📚 Elite University Admissions Tests Tutoring | 名校附加考试辅导
Getting a strong A-Level or IB profile is no longer enough for the most competitive courses in the UK. Oxford, Cambridge, Imperial, LSE, UCL and Warwick now screen applicants with their own admissions tests - the MAT, PAT, ESAT, TMUA, STEP, TSA and LNAT - and these papers are sat months before results day, on a fixed national date, with no second chance. This guide explains what those tests are, how they differ from school exams, and how structured tutoring turns a nervous candidate into a calm, well-drilled one.
在英国最热门的专业申请中，仅有漂亮的 A-Level 或 IB 成绩已经不够。牛津、剑桥、帝国理工、LSE、UCL、华威等院校纷纷用自设的附加考试来筛选申请者——MAT、PAT、ESAT、TMUA、STEP、TSA、LNAT——这些考试在放榜前几个月举行，全国统一日期，一年只有一次机会。本文说明这些考试是什么、与校内考试有何不同，以及系统化辅导如何把紧张的学生训练成沉着、熟练的应考者。
1. What Additional Admissions Tests Really Are | 附加考试到底是什么
Additional admissions tests are university-owned examinations sat in addition to, not instead of, your school qualifications. They are not designed to check whether you have covered the syllabus - your A-Levels already do that. They are designed to separate the top five per cent of applicants from each other, which means the questions are unfamiliar, the time limit is brutal, and the mark scheme rewards insight rather than routine.
附加考试是大学自行命题的考试，它与中学学历并行，而不是替代它。它不是用来检查你是否学完大纲——那是 A-Level 的任务。它的作用是把你和其他最顶尖的 5% 申请者区分开，因此题目陌生、时间极紧，评分标准奖励的是洞察力而非套路。
Three features define almost every one of these papers. First, they reward problem-solving over recall: you will rarely be asked to state a formula, but you will be asked to use it in a situation you have never seen. Second, they are sat under computer-based or tightly invigilated conditions at fixed centres, so practical logistics matter. Third, they feed directly into shortlisting, which means a weak score can end an application before a personal statement is ever read properly.
这类试卷几乎都具备三个共同特征。第一，考查解题能力而非记忆：很少要求你背公式，而是要求你把它用在一个从未见过的新情境里。第二，多数为机考或在固定考点严格监考，因此流程细节也很重要。第三，成绩直接决定能否进入面试筛选，一份低分可能在个人陈述被认真阅读之前就让申请终止。
Understanding this changes how you prepare. Revision alone is not enough; you need diagnosis, targeted teaching, timed practice and honest review, repeated in a cycle until speed and accuracy both hold up under pressure.
理解这一点会改变你的备考方式。单纯刷题不够，你需要诊断、针对性讲解、限时训练和诚实复盘，循环往复，直到速度与准确率在压力下同时站得住。
2. The Test Landscape: Which Exam for Which Course | 考试版图:哪个专业考哪一门
Different universities use different tests, and the same university may use several across different departments. The table below summarises the main papers currently in use for UK applications, together with the courses that typically require them.
不同大学使用不同考试，同一所大学的不同院系也可能各用一套。下表汇总了目前英国申请中最常用的主要考试，以及通常要求这些考试的专业。
| Test | Typical courses | Format |
| MAT | Oxford and Imperial Mathematics, Computer Science, joint honours | 2 h 30 min; multiple-choice plus longer written questions |
| PAT | Oxford Physics, Engineering Science, Materials Science | 2 h; mathematics and physics, computer-based |
| ESAT | Cambridge and Imperial Engineering, Natural Sciences, Physics, Chemical Engineering, Veterinary Medicine | Modular; each module 27 questions in 40 min, candidates normally sit Mathematics 1 plus two others |
| TMUA | Cambridge and Imperial Economics and Computer Science, LSE, Warwick, Durham, Bath, Southampton | Two papers, 75 min each, 20 multiple-choice questions per paper |
| STEP | Cambridge Mathematics, Imperial, Warwick | STEP 2 and STEP 3, 3 h each, 12 questions, answer 6 |
| TSA | Oxford PPE, Economics and Management, Experimental Psychology, History and Politics | Section 1: 50 multiple-choice in 90 min; Section 2: 30 min essay (course dependent) |
| LNAT | Oxford, UCL, LSE, KCL, Durham, Bristol, Glasgow, Nottingham, SOAS Law | Section A: 42 multiple-choice in 95 min; Section B: 40 min essay |
| UCAT | Medicine and Dentistry at most UK medical schools | About 2 h, five subtests, computer-based |
The landscape keeps shifting, so always verify against the official page for your year of entry. The BMAT for medicine was withdrawn after the 2023 cycle, and the old NSAA and ENGAA were replaced by the ESAT in 2024. Oxford also runs smaller subject tests such as the HAT for History, the MLAT for Modern Languages, the PhilAT for Philosophy and the CAT for Classics. Rumours travel fast in school corridors; the admissions office website does not lie.
