📚 Edexcel Maths Psychology | Edexcel 数学心理学
Psychology is often seen as a subject about human behaviour, but modern psychology depends heavily on statistical evidence. The Edexcel A-Level Mathematics course provides the core statistical tools needed to design experiments, summarise data, test hypotheses and draw valid conclusions in psychological research.
心理学常被视为关于人类行为的学科，但现代心理学在很大程度上依赖统计证据。Edexcel A-Level 数学课程提供了设计实验、汇总数据、检验假设并在心理学研究中得出有效结论所需的核心统计工具。
1. Why Psychology Needs Edexcel Maths | 为什么心理学需要 Edexcel 数学
Psychology experiments measure variables such as reaction time, memory score or anxiety level. Reaction time might be recorded in milliseconds, memory as the number of words recalled out of 20, and anxiety on a standardised questionnaire score. Every one of these measurements contains random variation, so psychologists cannot rely on intuition alone.
心理学实验测量反应时间、记忆分数或焦虑水平等变量。反应时间可能以毫秒记录，记忆以 20 个词中回忆出的个数表示，焦虑则以标准化问卷得分表示。这些测量都包含随机变异，因此心理学家不能仅靠直觉。
Edexcel statistics allow researchers to separate a real effect from chance. If a treatment group recalls two more words than a control group, a statistical test tells us how likely such a difference would be if the treatment actually did nothing at all.
Edexcel 统计方法使研究者能够将真实效应与偶然波动区分开。如果实验组比控制组多回忆出两个词，统计检验可以告诉我们：若处理实际上毫无作用，出现这样差异的可能性有多大。
The mathematics skills in Edexcel include probability, statistical distributions, correlation and hypothesis testing. Each of these appears in nearly every research paper in psychology, making the link between the two subjects very strong.
Edexcel 数学中的技能包括概率、统计分布、相关和假设检验。这些内容几乎出现在每一篇心理学研究论文中，因此两个学科之间的联系非常紧密。
- Probability - deciding how likely an event is under a stated assumption. | 概率——在给定假设下判断事件发生的可能性。
- Distributions - the normal distribution and the binomial distribution. | 分布——正态分布与二项分布。
- Correlation - measuring the strength of a relationship between two variables. | 相关——衡量两个变量之间关系的强度。
- Hypothesis testing - deciding whether to accept or reject a null hypothesis. | 假设检验——决定是否接受或拒绝零假设。
2. Types of Data and Measurement Scales | 数据类型与测量尺度
Psychological data can be qualitative or quantitative. Quantitative data are numerical, such as scores on a depression scale or reaction times in milliseconds; qualitative data are non-numerical, such as participants' descriptions of their mood in an interview.
心理数据可以是定性的或定量的。定量数据是数值型的，例如抑郁量表得分或以毫秒计的反应时；定性数据是非数值型的，例如参与者在访谈中对情绪的描述。
A second distinction concerns how the data were obtained. Primary data are collected first-hand by the researcher, while secondary data already exist, for example in official mental-health statistics. Edexcel data-handling questions expect you to identify both the type and the source of a data set.
第二个区分涉及数据的来源。一手数据由研究者亲自收集，二手数据则已经存在，例如官方的心理健康统计数据。Edexcel 的数据处理题要求你同时识别数据的类型与来源。
Measurement scales in psychology are often classified as nominal, ordinal, interval or ratio. Nominal data are categories without order, ordinal data have order but unequal gaps, interval data have equal gaps but no true zero, and ratio data have a true zero.
心理学中的测量尺度通常分为称名、顺序、等距和比率四类。称名数据是无顺序的类别，顺序数据有顺序但间隔不等，等距数据间隔相等但没有真实零点，比率数据则有真实零点。
| Scale | 尺度 | Example | 例子 | Key property | 关键性质 |
| Nominal 称名 | Diagnosis category, gender | No order; mode only 无顺序；只能求众数 |
| Ordinal 顺序 | Likert rating, fear ranking | Order, unequal gaps 有顺序，间隔不等 |
| Interval 等距 | Temperature in °C, IQ score | Equal gaps, no true zero 间隔相等，无真零点 |
| Ratio 比率 | Reaction time in ms, words recalled | True zero exists 存在真实零点 |
Edexcel questions may ask you to recognise whether a variable is nominal, ordinal, interval or ratio, and then to choose an appropriate average, graph or statistical test. Using the mean on ordinal data, for instance, is a classic error because the gaps between the categories are not equal.
