Stuck at 5.5 in IELTS Writing: The Four Things That Hold the Score There

A 5.5 in IELTS Writing is the score that stops an application. Many universities want 6.0 in each part, and many visa routes want 6.0 or 6.5, so half a band decides the outcome. Candidates who sit at 5.5 usually resit, practise more, and get 5.5 again.

That is not bad luck. A 5.5 is a stable score because it comes from habits that repeat in every answer you write. The way out is to find which habit sits on which criterion, then change that one thing. This post works backwards from the public band descriptors to the four habits, with the band 5 version and the band 6 version of the same sentence.

First, what 5.5 means

Writing is marked on four criteria, each given a whole band from 1 to 9:

• Task Response in Task 2, or Task Achievement in Task 1. Did you answer the question?
• Coherence and Cohesion. Can a reader follow it?
• Lexical Resource. Vocabulary.
• Grammatical Range and Accuracy. Grammar and punctuation.

The four are averaged, and the average is rounded to the nearest half band. So 5.5 is not a thing the marker writes down. It is what four whole numbers produce. The commonest route to it is three fives and a six, or two fives and two sixes.

That matters because it tells you where the work is. You do not need to improve everything by half a band. You need to move one or two criteria from 5 to 6, and you need to know which ones.

The four habits

The public descriptors describe band 5 and band 6 in short phrases. Below, each habit is the practical form of one of those phrases.

1. Task Response: the question is half answered

At band 5 the descriptors say the answer addresses the task only partially, and that main ideas are limited and not sufficiently developed. At band 6 the answer addresses all parts of the task, although some parts may be covered more fully than others.

This is the single most common reason a score stays at 5.5, and it has nothing to do with English. A two part question gets one part answered well and the other part forgotten. An opinion question gets both sides described and no opinion stated.

Band 5 behaviour. The prompt asks whether a development is positive or negative. The essay explains the development in detail and never says which it is.

Band 6 behaviour. The essay says, in the introduction and again in the conclusion, that it is mainly negative, and each body paragraph supports that.

The fix costs two minutes. Before writing, copy the question's instruction words onto your plan and number the parts. After writing, check each number has a paragraph. Half answered questions are usually not a decision; they are a part the candidate read and then forgot while writing.

2. Coherence and Cohesion: linking words instead of progression

At band 5 the descriptors mention a lack of overall progression and inadequate, inaccurate or over-use of cohesive devices. At band 6 ideas are arranged coherently with a clear overall progression.

This surprises candidates who have memorised a list of connectives. Overuse is named in the band 5 wording. Adding Moreover, Furthermore and In addition to the front of every sentence moves you towards 5, not towards 6, because the words claim a relationship the sentences do not have.

Before. Firstly, technology is very useful. Moreover, students can study online. Furthermore, it is cheap. In addition, teachers can also use it. Finally, it is the future.

Five sentences, five connectives, no argument. Each sentence restarts.

After. Technology has made study possible for people who cannot attend a campus. A student in a rural area can now take the same course as one in a capital city, and the cost of doing so is a fraction of the cost of moving. That access is what makes the change significant.

The second version has one connective in it. It progresses because each sentence develops the one before it rather than announcing itself.

3. Lexical Resource: a narrow range, stretched

At band 5 the descriptors describe a limited range of vocabulary that is minimally adequate for the task, with noticeable errors in spelling and/or word formation. At band 6 the range is adequate, with attempts at less common vocabulary and errors that do not impede communication.

Two separate habits sit here, and they pull in opposite directions.

The first is repetition. The word from the question appears in every sentence, unchanged, twenty times.

The second is the overcorrection: a memorised, very long phrase dropped into the wrong place. In this contemporary epoch of globalisation is not stronger than today. A marker reads it as a phrase learned for the exam rather than as range, and a phrase used wrongly is an error, not a flourish.

Word formation deserves its own line, because it is named in the descriptor and it is cheap to fix. Learn the family, not the word: pollute, pollution, polluted, pollutant. Most errors at this level are a noun used as a verb, or an adjective used as a noun.

4. Grammatical Range and Accuracy: complex sentences that break

At band 5 the descriptors say the writer uses only a limited range of structures and attempts complex sentences but these tend to be less accurate than simple sentences, with errors that can cause some difficulty for the reader. At band 6 there is a mix of simple and complex sentence forms, with errors that rarely reduce communication.

Read the band 5 wording again, because it describes something specific: the complex sentences are worse than the simple ones. That is the signature of a 5. The candidate can write correct short sentences, and attempts a long one to show range, and the long one has three errors in it.

Before. Although many people believes that the government should to spend money on public transport, but in my opinion is better to improve the roads.

Four problems in one sentence: subject and verb agreement, should to, although with but, and a missing subject before is better.

After. Many people believe that the government should spend money on public transport. In my opinion, improving the roads matters more.

Two correct sentences score better than one broken long one. Then build complexity back in deliberately, one structure at a time.

Then. Although many people believe that the government should spend more on public transport, improving the roads would benefit a larger number of travellers.

That is one complex sentence, one structure, correct. Range at band 6 is a mix of simple and complex forms, not a wall of subordinate clauses.

How to find your own two criteria

You do not have four problems. You have one or two, and they are in every answer you write.

1. Take three answers you have already written. Not new ones. The habits are in the old ones.
2. Mark each answer against the four criteria separately and write a whole number for each. No halves.
3. Look for the criterion that is lowest in all three. That is the one holding your score.
4. Fix that one habit only, for two weeks, then remark.

Working on all four at once is how a 5.5 becomes a 5.5 again. The criteria are marked independently, so a half band of improvement spread thinly across all four may round to nothing, while a whole band on one criterion moves the average.

What to do today

• Write the four criteria names on one line and put a whole number next to each, from memory, for your last answer. The one you are least sure about is usually the one you have never checked.
• Take your most recent essay and count the connectives at the start of sentences. Delete half of them and read it again.
• Find the longest sentence you have written this week and split it in two. If both halves are correct and the original was not, that is the habit from section four.

Marking your own writing on four criteria at once is hard, and the criterion you are weakest at is the one you are least able to judge. Pasting an answer into Bandly returns an estimated band for each of the four criteria separately, plus the sentences behind each estimate, which is what turns "about 5.5" into a specific habit to work on. The bands are estimates produced by a scoring engine, not marks from a qualified examiner, so treat them as a way to find your weakest criterion rather than as a predicted test result.