这个版图一直在变化，因此务必以你所申请年度的官网信息为准。医学方向的 BMAT 已在 2023 届之后停办，旧的 NSAA 与 ENGAA 于 2024 年被 ESAT 取代。牛津还设有较小的科目测试，例如历史的 HAT、现代语言的 MLAT、哲学的 PhilAT、古典学的 CAT。学校走廊里的传闻传得很快，招生办官网不会说谎。
Registration usually opens in late summer and closes in early autumn, with tests sat in the middle or second half of October, or in early November for STEP. Missing the registration window is the single most avoidable disaster in this whole process.
报名通常在夏末开放、初秋截止，考试多在 10 月中旬至下旬举行，STEP 则在 11 月初。错过报名窗口，是整个流程中最容易避免的一场灾难。
3. Why These Tests Decide So Much | 为什么这些考试如此关键
For Oxford and Cambridge, admissions tests are the primary filter before interview. A candidate with three A* predictions but a bottom-quartile MAT score is usually not interviewed for Mathematics, while a candidate with slightly weaker predictions and a top-decile score very often is. The test is treated as a standardised measure that levels out differences between schools, syllabuses and grade inflation.
对牛剑而言，附加考试是面试前的主要筛选工具。三门 A* 预估但 MAT 成绩落在最低四分位的申请者，通常拿不到数学系的面试；而预估稍弱但成绩进入前 10% 的申请者，往往能拿到。考试被视为一种标准化的衡量方式，可以抹平学校、课程体系与分数膨胀带来的差异。
For Imperial, LSE, Warwick, Durham, UCL and others, the test may appear as a condition in the offer, as a threshold for shortlisting, or as a tie-breaker between candidates with identical grades. Some departments publish only a "competitive score" range; others publish nothing at all and simply rank. Either way, the logic is the same: a high score buys you options, a low score removes them.
对帝国理工、LSE、华威、杜伦、UCL 等校，考试成绩可能写进有条件录取、作为进入下一轮的分数线，或在成绩相同的候选人之间充当决胜依据。有的院系只公布一个"有竞争力的分数区间"，有的什么都不公布，直接排名。无论哪种，逻辑都一样：高分给你选择权，低分把选择权拿走。
There is also a psychological dimension. Candidates who have never sat a test they cannot finish in comfort often panic in October. Candidates who have sat fifteen timed papers are bored by October, and boredom is a much better state to be in.
还有一个心理层面。从未经历过"做不完的考试"的学生，往往在十月慌神；已经做过十五套限时卷的学生，十月只会觉得无聊——而无聊是更好的状态。
4. Step One: Honest Diagnostic Assessment | 第一步:诚实的诊断测评
Good tutoring starts with measurement, not with teaching. In the first session we set a full past paper under real conditions: same duration, same calculator rules, no pauses, no hints. The raw score matters less than the pattern behind it, so every question is then classified into one of five categories.
好的辅导从测量开始，而不是从讲课开始。第一节课我们就按真实条件做一套完整真题：同样的时长、同样的计算器规定、不暂停、不提示。原始分数本身不如它背后的模式重要，因此每道题随后会被归入以下五类之一。
- Knowledge gap - the topic was never taught or has been forgotten.
- Method gap - the topic is known, but the standard approach was not recognised.
- Speed gap - the method was right, but the clock was the enemy.
- Precision gap - signs, units, algebra slips and misread questions.
- Strategy gap - poor question selection and time allocation across the paper.