Edexcel 题目可能要求你判断一个变量属于称名、顺序、等距还是比率，然后选择合适的平均数、图表或统计检验。例如，对顺序数据使用平均数就是典型错误，因为各类别之间的间隔并不相等。
3. Sampling and Representativeness | 抽样与代表性
Statistics only makes sense in relation to a population. Psychologists almost never test everyone in a population, so they draw a sample and use it to estimate population values. A biased sample produces conclusions that do not generalise.
统计学只有相对于总体才有意义。心理学家几乎不可能测试总体中的每一个人，因此他们抽取样本，并用样本估计总体参数。有偏的样本会产生无法推广的结论。
Edexcel covers random, systematic, stratified, opportunity and volunteer sampling, together with the idea of a sampling frame. In psychology these same methods appear whenever a study recruits participants, and examiners expect you to evaluate them rather than simply name them.
Edexcel 涵盖简单随机、系统、分层、机会和志愿抽样，以及抽样框的概念。在心理学中，研究招募参与者时使用的正是这些方法，考官期望你对其进行评价，而不仅仅是说出名称。
- Random sampling - every member has an equal chance; unbiased but hard to organise. | 随机抽样——每个成员机会均等；无偏但难以组织。
- Opportunity sampling - whoever is available; quick but often unrepresentative. | 机会抽样——谁方便就选谁；快速但常常不具代表性。
- Stratified sampling - the sample mirrors population proportions; more representative but needs prior knowledge. | 分层抽样——样本反映总体比例；更具代表性但需要事先掌握信息。
- Volunteer sampling - self-selected participants; may share personality traits such as willingness to help. | 志愿抽样——自我选择的参与者；可能具有共同特质，如乐于助人。
4. Descriptive Statistics: Summarising Psychological Data | 描述统计：汇总心理数据
Before any inferential test, psychologists summarise data using a measure of central tendency and a measure of dispersion. The mean, median and mode describe the typical score; the range, interquartile range and standard deviation describe how spread out the scores are.
在进行任何推断检验之前，心理学家会用集中量数和离散量数来汇总数据。平均数、中位数和众数描述典型得分；全距、四分位距和标准差描述得分的离散程度。
The standard deviation is the most informative measure of spread because it uses every score. It measures the typical distance of a score from the mean, so it is expressed in the same units as the original data.
标准差是信息量最大的离散量数，因为它用到了每一个得分。它衡量得分与平均数之间的典型距离，因此与原数据的单位相同。
s = √( Σ (x − x̄)² ÷ (n − 1) )
In Edexcel you must be able to calculate s from raw data and to explain what a large standard deviation means in context: it indicates that participants differed greatly from one another, so the treatment effect may be inconsistent.
在 Edexcel 中，你必须能够用原始数据计算 s，并在语境中解释较大的标准差意味着什么：它说明参与者之间差异很大，因此处理效果可能并不一致。
Choosing the right average matters. The mean is sensitive to extreme scores, the median is robust and suits ordinal data, and the mode is the only option for nominal categories.
选择合适的平均数很重要。平均数对极端值敏感，中位数稳健且适合顺序数据，而众数则是称名类别唯一可用的选择。
5. Probability and the Normal Distribution | 概率与正态分布
Probability in Edexcel is the language of uncertainty. If a participant guesses on every trial of a memory test, the probability of a correct answer is 0.5, and the binomial distribution describes how many correct answers are expected by chance alone.
Edexcel 中的概率是不确定性的语言。如果参与者在记忆测验的每一试次上都靠猜，那么答对的概率为 0.5，而二项分布描述了仅凭偶然可以预期答对多少题。
Many psychological variables, such as IQ or simple reaction time, are approximately normally distributed. The normal curve is symmetrical about the mean, and about 68% of scores lie within one standard deviation of the mean, 95% within two and about 99.7% within three.
许多心理变量，如智商或简单反应时，近似服从正态分布。正态曲线关于平均数对称，约 68% 的得分落在平均数一个标准差之内，95% 在两个标准差之内，约 99.7% 在三个标准差之内。
z = (x − μ) ÷ σ
are interpreted. A score of z = +1.5 on a screening test is above average but not extreme.
z 分数告诉你某个原始分数距离平均数有多少个标准差，标准化心理测评正是以此解释结果。在筛查测验中 z = +1.5 表示高于平均，但并非极端。
6. Correlation and Scatter Graphs | 相关与散点图
Correlation measures the strength and direction of a relationship between two co-variables. Pearson's r is used for interval or ratio data, while Spearman's rho is used when at least one variable is ordinal, which is common in psychology because rating scales dominate.
