📚 Pandemic Learning Loss and Tutoring Support: A Survey of 16-19 Students | 疫情学习损失与补课支持：16至19岁学生调查
The COVID-19 period was not a short interruption for sixth-form and college students. For learners aged 16 to 19, it landed exactly on the two years that carry the highest academic stakes: the transition into advanced level study, the build-up to linear exams, and the production of coursework and portfolios. This article examines what surveys of this age group consistently report about learning loss, how tutoring was funded and delivered as a response, and what the evidence says about which forms of support actually worked. It combines published research from bodies such as Ofqual, Ofsted, the Education Policy Institute, the Sutton Trust and the Education Endowment Foundation with the structure of a school-level survey that any sixth form or college can replicate.
对16至19岁的学生而言，新冠疫情并不是一次短暂的中断。它恰好落在学业风险最高的两年：从GCSE升入高阶课程的关键过渡期、线性考试的准备期，以及课程作业与作品集的产出期。本文梳理针对这一年龄段的各类调查在”学习损失”上的共同结论，分析补课支持的资金来源与实施方式，并讨论哪些形式的支持在证据上真正有效。文章结合 Ofqual、Ofsted、Education Policy Institute、Sutton Trust 与 Education Endowment Foundation 等机构已发表的研究，并给出任何一所高中或学院都可以复制的校内调查框架。
1. Why 16-19 Learners Need Their Own Survey | 为什么16至19岁学生需要单独调查
Most learning-loss research focuses on primary and lower secondary pupils, where standardised tests make measurement easier. Key Stage 5 is different: the cohort is small, the curriculum is specialised, and outcomes depend heavily on independent study habits rather than on supervised classroom time. A survey aimed at 16-19 learners therefore has to ask different questions from one aimed at Year 6 or Year 9.
大多数关于学习损失的研究聚焦于小学和初中阶段，因为那里有标准化测试，测量更容易。Key Stage 5 完全不同：学生人数少、课程高度专业化，学习成果更多依赖自主学习习惯，而不是被监督的课堂时间。因此，面向16至19岁学生的调查，必须提出与面向六年级或九年级学生不同的问题。
The second reason is timing. Disruption hit this cohort at the point of transition from GCSE to A-Level, IB, BTEC or equivalent qualifications. Students who never experienced a normal first term of Year 12 lost the induction into essay writing, laboratory technique, source analysis and academic reading that normally happens in that term.
第二个原因是时间点。中断恰好发生在从GCSE升入A-Level、IB、BTEC或同等资历的过渡阶段。没有经历过正常Year 12第一学期的学生，错过了通常在那一学期完成的学术入门训练：论文写作、实验技术、史料分析以及学术阅读方法。
The third reason is stakes. A two-year course with linear assessment leaves little room for recovery. Unlike younger pupils, who can catch up across several academic years, a Year 13 student has roughly one year of usable time before final assessment.
第三个原因是风险。两年制课程配线性评估，几乎没有恢复空间。年幼的学生可以跨越几个学年慢慢追赶，而Year 13的学生在最终评估之前，大约只剩一年的可用时间。
Finally, 16-19 students are old enough to report their own experience accurately, which makes self-report survey data unusually valuable. They can describe what they missed, what they did instead, and what support they received in precise terms.
最后，16至19岁的学生已经有能力准确描述自身经历，这使自陈式调查数据在这一年龄段格外有价值。他们能具体说明自己错过了什么、当时做了什么替代、以及获得过哪些支持。
2. Designing the Survey: Sample, Instrument and Ethics | 调查设计：样本、工具与伦理
A credible survey of this kind needs a stratified sample. If you only survey one school, you are describing that school, not the cohort. A workable design draws students from a mix of state comprehensives, selective schools, sixth form colleges, independent schools and further education colleges, across at least three regions with different levels of deprivation.
可信的调查需要分层抽样。如果只调查一所学校，你描述的是那所学校，而不是整个群体。可行的设计应当覆盖公立综合中学、选拔性学校、六年级学院、私立学校与继续教育学院，并跨越至少三个贫困程度不同的地区。
For a total population of roughly 500,000 students in a national cohort, a sample of about 2,400 gives a margin of error near ±2 percentage points at 95% confidence, which is adequate for headline figures.
假设全国该年龄段学生总数约为50万人，抽取约2400人的样本，在95%置信水平下，误差范围约为 ±2个百分点，用于报告总体数据已经足够。误差范围可用下式估算：
Margin of error = ±1.96 × √( p × (1 − p) ÷ n )
With p = 0.5 and n = 2,400, the calculation gives ±0.020, that is ±2.0 percentage points. Subgroup analysis, for example comparing disadvantaged and non-disadvantaged students, needs larger numbers because the margin of error grows as the subgroup shrinks.
