines and Hong Kong the three fastest growing users by country.

In 2011, there were 800 IELTS testing locations, in 130 countries (British Council, 2011). More than 6000 institutions around the world use the IELTS test. Over 6000 institutions around the world use the IELTS test as a guideline or requisite, including universities, immigrations authorities, professional bodies and employers (British Council, 2011).

The number one reason internationally for taking the IELTS test is for entry to an educational institution (British Council, 2011). The next most used reason being for immigration.

In Australia, various academic and training courses require different level of English proficiency (QUT, 2012). The academic IELTS test is used for entry to tertiary institutions such as universities, and the general training IELTS test is used for access to vocational training such as TAFE colleges and Registered Training Organisations (RTOs):

Academic study requiring high level language skills, such as mathematics, IT and telecommunications requires a minimum IELTS sore of 7.0-7.5
Academic study requiring lower level language skills, such as medicine, law journalism or linguistics, requires a minimum IELTS sore of 6.5-7.0
Training course requiring reasonably high level language skills, such as engineering, applied sciences or air traffic control, requires a minimum IELTS sore of 6.0-6.5
Training course requiring moderate level language skills, such as catering or fire services, requires a minimum IELTS sore of 5.5-6.0

For immigration purpose in Australia, different jobs have different IELTS requirements, roughly similar to academic entry requirements (Australian Immigration, 2013). The general training IELTS test is used for Australian immigration purposes.

Other countries that use IELTS for immigration include United Kingdom, Canada and New Zealand (British Council, 2011). Amongst international organisations who also use the IELTS test as a measure of ability are the International monetary Fund (IMF) and the Commission on Graduates of Foreign Nursing Schools and National Council of State Boards of Nursing (NCSBN) in the USA (British Council, 2011).

Strengths and weaknesses of the IELTS Speaking Test

The strengths and weaknesses of the IELTS Speaking Test can best be demonstrated by examining individually the six qualities of tests, which together are complementary in defining the test’s usefulness. The usefulness of the test should be considered by its overall value rather than the value of each of the six qualities individually (Bachman & Palmer, 1996). However, for the purpose of this study, each quality will be examined separately:

Reliability

Reliability is consistency of measurement (Bachman & Palmer, 1996). Test scores need to be relatively consistent, so the speaking ability of one student can be compared to the speaking ability of another.

At first glance, it would appear that the rubrics used for the IELTS Speaking Test would ensure consistency across al candidates as the same criteria are applied for all candidates (IELTS, 2013). However, as is stated in Zahedi & Shamsaee (2012), difficulties arise when the interviewers apply subjective ratings to the candidates.

Additionally, different interviewers can use different sorts of questioning, style and approach (Brown, 2003). The different styles, or even different genders, can affect candidate’s responses. Additionally, the different personalities of the interviewers can mean they have different perceptions of the candidates’ speaking abilities.

Chalhoub-Deville & Turner (2000) discuss whether tests are taken under optimal conditions. Exactly what optimal conditions are is unclear, except that ideal conditions can never be achieved, and rarely even approached.

Chalhoub-Deville & Turner (2000) look explicitly at reliability in IELTS speaking tests. They state that any variable that affcts test scores, other than the language itself, is considered measurement error. This in turn affects the reliability. The possibility of this occurring at different tests is almost inescapable.

However, reliability is seen as being positive in consistency of assessment according to the rating scale, if open ended items are used in the interviews by all interviewers (Chalhoub-Deville & Turner, 2000).

These different factors have different weightings, but overall, the method of production and assessment appears to make the IELTS speaking test reliable.

Construct Validity

Construct validity refers to how meaningful and appropriate the interpretations of the test score are (Bachman & Palmer, 1996). Here must be evidence that the test score only reflects the language ability being tested (in this case speaking) and nothing else. It also must test each of the four criteria in terms of the band descriptors (Appendix 3)

Zahedi & Shamsaee (2012) state that the construct validity of the IELTS speaking test is almost universally trusted. The relationships between the candidates’ speaking and actual scores show little variation.

In their study, Zahedi & Shamsaee (2012) compared the results of the candidates’ speaking tests with their results from the other three tests, as well as with their overall result. The correlation coefficient in all cases showed a strong relationship, which in turn shows that the speaking test is meaningful and appropriate.