相关衡量两个共变量之间关系的强度和方向。皮尔逊 r 用于等距或比率数据，而当至少一个变量为顺序数据时使用斯皮尔曼 ρ，这在心理学中很常见，因为评定量表占主导地位。
A scatter graph makes the relationship visible and reveals outliers or non-linear patterns that a single correlation coefficient would hide. The coefficient lies between −1 and +1, where the sign shows direction and the magnitude shows strength.
散点图使关系可视化，并揭示单个相关系数会掩盖的异常值或非线性模式。系数介于 −1 与 +1 之间，符号表示方向，绝对值表示强度。
The most important warning in both subjects is that correlation does not imply causation. A negative correlation between sleep duration and anxiety symptoms may be explained by a third variable, or the causal direction may run the other way.
两个学科中最重要的警示是：相关不等于因果。睡眠时长与焦虑症状之间的负相关可能由第三变量解释，或者因果方向恰好相反。
.25, meaning only a quarter of the variation in one variable is associated with the other.
相关系数的平方得到 r²，即两个变量共享方差的比�例。若 r = 0.50，则 r² = 0.25，意味着一个变量的变异中只有四分之一与另一个变量相关。
7. Hypothesis Testing and Significance Levels | 假设检验与显著性水平
A hypothesis test begins with a null hypothesis (H₀) stating that there is no effect or no difference, and an alternative hypothesis (H₁) stating that there is. The test asks how unlikely the observed data would be if H₀ were true.
假设检验始于零假设（H₀），即不存在效应或差异；以及备择假设（H₁），即存在效应或差异。检验要回答的问题是：若 H₀ 为真，观测到当前数据的可能性有多大。
Psychologists conventionally use a significance level of α = 0.05, meaning a 5% lly true. This mistake is a Type I error; failing to reject a false H₀ is a Type II error.
心理学家通常采用 α = 0.05 的显著性水平，意味着当 H₀ 实际为真时拒绝它的风险为 5%。这种错误称为第一类错误；未能拒绝错误的 H₀ 则称为第二类错误。
You must also decide whether the test is one-tailed or two-tailed. A one-tailed test is used when a directional hypothesis is stated, and its critical value is smaller, making it easier to reach significance, but only in the predicted direction.
你还必须决定使用单尾还是双尾检验。当提出方向性假设时使用单尾检验，其临界值更小，更容易达到显著，但仅在预测的方向上有效。
The decision rule is to compare the observed value with the critical value from the table, using the correct number of participants (N) and significance level. If the observed value is more extreme than the critical value, the null hypothesis is rejected.
判定规则是将观测值与表中临界值进行比较，并使用正确的参与者人数（N）和显著性水平。若观测值比临界值更极端，则拒绝零假设。
8. Choosing the Right Statistical Test | 选择正确的统计检验
Choosing a test depends on three questions: what level of measurement the data are, whether the design is independent or repeated measures, and whether the hypothesis is directional. Edexcel and psychology exams both reward this systematic decision-making.
选择检验取决于三个问题：数据属于什么测量水平，设计是独立组还是重复测量，以及假设是否具有方向性。Edexcel 与心理学考试都奖励这种系统化的决策过程。
| Test | 检验 | Data / design | 数据与设计 |
| Sign test 符号检验 | Nominal, repeated measures 称名，重复测量 |
| Chi-squared χ² 卡方检验 | Nominal, independent groups 称名，独立组 |
| Spearman's rho ρ 斯皮尔曼 | Ordinal, correlation 顺序，相关 |
| Mann-Whitney U | Ordinal, independent groups 顺序，独立组 |
| Wilcoxon T | Ordinal, repeated measures 顺序，重复测量 |
| Unrelated t-test | Interval/ratio, independent 等距或比率，独立组 |
| Related t-test | Interval/ratio, repeated 等距或比率，重复测量 |
Non-parametric tests such as the sign test and chi-squared make fewer assumptions but are less powerful; parametric tests such as the t-test use more information and are more likely to detect a genuine effect.