当 p = 0.5、n = 2400 时，计算结果为 ±0.020，即 ±2.0个百分点。若要做亚组分析（例如比较弱势学生与非弱势学生），则需要更大的样本量，因为亚组越小，误差范围越大。
The instrument itself usually combines four elements: a Likert-scale section on perceived learning loss by subject, a factual section on tutoring received (hours, provider, format, cost), a wellbeing section using validated short scales, and an open-response section. Adding attendance and assessment data from school records turns self-report into a triangulated dataset.
调查工具通常包含四个部分：按学科评估”感知学习损失”的李克特量表；关于所获补课的事实性问题（时数、提供方、形式、费用）；使用经验证短量表的心理健康部分；以及开放式问题。若能从学校记录中补充出勤与测评数据，自陈数据就可以被三角验证。
Ethics are not optional. Participants under 18 require guardian consent, all responses must be anonymous, and questions about mental health must be accompanied by signposting to pastoral support. A survey that measures distress without offering help is ethically incomplete.
伦理问题不可省略。未满18岁的参与者需要监护人同意，所有回答必须匿名，涉及心理健康的问题必须同时提供校内关怀支持的指引。一项只测量痛苦却不提供帮助的调查，在伦理上是不完整的。
| Sample characteristic | 样本特征 | Target proportion | 目标比例 | Why it matters | 为什么重要 |
|---|---|---|
| Disadvantaged (Pupil Premium / FSM eligible) | 弱势学生（曾获补助） | 25-30% | Losses are uneven; averages hide the gap | 损失并不均等，平均值会掩盖差距 |
| STEM vs humanities vs practical subjects | 理科 vs 人文 vs 实践类学科 | Balanced | Loss differs by subject type | 不同学科类型的损失不同 |
| School type (comprehensive, selective, college, independent) | 学校类型 | Reflect national mix | Tutoring access varies sharply | 补课可及性差异极大 |
| Region and deprivation decile | 地区与贫困十分位 | At least 3 regions | Local labour markets and internet quality differ | 当地就业与网络质量不同 |
3. Headline Findings: Where the Loss Concentrated | 核心发现：损失集中在哪些学科
Surveys of this cohort repeatedly show the same pattern: self-reported loss is largest in mathematics, the sciences and modern foreign languages, and smallest in essay-based humanities subjects where independent reading partially substituted for lessons. Practical and portfolio subjects report losses as large as mathematics, but for different reasons.
针对这一群体的多项调查反复呈现同一模式：自陈损失最大的是数学、自然科学和现代外语；以论文为主的人文学科损失最小，因为独立阅读在一定程度上替代了课堂。实践类与作品集类学科报告的损失与数学相当，但原因不同。
Mathematics suffers because it is cumulative. If a student never securely learns algebraic manipulation or calculus fundamentals, later topics become inaccessible. The dependency is sequential, so a gap at one point propagates forward.
数学之所以受损，是因为它具有累积性。如果学生没有牢固掌握代数运算或微积分基础，后续内容就变得无法理解。知识点之间是顺序依赖的，因此某一处的缺口会向前传导。
Modern languages suffer because they depend on spoken interaction, and remote learning removed almost all low-stakes oral practice. A student can memorise vocabulary alone; it is much harder to develop fluency and listening speed alone.
现代外语受损，是因为它高度依赖口语互动，而线上学习几乎消除了所有低风险的口语练习。学生可以独自背单词，但很难独自培养流利度与听力反应速度。
Sciences suffer because of practical work. Experimental technique, apparatus handling and data evaluation cannot be learned from a video of someone else doing it, and many students entered Year 13 having completed very few full practicals.
自然科学受损则源于实验课。实验技术、仪器操作与数据评估，无法通过观看他人操作的视频来掌握，而许多学生进入Year 13时，完整实验的次数极少。
| Subject group | 学科组 | Median self-reported lost teaching weeks | 自陈损失教学周中位数 | % reporting a “large” gap | 自认”缺口很大”的比例 |
|---|---|---|
| Mathematics | 数学 | 6 | 54% |
| Modern foreign languages | 现代外语 | 6 | 52% |
| Sciences | 自然科学 | 5 | 49% |
| Business and economics | 商科与经济学 | 4 | 33% |
| History and English | 历史与英语 | 3 | 27% |
| Practical and portfolio subjects | 实践与作品集类 | 7 | 58% |
The figures above are indicative of the pattern reported across published surveys rather than a single dataset, and any school running its own survey should expect local variation. The important methodological point is that self-reported loss and objective loss are not the same thing.
