Planning, Analysis and Evaluation | 计划、分析与评价

📚 Planning, Analysis and Evaluation | 计划、分析与评价

Planning, analysis and evaluation form the core of the A-Level Biology practical assessment. This chapter explores how to design rigorous experiments, select appropriate statistical tools, and critically evaluate data to draw valid conclusions. Mastery of these skills not only ensures success in Paper 5 but also builds a genuine scientific mindset.

计划、分析与评价构成了A-Level生物实验评估的核心。本章探讨如何设计严谨的实验、选择恰当的统计工具，并批判性地评价数据以得出有效结论。掌握这些技能不仅能确保在Paper 5中取得成功，更能培养真正的科学思维。

1. Variables and Hypotheses | 变量与假设

Every experiment begins by identifying the independent variable (the factor you change) and the dependent variable (the factor you measure). All other variables that could influence the outcome must be controlled or kept constant. A testable hypothesis must then be written as a clear statement predicting the relationship between these variables, often using ‘If… then…’ logic.

每个实验都始于识别自变量（你改变的因素）和因变量（你测量的因素）。所有其他可能影响结果的变量都必须加以控制或保持恒定。之后，必须将可检验的假设写成清晰的陈述，预测这些变量之间的关系，通常采用“如果……那么……”的逻辑。

In A-Level Biology, a null hypothesis is also essential for statistical testing. It states that there is no significant difference or relationship between variables. The alternative hypothesis holds that a significant difference or correlation does exist. For example, ‘If light intensity increases, then the rate of photosynthesis will increase’ is an alternative hypothesis, while the null version would assert that light intensity has no effect.

在A-Level生物中，零假设对统计检验也至关重要。它表明变量之间没有显著差异或关系。备择假设则认为存在显著差异或相关性。例如，“如果光照强度增加，那么光合作用速率将增加”是备择假设，而零假设则会断言光照强度没有影响。

2. Experimental Design and Controls | 实验设计与对照

A robust design uses a control group to provide a baseline for comparison. The control is identical to the experimental group in every way except for the independent variable. This ensures that any observed effect can be attributed solely to the factor being tested. Without appropriate controls, results may be invalid due to confounding variables.

稳健的设计使用对照组提供比较的基线。对照组除自变量外，与实验组在各方面均相同。这确保了任何观察到的效应都可唯一归因于被测试的因素。如果没有适当的对照，结果可能由于混杂变量而无效。

Replication is equally important: each treatment should have at least three replicates (three petri dishes, three test tubes, etc.). Replication allows calculation of means and assessment of variability. When designing the protocol, state precisely the volumes, concentrations, incubation times and equipment used, so that the experiment is reproducible by another scientist.

重复同样重要：每个处理应至少有三个重复（三个培养皿、三支试管等）。重复允许计算平均值并评估变异性。在设计方案时，要精确说明所用的体积、浓度、孵育时间和设备，以便其他科学家能够重复该实验。

3. Data Collection and Recording | 数据收集与记录

Data must be collected systematically and presented in clearly labelled tables. The independent variable goes in the first column, and the dependent variable in subsequent columns, with units in the header only. For example, ‘Temperature / °C’ and ‘Rate of reaction / cm³ min⁻¹’. All raw data should be recorded to the same number of decimal places determined by the precision of the measuring instrument.

数据必须系统地收集，并呈现在清晰标记的表格中。自变量放在第一列，因变量放在后续列中，单位仅放在表头。例如，“温度 / °C”和“反应速率 / cm³ min⁻¹”。所有原始数据都应记录到由测量仪器精度决定的相同小数位数。

If you are measuring something like the time for a colour change, use a stopwatch that reads to 0.01 s. Do not mix decimal places: if the instrument gives readings to 0.1 cm³, record all volumes as 2.0, 3.5, 4.0, not 2, 3.5, 4. Any anomalies should be flagged but not removed unless a clear error in procedure can be identified.