Chalhoub-Deville & Turner (2000) state that the validation process of IELTS, including the speaking test, is an ongoing process, rather than a one-time action. IELTS developers have shown they are committed to creating test practices centred on research findings.

However, Chalhoub-Deville & Turner (2000) do express some concerns that IELTS was originally developed for use in the United Kingdom and Australia. With IELTS now being more widely used in USA and Canada, there may be a need to examine the suitability of the test to the language differences in North America.

Authenticity

Authenticity refers to the degree to which score tests relate to actual use in the target language domain (Bachman & Palmer, 1996). This in turn relates to the candidate’s opinions of the test, and ultimately will affect their performance in the test.

A first step in deciding on authenticity is to identify important features that associate the task with the target language use (Bachman & Palmer, 1996). In the case of the IELTS speaking test, the use of spoken language in the test closely resemble conversation in a real situation.

The descriptions of the books for IELTS test preparation cited in Terry (2003) favour those that prepare for conversations in real life conversations, similar to what the IELTS test does. Ironically, Terry (2003) criticises the Cambridge Practice Tests books, written by the tests designers, as not being the best for preparation. In that respect, more importance is put on real life situations than “studying for the test.”

Brown (2003) also speaks of engaging the candidate in conversation, rather than testing the candidate. Again, this shows the emphasis on target language use (TLU).

A criticism of the IELTS Speaking Test is that it is “high stakes,” and that this cause anxiety amongst the candidates that would not be experienced in a real life situation (Issitt, 2008). Getting he students into the right frame of mind is easier in IELTS preparation classrooms, but not so easy in the test itself.

Interactiveness

Interactiveness is the degree of participation the candidate shows of distinct features in completing the test task (Bachman & Palmer, 1996). The key characteristics are the candidate’s language ability, including strategic competence and language knowledge or metacognitive strategies, as well topical knowledge and affective schemata.

The manner of the test itself, as an interview between the tester and the candidate, is very interactive, as both sides rely on language skills to achieve the best results. In particular, the candidate must satisfy rubrics that were known before the test, being the band descriptors (Appendix 3) (Cambridge English, 2013).

Fulcher (2003) states that the primary concern in testing speaking is to ensure that enough speech is produced to allow the interviewer to make an accurate rating. The IELTS speaking test appears to satisfy that requirement, wit a variety of speaking tasks in real life situations.

It appears that the interactiveness quality of the speaking test can be improved by IELTS preparation courses (Gan, 2009). Preparation for the test through English language university studies played an important part of candidate’s improvement in language ability, and their ability to interact with the interviewer in Gan’s (2009) study.

Stones (2013) used transcription to analyse the interactiveness of the speaking test specifically. The students would transcribe and analyse the practice tests they did in IELTS preparation courses. They would then identify grammatical features, syntax and lexical items they had used, and hoe these could be improved. This strengthened their ability to interact with the interviewer, and increased confidence.

A criticism of interactiveness is that the test is video recorded, and a second rater assesses the candidate without any personal interaction (Macqueen & Harding, 2009). However, it is also possible that the second rater will be more objective about the assessment than the rater conducting the interview.

Impact

Impact refers to the effect if the IELTS Speaking Test on society and educational systems, and on the people contained in those systems (Bachman & Palmer, 1996). At a micro level, impact is concerned with the candidate taking the test, and at a macro level, the impact on society and various educational institutions.

The impact of the IELTS Speaking Test on society is undeniable. It is an institution within academic and immigration circles in the United Kingdom, Australia, New Zealand and Canada, and it is becoming increasingly important in the United States (IELTS, 2013)

For the individual, it can change lives. A good IELTS score can mean acceptance to international universities, and the opportunity for an education at an English language institution. It can also mean the possibility of immigration to an English speaking country (IELTS, 2013).

For society, it can help screen applicants for immigration based on understanding of English language, and, by extension, western culture. This can be for skill based migration, or for skills training in Australia (Australian Immigration, 2013).

A criticism of IELTS is that it is designed for the United Kingdom and Australia, but is now so international that it fails to be country-specific in many countries where it is gaining a foothold (Chalhoub-Deville & Turner, 2000). More research and development is needed to adapt it to the internationalisation of English and the various World Englishes.

