← 返回资讯屏轩教育 · 行业资讯

How IELTS Writing and Speaking Are Marked: The Official Band Descriptors Explained | 雅思写作与口语如何评分：官方评分标准说明

2026-10-15 08:06 · 本文部分内容由AI辅助创作

📚 How IELTS Writing and Speaking Are Marked: The Official Band Descriptors Explained | 雅思写作与口语如何评分：官方评分标准说明

IELTS Writing and Speaking are not marked by general impression. Every script and every interview is judged against the published band descriptors, which are public documents available from ielts.org. The descriptors list nine bands and four criteria, and they tell you exactly what an examiner must be able to see before awarding a particular score. If you understand these descriptors, you stop guessing and start training the specific behaviours that raise your band.

雅思写作与口语并非凭整体印象打分。每一份作文、每一次口试，都依据官方发布的评分标准（band descriptors，可在 ielts.org 公开查阅）进行评判。该标准列出九个分数段与四项评分维度，明确说明考官必须"看到"什么，才能给出某个分数。理解这套标准，你就不再靠猜，而是有针对性地训练能真正提分的具体表现。

1. Where the Band Descriptors Come From | 评分标准的来源与作用

IELTS is produced by the British Council, IDP: IELTS Australia and Cambridge University Press and Assessment. All examiners complete training and are certified, then re-standardised at regular intervals. Writing scripts are marked by certified examiners, and a proportion of scripts is double-marked for quality control. Speaking is assessed by the examiner who conducts the interview, using the same four criteria on a standardised rating scale.

雅思由英国文化教育协会、IDP：IELTS 澳大利亚与剑桥大学出版与考评院共同主办。所有考官都需完成培训与认证，并定期接受标准化复核。写作答卷由认证考官评分，其中一定比例会被双评以进行质量控制；口语则由主持考试的考官根据同一套四项标准、使用标准化的评分量表进行评定。

The band descriptors come in two versions. The "public" version is what you can download and read; the examiners' version is more detailed and adds notes on typical borderline scripts. Both describe the same four criteria, so the public version is a completely reliable study tool.

评分标准有两种版本。公开版本供考生下载阅读；考官版本更细致，并附有对典型"临界"答卷的说明。两者描述的都是同样四项标准，因此公开版本完全可以作为可靠的备考依据。

One important principle: descriptors are positive statements. A band is awarded when the examiner can see the features listed for that band most of the time, not merely once. A single good paragraph or one fluent answer does not lift a band.

一个重要原则是：评分标准的表述都是肯定性的。只有当考官在大部分时间都能看到该分数段所列特征时，才会给出这个分数，偶尔出现一次并不算。写出一段好段落或答出一段流利回答，并不能把分数抬上去。

2. The Four Criteria at a Glance | 四项评分标准一览

Writing uses four criteria, and Speaking uses four criteria. Three of them share the same names: Coherence and Cohesion, Lexical Resource, and Grammatical Range and Accuracy. The first criterion differs according to task type, and Speaking replaces it with Fluency and Coherence plus Pronunciation.

写作有四项标准，口语也有四项标准。其中三项同名：连贯与衔接、词汇资源、语法多样性与准确性。第一项标准按任务类型不同而变化，口语则以流利与连贯及发音取代之。

Criterion Writing Speaking 中文
Task fulfilment Task Achievement (T1) / Task Response (T2) Fluency and Coherence 任务完成／任务回应／流利与连贯
Organisation Coherence and Cohesion Fluency and Coherence 连贯与衔接
Vocabulary Lexical Resource Lexical Resource 词汇资源
Grammar Grammatical Range and Accuracy Grammatical Range and Accuracy 语法多样性与准确性
Sound — Pronunciation 发音

In both Writing and Speaking the four criteria carry equal weight. There is no bonus for length, for accent, or for using difficult words. Each criterion is scored on a whole or half band, and the four scores are averaged.

写作与口语中，四项标准权重完全相同。篇幅长、口音好、用难词，都没有额外加分。每项按整分或半分评定，四项取平均。

3. Writing Task 1: Task Achievement | 写作 Task 1：任务完成度

Task Achievement asks whether you have done what the task asked, in the right form. For Academic Task 1 this means presenting, describing and comparing data from a graph, chart, table, map or process diagram; for General Training Task 1 it means writing a letter that fulfils its purpose in an appropriate tone.

