ielts-yasi.englishlab.net  

Updated May 7, 2011

    Notes for the Detailed Band Score Descriptors

• I have taken the public version of the Band Descriptors, as well as the version that examiners are given, and added my personal interpretations of what these descriptors mean. (The public version is about 95% the same as the examiners' version.) I have also used the knowledge that I gained from the examiner training, which mostly consisted of studying representative examples of different Band levels in tape recordings and/or videos.

• Most current or former examiners would agree with most of what I have written for this detailed description of the band descriptors but there might be some specific points where we disagree slightly. For example, there might be some disagreement on the specific grammar points that I have listed for Band 5 and Band 6, which are mostly my personal interpretation of what these two Bands represent. As well as that, there might be slight disagreement about the numbers I use for the number of grammar errors. However, when the different points or items are averaged out, most examiners would agree with my descriptions here.

       Notes for Coherence

• "Coherence" refers to the degree of understandability of several ideas that are presented together; "cohesion" refers to the linkage of the ideas. For long or detailed utterances, cohesion is one of the most important ways to achieve good coherence.

• For Coherence, the examiner is not allowed to give a sub-score of 6 or more if the candidate does not show knowledge of a range of connective words and phrases (to show connection between sentences or ideas, and usually placed at the beginning of sentences).

• For the "Fluency and Coherence" sub-score, the examiner needs to assess a sub-score for each of Fluency and Coherence. If the two sub-scores are different, the lower of the two scores is the final sub-score. For example, sometimes a candidate gets 5 for the Fluency and Coherence sub-score even when he or she is as fluent as a Band 6 or even a Band 7 candidate. She usually fails to qualify for a 6 for this sub-score because she does not show a wide enough knowledge of connective phrases.

To repeat: If a candidate is a 7 for fluency but only a 5 for coherence (as shown by a limited usage of connectives), then that candidate gets a score of 5 for this sub-score, not the average of 7 and 5, which is 6. The sub-score here is the highest score that 'satisfies' both fluency and coherence – a score of 6 would satisfy the description of this candidate's fluency but would not satisfy a description of this candidate's coherence.

(Remember, “fluency” just refers to the flow of words from the mouth, i.e., the continuity and speed of this flow. But coherence is largely based on the choice of language spoken – mostly the words and phrases that are used to connect ideas so that ‘the ideas flow’.)

       Notes for Vocabulary

• "Idiomatic usage" refers mostly to phrasal verbs and other examples of idiomatic language, not proverbs or slang.
• "Style" refers to the formality of the language, usually the formality of the vocabulary.
• "Collocation" refers to commonly used word combinations. For example, in English we say, "heavy rain" but, "strong wind".

       Notes for Pronunciation

• "Accent" is the overall sound of a person's speech. It is the sum of the individual pronunciations of the vowels, consonants, diphthongs and other speech sounds of a person. When speaking about the accent of a person's English, we refer to it in terms of what English speaking country's accent it sounds like. There are two main accents in "Standard English" (i.e., two main varieties of pronunciation): the accent of English as spoken in southern England and the accent of English as spoken in most parts of the U.S.A. Both are equally acceptable in the IELTS test. But other national or regional English accents are also equally acceptable, for example: the Australian accent, the New Zealand accent, the Canadian accent, the South African accent, the accent from Northern England, the accent from Scotland, the accent from Ireland and the accent from the southern states of the U.S.A.

This creates a problem for the IELTS test. For example, if you study English in New Zealand for a year or two, you will start to speak English like a New Zealander. If you then come back to do the IELTS test in Beijing, it is possible that your examiner in Beijing is not used to the New Zealand accent and will have problems understanding you. But if you did the IELTS test in New Zealand, your New Zealand examiner would probably think you were speaking very naturally.

Although it is probably preferable to speak English using purely one accent, I think it is also acceptable if you speak English in a blend of two accents, as long as you sound like you have been mainly influenced by one accent and as long as this blend is a gradual blend, not a strongly contrasting mixture of different accents. For example, there are many native English speakers who speak mostly British English but who have been influenced by American English because they have lived in the U.S. for many years, and vice-versa. A mixture of accents like that is called, "international English".

• "Sentence stress" is the stress on one particular word (or more than one word) in a sentence. I have noticed that, for Chinese people, most errors in sentence stress are simply errors when speaking noun+ noun or adjective +noun combinations. One of the two words in these combinations is stressed more than the other. Probably this situation is the same for all learners of English, not just Chinese people. Another common sentence stress error among Chinese people is to stress the last word in a sentence when that word is a pronoun that is not stressed by a native English speaker. For example, "He gave it to me ", if spoken with the stress on the word "me" means that he gave it to me, not someone else. This is a situation of contrast. But if there is no contrast meant, the word "gave" would be the word that is stressed in that sentence.