Practicality

Practicality is different to the other five qualities of test usefulness, in that it is concerned mainly with the way the test will be implemented (Bachman & Palmer, 1996). The other five qualities are concerned with the uses of the test. Practicality also dictates whether the test will be developed at all.

Practicality is the relationship between the resources required to design, develop and implement the test, and the actual resources available. The main consideration when deciding whether it is practical to continue development of the IELTS speaking test is the impact the test has on all stakeholders (Bachman & Palmer, 1996).

It seems apparent that the IELTS Speaking test has enormous impact, and that it will continue to expand its influence globally (Chalhoub-Deville & Turner, 2000). In light of the fact that more than 1.5 million people have already used the IELTS test, it seems that allocation of vast resources is practical for further development.

Conclusion

Taking into account all the qualities of test usefulness, particularly he impact on the candidates and the global society, it would appear that the IELTS Speaking Test is, indeed, a useful test. It should also be remembered, when making this judgement, that each of the qualities does not stand alone, but is considered relative to all the other qualities. An optimal balance between all the qualities is also essential when deciding usefulness.

It is also practical, as well as desirable, to continue development of the IELTS test, particularly the speaking test, because of he considerable impact it will have in the future on many people from all countries, and using all the world’s Englishes.

One of the most important considerations is that the test qualities must be considered from the beginning of the test planning and through to development and practical use. It is not enough to just use the test qualities to analyse the test after it has been implemented (Bachman & Palmer, 1996). Given that the IELTS Speaking Test is constantly under review, it would appear that the test qualities, as interpreted in this analysis, are suitable for the evaluation of this test.

Example of an IELTS Speaking Test

Part 1 – Introduction and interview

[This part of the test begins with the examiner introducing himself or herself and checking the candidate’s identification. It then continues as an interview.]

Let’s talk about your home town or village.

What kind of place is it?
What’s the most interesting part of your town/village?
What kind of jobs do the people in your town/village do?
Would you say it’s a good place to live? (Why?)

Let’s move on to talk about accommodation.

Tell me about the kind of accommodation you live in?
How long have you lived there?
What do you like about living there?
What sort of accommodation would you most like to live in?

Part 2 – Individual long turn

Candidate Task Card

Describe something you own which is very important to you.

You should say:
where you got it from
how long you have had it
what you use it for

and explain why it is important to you.

You will have to talk about the topic for 1 to 2 minutes.
You have one minute to think about what you’re going to say.
You can make some notes to help you if you wish.

Rounding off questions

Is it valuable in terms of money?
Would it be easy to replace?

Part 3 – Two-way discussion

Let’s consider first of all how people’s values have changed.

What kind of things give status to people in your country?
Have things changed since your parents’ time?

Finally, let’s talk about the role of advertising.

Do you think advertising influences what people buy?

Sample IELTS Speaking Test Evaluation

Examiner’s Commentary

The student interviewed was Sandra, a Colombian female. The Speaking Test is in three sections. First of all let’s look at each of these sections in turn to identify the strong and weak parts of the interview.

Section 1

Sandra was a little bit nervous and not very confident in her speech. She answered all the questions but some of the questions, particularly in Topic 2, could have been longer and better developed. Most people find that Section 1 of the IELTS Speaking Test is the easiest section and therefore students should take the opportunity to talk as long and as fully as they can while the questions are at their simplest. For example, the examiner asks Sandra about the advantages and disadvantages of living in her area. She talks about the climate as an advantage, which is good, and then mentions that her area is near other good places but she does not talk more about this. She could talk about other good things regarding her area and she did not even mention one disadvantage. This was wasting an opportunity. On the other hand she answered most of the questions quite well and this was certainly no disaster.