- 知识缺口——该知识点从未学过或已经遗忘。
- 方法缺口——知识点会，但没有识别出标准解法。
- 速度缺口——思路正确，但时间不够。
- 精度缺口——符号、单位、代数变形和读题失误。
- 策略缺口——选题顺序与整卷时间分配失当。
These five categories need completely different remedies, and mixing them up wastes months. A knowledge gap is solved by teaching; a speed gap is solved by drilling; a strategy gap is solved by learning to abandon questions. We summarise the diagnostic with a simple readiness index, used only as an internal tracking tool.
这五类问题需要完全不同的处方，混淆它们会白白浪费几个月。知识缺口靠讲解，速度缺口靠重复训练，策略缺口靠学会放弃题目。我们用下面这个简单的准备度指数来概括诊断结果，它只是内部的追踪工具。
Readiness = 0.4 × accuracy (%) + 0.6 × completion rate (%)
Weighting completion more heavily reflects reality: in these papers, unanswered questions score zero regardless of how elegant the ones you did answer were. The diagnostic is repeated every four to six weeks so that progress is measured rather than assumed.
把完成率赋予更高权重，是为了贴近现实：在这类考试里，空着的题一律零分，无论你答的那几题多么漂亮。诊断每隔四到六周重复一次，让进步被测量，而不是被假设。
5. Closing the Content Gap Beyond A-Level | 补齐 A-Level 之外的知识缺口
Most admissions tests assume content that sits just outside, or at the very edge of, the standard A-Level syllabus. The gap is rarely huge, but it is decisive, because the questions that separate candidates are usually built exactly on those extra tools.
大多数附加考试所假设的知识，正好落在 A-Level 大纲的边缘或稍外一点。缺口通常不大，却具有决定性，因为那些拉开差距的题目，恰恰建立在这些额外工具之上。
For the MAT and TMUA, the recurring extras include elementary number theory (divisibility, modular reasoning, digit problems), inequalities and the behaviour of functions, graph sketching and transformation, sequences and series, counting and elementary probability, and the language of proof - necessary, sufficient, contradiction, counterexample. For the PAT, the extras cluster around mechanics treated with calculus, dimensional analysis, simple circuit reasoning, waves and superposition, and approximation without a calculator.
对 MAT 和 TMUA 来说，反复出现的补充内容有：初等数论（整除、同余思想、数字问题）、不等式与函数性质、图像绘制与变换、数列与级数、计数与初等概率，以及证明的语言——充分、必要、反证、反例。对 PAT 来说，补充内容集中在用微积分处理的力学、量纲分析、简单电路推理、波与叠加，以及无计算器条件下的估算。
For STEP, the biggest single gap is not content but fluency in writing multi-step proofs and in choosing a productive first move under uncertainty. For the TSA and LNAT, the "content" is entirely logical: argument structure, assumption, flaw, inference and the difference between a claim that is supported and one that merely feels true.
对 STEP 而言，最大的缺口不是知识，而是书写多步证明、以及在不确定中选出有效第一步的流畅度。对 TSA 和 LNAT 而言，"知识点"完全是逻辑性的：论证结构、隐含假设、逻辑谬误、推论，以及分清"被支持的主张"与"感觉上成立的主张"。
We teach these topics in short, dense blocks of forty to fifty minutes, each ending with three to five questions that use the idea immediately. Ideas that are not used within a day are usually lost within a week.
我们把这些内容放在四十到五十分钟的密集小单元里讲授，每个单元结束时立刻用三到五道题实践该思想。一个思想若不在当天被使用，通常一周之内就会消失。
6. Speed, Accuracy and the Arithmetic of Time | 速度、准确率与时间的算术
Time pressure is a designed feature, not an accident. Do the division and the design becomes obvious.
时间压力是刻意设计的，而不是意外。动手除一除，设计意图就一目了然。
TMUA: 75 min ÷ 20 questions = 3.75 min ≈ 225 s per question
ESAT module: 40 min ÷ 27 questions ≈ 89 s per question
TSA Section 1: 90 min ÷ 50 questions = 1.8 min ≈ 108 s per question
At roughly 90 seconds per question in the ESAT, there is no time to derive anything from first principles. The candidate must recognise a standard form within ten seconds, choose a route within twenty, and execute. This is a trained reflex, and it is built in a very specific way: same topic, short bursts, strict clock, immediate feedback.