任务完成度考察你是否按要求、以正确文体完成了题目。学术类 Task 1 要求呈现、描述并比较图、表、地图或流程图中的数据；培训类 Task 1 要求写一封目的明确、语气得体的信件。

The single most important word in the Academic Task 1 descriptors is overview. From Band 7 upwards the descriptor requires a "clear overview of main trends, differences or stages". A Band 6 response "may have an overview but it may be unclear or not fully developed". A Band 5 response "recounts detail mechanically with no clear overview".

学术类 Task 1 标准中最关键的一个词是概述。Band 7 及以上要求"对主要趋势、差异或阶段有清晰的概述"；Band 6 的回答"可能有概述，但不清晰或不够充分"；Band 5 的回答"机械地罗列细节，没有清晰的概述"。

For General Training, the same criterion checks purpose, tone and the bullet points. All three bullets must be covered, the register must fit the reader, and a clear reason or request must run through the letter.

培训类任务中，同一标准考察写信目的、语气与提示要点。三个提示点必须全部覆盖，语体要与读者匹配，全信要有清晰的缘由或请求。

4. Writing Task 2: Task Response | 写作 Task 2：任务回应

Task Response asks four questions: Have you addressed all parts of the question? Is your position clear throughout? Are your main ideas extended and supported? Are the ideas relevant? Task 2 carries twice the weight of Task 1, so weakness here is expensive.

任务回应考察四个问题：你是否回答了题目的所有部分？你的立场是否始终清晰？主要观点是否得到延伸与支撑？观点是否切题？Task 2 的分值是 Task 1 的两倍，因此这一项的失分代价很大。

Two practical consequences follow. First, an opinion essay must state an opinion early and hold it to the end. Second, a two-part question needs both parts answered with roughly equal weight - "discuss both views and give your own opinion" means three moves, not two.

由此得出两点实操结论。第一，观点类作文必须尽早表明立场并贯穿全文。第二，双问题型题目两部分都要答，且分量大致相当——"讨论双方观点并给出你自己的看法"包含三个动作，而不是两个。

5. Coherence and Cohesion | 连贯与衔接

BASIC READERS：美国学校现代英语阅读教材（套装共7册·英文原版）
BASIC READERS：美国学校现代英语阅读教材（套装共7册·英文原版）
¥120.0
📱 去京东购买
京东精选

Coherence is logical order; cohesion is the linguistic glue that makes that order visible. The descriptors reward progression and paragraphing far more than the number of linking words. At the top bands, cohesion should be invisible: Band 9 states that the writer "uses cohesion in such a way that it attracts no attention".

连贯指逻辑顺序，衔接指让这种顺序可见的语言黏合剂。评分标准奖励的是"推进"与"分段"，而不是连接词的数量。在高分段，衔接应当是无痕的：Band 9 的表述是"使用衔接手段的方式不引人注意"。

A common Band 6 trap is mechanical linking: "Firstly... Secondly... Moreover... In addition... Finally..." in every paragraph. Examiners note this as over-use. Better cohesion comes from referencing (this trend, such measures), substitution (do so, the former), and topic sentences that carry meaning.

常见的 Band 6 陷阱是机械连接：每段都是"Firstly... Secondly... Moreover... In addition... Finally..."。考官会把这视为过度使用。更好的衔接来自指代（this trend、such measures）、替代（do so、the former），以及真正承载意义的主题句。

6. Lexical Resource | 词汇资源

Lexical Resource rewards range, precision and natural collocation - not rarity. The descriptors repeatedly use words such as "flexibility", "precision", "style" and "collocation", and they penalise errors in spelling and word formation separately from errors in word choice.

词汇资源奖励的是广度、精准与自然搭配，而不是生僻。标准中反复出现"灵活性""精准""语体""搭配"等词，并把拼写与构词错误同选词错误区分开来扣分。

This is why a thesaurus is a dangerous tool. "The government should alleviate the problem of traffic" is weaker than "the government should reduce traffic", because the second collocates properly and is used with control. Precision beats decoration.

这就是为什么滥用同义词词典很危险。"The government should alleviate the problem of traffic"不如"the government should reduce traffic"，因为后者的搭配正确、使用有掌控力。精准胜过堆砌。

7. Grammatical Range and Accuracy | 语法多样性与准确性

This criterion is deliberately two-dimensional. Range means a variety of sentence forms - simple, compound and complex - used where they are needed. Accuracy means the proportion of error-free sentences and the severity of the errors that remain.

这一标准刻意设计为两个维度。多样性指句型丰富——简单句、并列句、复合句——且在需要时使用；准确性指无错句子的比例，以及剩余错误的严重程度。