符号检验和卡方检验等非参数检验假设更少，但统计效能较低；t 检验等参数检验利用了更多信息，更有可能检出真实效应。
9. Effect Size, Confidence and Reporting Results | 效应量、置信度与结果报告
Statistical significance is not the same as practical importance. With a large sample, a trivial difference can become significant, so psychologists also report an effect size such as Cohen's d or r² to show how large the effect actually is.
统计显著并不等于实际重要。在大样本中，微不足道的差异也可能变得显著，因此心理学家还会报告效应量，如科恩 d 或 r²，以显示效应实际有多大。
ic, the degrees of freedom and the probability value, for example t(28) = 2.31, p = 0.028.
结果通常以标准格式报告，包括平均数、标准差、检验统计量、自由度和概率值，例如 t(28) = 2.31，p = 0.028。
A confidence interval gives a range of plausible values for the population parameter. If a 95% confidence interval for the difference between two means does not include zero, the difference is significant at the 5% level.
置信区间给出总体参数可能取值的范围。如果两个平均数之差的 95% 置信区间不包含零，则该差异在 5% 水平上显著。
10. Exam Technique and Command Words | 考试技巧与指令词
Edexcel and psychology examiners use specific command words. Calculate means show your working and round appropriately; comment means interpret the result in context; justify means give a statistical reason; evaluate means weigh strengths against limitations.
Edexcel 与心理学考官使用特定的指令词。Calculate 意味着展示过程并恰当取整；comment 意味着结合语境解释结果；justify 意味着给出统计学理由；evaluate 意味着权衡优点与局限。
Always state units, such as milliseconds or percentage points, and give answers to the same level of accuracy as the data. In psychology papers, unfinished sentences such as 'the result was significant' without the direction or the test used lose marks.
始终写出单位，如毫秒或百分点，并使答案与数据的精确度一致。在心理学试卷中，像"结果显著"这样未说明方向或所用检验的不完整表述会失分。
Where a question gives a statistical table, read the row for the correct N or degrees of freedom and the column for the correct significance level before comparing the observed value with the critical value.
当题目给出统计表时，在将观测值与临界值比较之前，先找到正确的 N 或自由度所在行，以及正确的显著性水平所在列。
11. Common Mistakes in Psychology Statistics | 心理学统计中的常见错误
The most common errors are conceptual rather than arithmetic. Using the mean on ordinal data, treating a correlation as proof of causation or ignoring outliers are all penalised in both subjects.
最常见的错误是概念性的，而非计算性的。对顺序数据使用平均数、把相关当作因果关系证据、忽略异常值，在两个学科中都会被扣分。
- Confusing the standard deviation with the standard error. | 混淆标准差与标准误。
- Using a one-tailed critical value when a two-tailed hypothesis was tested. | 在双尾假设下使用单尾临界值。
- Reporting p = 0.000 instead of p < 0.001. | 用 p = 0.000 代替 p < 0.001。
- Saying a result "proves" a theory instead of "supports" it. | 说结果"证明"了理论，而非"支持"。
- Rounding intermediate values too early and losing accuracy. | 过早对中间值取整而损失精度。
A second group of errors concerns design. A repeated-measures design removes individual differences but introduces order effects, and these must be acknowledged when interpreting any t-test or Wilcoxon result.
第二类错误涉及设计。重复测量设计消除了个体差异，但引入顺序效应，在解释任何 t 检验或 Wilcoxon 结果时都必须考虑这一点。
12. Transferable Skills Beyond A-Level | 超越 A-Level 的可迁移技能
The statistical skills learned in Edexcel Mathematics transfer directly to psychology coursework, the EPQ, undergraduate research methods modules and eventually to professional work in data analytics, clinical trials and educational assessment.
在 Edexcel 数学中学到的统计技能可直接迁移到心理学课程作业、EPQ、本科研究方法模块，并最终应用于数据分析、临床试验和教育测评等专业工作。
The Edexcel Large Data Set also builds the habit of handling real, messy data with missing values, which mirrors how psychological data sets actually look before cleaning.
Edexcel 的 Large Data Set 还培养处理真实、杂乱且有缺失值数据的习惯，这正反映了心理数据集在清洗之前的真实样貌。
Mastering these ideas is therefore not a detour from psychology but part of its core. A psychologist who understands sampling, distributions and hypothesis testing can read a study critically, design a better one, and report results honestly.
因此，掌握这些内容
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