Section 2

In Section 2 Sandra managed to get over the one minute mark without any problems but it appeared as though it would have been hard for her to go any longer than she did. Like Ilaria in the recording for the Academic Speaking Test 1, Sandra did not make use of the one minute preparation time. It sounds as though the examiner got her started early but it was Sandra who indicated that she was ready to begin. By not taking the time to make notes on each part of the question, Sandra’s talk was a bit disjointed and lost coherence at times. Here is an example where using the one minute preparation time would have had great benefit. On the other hand Sandra did manage to answer most parts of the question. Looking at her actual speaking, Sandra often had long hesitations as she searched for things to say and to connect her ideas. The connections were not always clear and she shifted subject a couple of times in one sentence. If she could have finished each of the sentences by developing the subjects of each sentence and then change the subject coherently, it would have been a much better talk. Again, preparation would have benefited her. She did not lose marks for not preparing but the lack of coherence in her talk that the lack of preparation caused affected her score.

Section 3

Section 3 started very well with Sandra giving long and full answers to the questions in Topic 1. In Topic 2 however the answers were not as long and could have been better developed, speaking about the subjects in more depth. For example the last question was not really answered properly at all with Sandra only saying that holidays would change by having more amenities. If Topic 2 could have been answered as fully as Topic 1 then she would have made a much better impression.

Marking

The Marking of the IELTS Speaking Test is done in 4 parts. Below is how the examiner evaluates this student.

Pronunciation

Sandra’s pronunciation was not that clear. There were areas where there were no problems but there were also times when her strong accent, intonation and stress problems caused difficulties for the listener.

Grammatical Range and Accuracy

Sandra’s grammar was often faulty. Sandra’s basic sentences were controlled fairly well but more complex grammar often broke down. There were lots of mistakes scattered throughout her speech as well, some of them quite basic.

Lexical Resource

Sandra’s vocabulary was good enough to discuss familiar and unfamiliar topics but she had limited flexibility. There were quite a few pauses where she had to stop to search for the right word and not always with success.

Fluency and Coherence

Sandra’s fluency and coherence was sometimes quite good and she showed the ability to talk independently and at some length to questions. On areas that she was not so sure about though her fluency and coherence sometimes broke down and this led to quite a lot of hesitation and a bit of repetition.

Estimated IELTS Speaking Band: 5

Reference List

Australian Immigration website (2013) www.immi.gov.au

Bachman, Lyle F. and Palmer, Adrian S. (1996). Chapter 2 : Test Usefulness : Qualities of Language Tests in Bachman, Lyle F. and Palmer, Adrian S, Language testing in practice : designing and developing useful language tests, Oxford: OUP, pp.17-42.

British Council (2011). IELTS breaks the 1.5 million tests mark. PR Newswire, (2011, Mar 13)Retrieved from http://search.proquest.com/docview/856700817?accountid=13380

Brown, A. (2003). Interviewer variation and the co-construction of speaking proficiency Language Testing, 20 (1), 1-25.

Cambridge English website (2013) www.cambridgeenglish.org

Chalhoub-Deville, M., & Turner, C. E. (2000). What to look for in ESL admission tests: Cambridge certificate exams, IELTS, and TOEFL. System, 28(4), 523-39.

Fulcher, Glenn. (2003). Chapter 3 : Tasks for Second Language Speaking Tests in Fulcher, Glenn, Testing second language speaking, London: Pearson Longman, pp.50-87.

Gan Zhengdong. (2009). “IELTS Preparation Course and Student IELTS Performance: A Case Study in Hong Kong”. RELC journal , 40 (1), p. 23.

IELTS website (2013) www.ielts.org

Issitt, S. (2008). “Improving scores on the IELTS speaking test”. ELT journal , 62 (2), p. 131.

Terry, M. (2003). “IELTS Preparation Materials”. ELT journal , 57 (1), p. 66.

Luoma, S. 2004, Assessing speaking, Cambridge University Press, Cambridge.

Macqueen, S., & Harding, L. (2009). Test review: Review of the certificate of proficiency in english (CPE) speaking test. Language Testing, 26(3), 467-475

QUT (2012). Class handout for CLN616 Using IELTS Test Scores

Stones, Thomas P (01/01/2013). “Transcription and the IELTS Speaking Test: Facilitating Development”. ELT journal , 67 (1), p. 20.

Zahedi, K., & Shamsaee, S. (2012). Viability of construct validity of the speaking modules of international language examinations (IELTS vs. TOEFL iBT): Evidence from iranian test-takers. Educational Assessment, Evaluation and Accountability, 24(3), 263-277.