• "Word stress" is the stress on one particular syllable in a multi-syllable word.

• "Intonation" is the rise or fall in the tone of an utterance. This is closely related to sentence stress but is not always the same. Different nationalities of English speakers tend to have a few differences in their use of intonation.

• At some time in the future I will include sound recordings on this website to illustrate examples of sentence stress and intonation. One important example of sentence stress is the stress on noun + noun or adjective + noun combinations. Information on this can be found here.

• The meaning of, "chunking" is included below. I am indebted to Matt Clark for the information shown below, which is about the changes to the Pronunciation sub-score that came into effect on August 1, 2008. (Although Matt Clark might not have intended his notes to be placed on the internet, the notes below were found at http://blog.sina.com.cn/s/blog_602badbd0100dwe1.html)

          -----------------------------------------------------

        From Matt Clark

With effect from August 1^st 2008, the marking scale for pronunciation will change in all IELTS speaking tests worldwide. Examiners will now be able to award band scores 1-9 instead of the older system of 2-4-6-8. 

Candidate’s pronunciation will be marked using the same criteria as in the previous system, but for the higher scores (7-8-9) there are one or two areas that have become more emphasized. 

The lower scores (2-4)

The lower scores will continue to focus on “persistent” or “fossilized” pronunciation problems. These include consistent mispronunciation of certain sounds, such as “r” “l” and “th”. If a candidate is continually mispronouncing the same sounds throughout the test the examiner will not be able to award a score of 6. 

Another key feature of the lower scores is a lack of overall intelligibility of “utterances” or sentence parts. If the examiner has difficulty in understanding the general message of a sentence or sentence-part within many answers then a lower score will be awarded. An effective way for candidates to “self-diagnose” their intelligibility is to record themselves delivering answers (not reading) and then to playback their answers and ask a friend or classmate if they can understand everything that was said.  

The mid-range scores (5-6)

The band scores 5 and 6 in the new system will be awarded when a candidate can successfully replicate the sounds of English on “word-level” and attempt to produce sentence stress and intonation. The good news for candidates is that the intonation and sentence stress do not to be correct in every “utterance”.

The key to scoring 6 in pronunciation is to speak English clearly enough to be understood throughout. There is no need to reproduce a native-speaker accent (British, American, and Australian) in this band score. Candidates with a Chinese accent can easily score 6 in this section. 

The higher scores (7+)

The introduction of the band score 7 for pronunciation is probably the most important development in recent years in the IELTS test. Previously many candidates could get close to an overall 7.0 in speaking but their pronunciation score (usually 6) would reduce the overall score to either 6.0 or 6.5. The pronunciation score of 7 has made it easier to achieve an overall7 in Speaking. 

What is needed for a pronunciation 7?

Candidates will be awarded a 7 if they fulfill the following criteria.

The examiner can fully understand candidate’s language. (even with occasional problems)

Examples of correct word /sentence stress to emphasize or focus meaning.

Use of intonation to emphasize important meaning.

Able to vary speed of delivery to affect meaning.

Able to divide utterances in “chunks” (see below) 

There may be some mispronounced words.

A first language accent may still be present. 

The features listed in the band descriptor for 7 are quite self-explanatory with the exception of  “chunking” therefore some more detail is provided below.

What is Chunking?

Chunking is actually quite a new “label” or “buzzword” in the EFL world and for many experienced EFL teachers and even for IELTS examiners it is something which is difficult to accurately define. 

When speech is produced by native speakers the words are not pronounced individually; groups of words (within an utterance) are pushed together into “multi-word chunks”. An ability to demonstrate this skill is required for a band 7. 

These “word groups” are not random; there are rules as to which words should be “chunked” together. Unfortunately, as with many features of English pronunciation, these rules are somewhat “fuzzy”. 

As a general guideline most “sentence fillers” or “connecting phrases” should be chunked into a “multi-word group”. Look at the following examples. 

“and all of those kinds of things”

“and stuff like that”

“I’m not really sure but..”

“a lot of people think that…”

“maybe it’s something to do with…”

“I guess that it could be because…” 

These phrases should be pronounced as one flowing sound with little or no audible gap between words. This skill becomes more obvious to the listener (or examiner) when the “word chunks” are surrounded by examples of speed or volume variation in important parts of the message. 

“Well I guess that in many ways, nowadays learning a language is quite important, and the main reason would probably be that it can help you get a job, it can make your more aware of other cultures and stuff like that”. 

The underlined phrases in the example above should be “chunked” together in a “word-group”. Also it becomes easier to see that the “message” parts of the example should contain variations in speed and volume to emphasize important meaning. 

A candidate who can demonstrate some ability to chunk will be awarded a 7. When the chunking becomes more natural and consistently correct an 8 may be awarded (if the other criteria are met).

          --------------------------------------