上表中的数字反映的是各类已发表调查呈现的整体模式，而非某一单独数据集；任何开展校内调查的学校都应当预期本地数据存在差异。关键的方法论要点是：自陈损失与客观损失并不等同。
Self-report tends to be inflated for subjects where students feel anxious and compressed for subjects where students feel confident. Wherever possible, pair perception data with actual assessment evidence, such as a diagnostic test or a marked past paper.
在学生感到焦虑的学科上，自陈损失往往被高估；在学生感到自信的学科上则被压缩。只要条件允许，就应当把感知数据与实际测评证据配对使用，例如一次诊断性测试或一份已批改的历年真题。
4. Practical, Oral and Portfolio Losses Are Hardest to Repair | 实践、口语与作品集的损失最难弥补
Practical subjects were hit differently. A student cannot complete a titration, a sculpture or a music performance remotely, and coursework evidence depends on access to studios, workshops, laboratories and rehearsal space. Interruptions here do not just reduce learning; they reduce the material available for assessment.
实践类学科的受损方式不同。学生无法远程完成滴定实验、雕塑创作或音乐演奏，而课程作业的证据依赖于工作室、车间、实验室与排练场地的使用权。这里的中断不仅减少了学习，也减少了可用于评分的材料。
Oral skills show a similar pattern. Debating, discussion, oral examinations and presentations disappeared for long periods, and students who returned reported lower confidence speaking in front of others even where their written work held up.
口语能力呈现类似规律。辩论、讨论、口试与演讲长时间消失，即便书面作业水平维持住了，返校后的学生在公开表达上的信心也明显偏低。
These deficits are the most expensive to fix because they need facilities, specialist staff and time in the timetable. A maths gap can be addressed in a small tutorial room; a practical gap often cannot.
这类缺口的修复成本最高，因为它们需要场地、专业教师以及课表中的时间。数学的缺口可以在小教室的辅导课里解决，实践技能的缺口往往不行。
Schools that recovered fastest typically re-sequenced the practical curriculum, front-loading essential techniques and moving written analysis later, rather than trying to cover everything in the original order.
恢复最快的学校通常重排了实践课程顺序：把核心技术提前，把书面分析后移，而不是按原顺序试图覆盖全部内容。
5. Wellbeing, Motivation and Exam Confidence | 心理健康、动力与考试信心
Learning loss and wellbeing are entangled. Surveys of 16-19 students consistently find that those reporting the largest academic gaps also report the highest anxiety, the lowest motivation and the poorest sleep. Causation runs in both directions.
学习损失与心理健康彼此纠缠。针对16至19岁学生的调查一致发现：自认学业缺口最大的人，同时报告的焦虑最高、动力最低、睡眠最差。因果关系是双向的。
Anxiety reduces working memory capacity in high-stakes situations, which lowers performance independently of knowledge. In other words, a student can know the content and still underperform because of assessment conditions.
焦虑会降低高压情境下的工作记忆容量，从而在知识水平不变的情况下拉低表现。换句话说，学生可能已经掌握了内容，却仅因考试情境而表现不佳。
A specific feature of this cohort is the sense of a “lost two years” that cannot be repeated. Students who lost sports seasons, performances, trips, competitions and social milestones report a grief-like response that is distinct from ordinary academic stress.
这一群体有一个特殊现象：感觉”失去了无法重来的两年”。错过运动赛季、演出、旅行、竞赛与社交里程碑的学生，报告的是一种类似丧失的哀伤反应，与日常学业压力性质不同。
Motivation decay is measurable over time. Many students began remote learning with high engagement and declined steadily across successive lockdowns. A simple proxy for this in survey data is the change in reported study hours per week across terms.
动力的衰减在时间上可以测量。许多学生在线学习初期参与度很高，之后随着一轮又一轮封锁持续下滑。调查数据中一个简单的代理指标，就是每周学习时数在不同学期之间的变化量：
Δ Study hours = Hours (Term 3) − Hours (Term 1)
A negative value indicates motivational decay and is itself a warning sign for exam performance, because total study time is one of the strongest predictors of final grades.
若结果为负值，说明动力在衰减，这本身就是考试成绩的风险信号，因为总学习时长是预测最终成绩最有力的因素之一。
6. Unequal Impact: the Gap Widened | 不平等影响：差距被拉大
The single most consistent finding across all surveys of this age group is that the pandemic widened existing gaps. Students from disadvantaged backgrounds lost more, and the difference was visible even after controlling for prior attainment.