在 ESAT 里平均每题约 90 秒，根本没有时间从基本原理推导。考生必须在十秒内辨认出标准形式，二十秒内选定路径，然后执行。这是一种被训练出来的条件反射，需要用非常具体的方式建立：同一主题、短时爆发、严格计时、即时反馈。
We use a three-stage drill. Stage one is untimed accuracy, where the only goal is a correct method. Stage two is generous timing, at roughly 150 per cent of exam pace. Stage three is true exam pace, and no question is allowed to consume more than three times the average allocation. Accuracy is monitored with a simple ratio that keeps both failure modes visible.
我们采用三阶段训练。第一阶段不计时，唯一目标是方法正确。第二阶段宽松计时，约为考试节奏的 150%。第三阶段按真实考试节奏，任何一题都不允许超过平均配时的三倍。准确率用下面这个简单比值监控，让两种失败模式同时可见。
Precision ratio = correct answers ÷ attempted questions × 100%
A candidate with a high precision ratio but a low completion rate is being too careful, and must be pushed to move. A candidate who finishes everything with a low precision ratio is guessing, and must be slowed down and forced to check. The two need opposite coaching, and only measurement can tell them apart.
准确率高但完成率低的考生过于谨慎，需要被推着往前走；全部做完但准确率低的考生在猜，需要放慢并强制检查。这两类人需要相反的训练方向，而只有测量才能把他们区分开。
7. The Tutoring Cycle: Plan, Teach, Practise, Review | 辅导周期:计划、讲解、练习、复盘
A productive tutoring relationship is not a series of lectures. It is a repeating four-phase cycle, and each week should visibly contain all four phases.
有效的辅导关系不是一连串讲座，而是一个反复运行的四阶段循环，每一周都应当看得见这四个阶段。
- Plan - set the week's target topic list and the exact papers to be attempted, with dates.
- Teach - fill the specific conceptual gap, forty to fifty minutes, no filler.
- Practise - timed work done independently, between sessions, with the clock running.
- Review - inspect the errors, reclassify them, and adjust next week's plan.
- 计划——确定本周目标题型清单和要做的具体试卷，并写明日期。
- 讲解——只针对那一个概念缺口，四十到五十分钟，不掺水分。
- 练习——在两次课之间独立完成限时训练，计时器必须开着。
- 复盘——逐题检查错误、重新分类，并据此调整下周计划。
The review phase is where most of the value is created and where most self-study collapses. A student who does four papers and reviews none of them carefully has practised making the same mistakes four times. A student who does two papers and spends two hours dissecting each error has genuinely improved.
复盘阶段创造的价值最多，也恰恰是自学最容易崩塌的环节。做四套卷却从不细致复盘的学生，只是把同样的错误练了四遍；做两套卷、每套花两小时逐题剖析错误的学生，才是真正在进步。
A useful discipline is the error log. Every mistake is recorded with four fields: the question reference, the category from Section 4, the correct idea in one sentence, and the trigger that should have prompted it. Reviewed weekly, forty entries become a personal map of weakness, far more useful than any generic revision guide.
一个很有用的纪律是错题本。每道错题记录四个字段：题号出处、第 4 节中的错误类别、用一句话写清的正确思路、以及本应触发该思路的线索。每周回看，四十条记录就会变成一张专属的弱点地图，比任何通用教辅都有用。
8. Past Papers, Mocks and Realistic Conditions | 真题、模考与真实条件
Past papers are the single most valuable resource, but only if they are used honestly. Sitting a paper with music on, pausing the clock, and checking the mark scheme mid-question produces a comfortable feeling and almost no improvement.
真题是最有价值的资源，但前提是诚实使用。边听音乐边做、中途停表、做到一半翻答案，只会带来舒适感，几乎不带来进步。
We recommend reserving at least six full papers for genuine mock conditions, and rationing them across the preparation period. Earlier papers, where the format has since changed, are still useful for topic drilling but
更多咨询请联系16621398022（同微信）
或截图后用「扫一扫 → 相册」选择本图