在所有针对该年龄段的调查中，最一致的发现是：疫情拉大了原本存在的差距。来自弱势背景的学生损失更大，而且即使在控制了先前成绩之后，这种差异依然存在。
The mechanisms were practical. Home study space, a quiet room, a reliable internet connection, a personal device, and an adult at home who could help or simply enforce routine were unevenly distributed. Sharing a single laptop with siblings is a measurable educational disadvantage.
其作用机制非常具体。独立的学习空间、安静的房间、稳定的网络、专属设备、以及能够在旁协助或至少监督作息的成年人，这些条件的分布极不均衡。与兄弟姐妹共用一台笔记本电脑，本身就是一种可测量的教育劣势。
Employment mattered too. Many 16-19 students increased paid work hours during the pandemic because family incomes fell, and those hours competed directly with study time. Surveys frequently report that disadvantaged students worked longer weekly hours than their peers.
就业也是重要因素。疫情期间许多16至19岁的学生因为家庭收入下降而增加了打工时数，这些时间直接挤占学习。调查常报告：弱势学生每周工作时数高于同龄人。
The size of the gap can be expressed simply as the difference in mean attainment between groups:
差距的大小可以用两组平均成绩之差简单表示：
Gap G = M (advantaged) − M (disadvantaged)
Reporting this as a raw difference is useful for communication, but reporting it as an effect size divided by the pooled standard deviation allows comparison across subjects and years.
用原始差值报告便于沟通，但若除以合并标准差、转换成效应量，就能在不同学科与不同年份之间进行比较。
There is also a reporting effect worth noting. Disadvantaged students are less likely to respond to optional surveys, which means the true gap may be larger than the survey implies. High response rates and deliberate over-sampling are therefore essential to accuracy.
还有一个值得注意的回应偏差。弱势学生更少回应自愿参与的调查，这意味着真实差距可能比调查显示的更大。因此，高回应率与有意的超额抽样对数据准确性至关重要。
7. What Tutoring Support Was Actually Available | 实际存在哪些补课支持
Tutoring became the flagship policy response for this cohort. In England, the National Tutoring Programme was funded at approximately £1 billion across three years, and a separate 16-19 Tuition Fund provided around £100 million per year directly to schools and colleges for small-group tuition in English and mathematics, and for other subjects where disruption was severe.
补课成为针对这一群体的旗舰政策应对。在英格兰，National Tutoring Programme 三年间投入约10亿英镑；另有一个独立的 16-19 Tuition Fund，每年直接向学校与学院提供约1亿英镑，用于英语和数学的小组补课，以及中断严重学科的支持。
The 16-19 fund had an important design feature: it required tuition to be delivered in small groups and targeted at students below a certain GCSE attainment threshold. That targeting was intended to concentrate resources where the gap was widest.
16-19基金有一个重要的设计特征：它要求补课必须以小组形式开展，并针对GCSE成绩低于某一门槛的学生。这种定向设计意在把资源集中到差距最大处。
Beyond government schemes, four other delivery models were widely used: school-employed tutors, external tutoring agencies, academic mentors placed in schools, and peer or near-peer tutoring by older students. Online one-to-one platforms expanded rapidly and continued afterwards.
除政府方案之外，还有四种被广泛采用的实施模式：学校自聘导师、外部补课机构、驻校学业辅导员，以及由高年级学生承担的同辈辅导。线上一对一平台迅速扩张，并在之后延续下来。
| Model | 模式 | Typical dosage | 典型剂量 | Cost per hour | 每小时成本 | Evidence strength | 证据强度 |
|---|---|---|---|
| One-to-one, qualified teacher | 一对一，持证教师 | 1 hour, 1-2× weekly | High | Strong but expensive | 强，但成本高 |
| Small group (3-6) | 小组（3-6人） | 1 hour, 1-2× weekly | Moderate | Strong, cost-effective | 强，性价比高 |
| Academic mentor in school | 驻校学业辅导员 | Multiple
Published by TutorHao | Exam Prep Revision Series | aleveler.com 更多咨询请联系16621398022（同微信） CommentsMore posts |
屏轩国际教育cambridge primary/secondary checkpoint, cat4, ukiset,ukcat,igcse,alevel,PAT,STEP,MAT, ibdp,ap,ssat,sat,sat2课程辅导，国外大学本科硕士研究生博士课程论文辅导