如果你正在测量诸如颜色变化的时间，使用可精确到0.01秒的秒表。不要混合小数位数：如果仪器给出的读数精确到0.1 cm³，则将所有体积记录为2.0、3.5、4.0，而非2、3.5、4。任何异常值应予以标记，但除非能识别出操作中的明显错误，否则不应删除。

4. Presenting Results in Graphs | 用图表呈现结果

Graphs should be drawn with the independent variable on the x-axis and the dependent variable on the y-axis. Use sharp pencil lines, label both axes with units, and choose a scale that occupies at least half the grid. For line graphs, plotted points must be clearly marked (without connecting them with a dot-to-dot line unless asked) and a line or curve of best fit drawn to reveal the trend. Bar charts are used for categorical independent variables.

图表应将自变量置于x轴，因变量置于y轴。使用尖锐的铅笔线条，给两个坐标轴标上单位，并选择至少占据一半网格的比例尺。对于折线图，标出的点必须清晰（除非要求，不要用点对点连线），并画出最佳拟合线或曲线以揭示趋势。柱状图用于分类自变量。

When displaying variation around means, add error bars representing ±1 standard deviation or ±1 standard error of the mean. This instantly shows the overlap between data sets. When asked to estimate a gradient, draw a large triangle on the linear portion of the curve and use ‘rise over run’. Always include the units of the gradient if it represents a biological rate.

在显示平均值周围的变异时，添加代表±1标准差或±1平均值标准误的误差线。这立即可显示数据集之间的重叠。当被要求估算梯度时，在曲线的线性部分画一个大三角形，使用“纵增量/横增量”。如果梯度代表生物速率，请始终包含单位。

5. Descriptive Statistics: Mean, SD and SEM | 描述统计：平均值、标准差与标准误

Descriptive statistics summarise data. The mean is the sum of all values divided by the number of replicates (x̄ = Σx / n). The standard deviation (SD) measures the spread of data around the mean. A small SD indicates that the data points are tightly clustered, suggesting high precision. The formula for SD is:

描述统计用于总结数据。平均值是所有数值之和除以重复次数 (x̄ = Σx / n)。标准差 (SD) 衡量数据围绕平均值的分散程度。小的SD表明数据点紧密聚集，意味着高精度。SD的公式为：

SD = √[ Σ(x – x̄)² / (n – 1) ]

The standard error of the mean (SEM) is SD divided by √n. SEM estimates how close the sample mean is to the true population mean. Larger samples yield smaller SEMs. In A-Level Biology, you are expected to calculate SD and SEM, and to use them to interpret error bar overlap: if error bars do not overlap, the difference is likely significant.

平均值标准误 (SEM) 是SD除以√n。SEM估计样本平均值与真实总体平均值的接近程度。样本量越大，SEM越小。在A-Level生物中，要求你计算SD和SEM，并用它们解释误差线重叠：如果误差线不重叠，差异很可能显著。

6. Choosing the Right Statistical Test | 选择合适的统计检验

The choice of inferential statistical test depends on the type of data and the question being asked. The Student’s t-test compares the means of two sets of normally distributed data. The Chi-squared (X²) test compares observed frequencies with expected frequencies for categorical data. Using the wrong test can lead to erroneous conclusions and will lose marks in evaluation.

推断统计检验的选择取决于数据类型和所问的问题。学生t检验比较两组正态分布数据的平均值。卡方（X²）检验比较分类数据的观察频率与预期频率。使用错误的检验可能导致错误的结论，并在评价中失分。

t-test: suitable for continuous data (e.g. length, mass, heart rate) when comparing two groups.
t检验：适用于比较两组连续数据（如长度、质量、心率）。
Chi-squared test: suitable for frequencies or counts in discrete categories (e.g. phenotype ratios).
卡方检验：适用于离散类别中的频率或计数（如表型比例）。

Always check that the data meet the assumptions: t-test requires normal distribution and roughly equal variances; Chi-squared requires raw counts, not percentages, and no expected value below 5 in more than 20% of cells. If assumptions are violated, consider an alternative test like the Mann-Whitney U-test (non-parametric).

务必检查数据是否满足假设：t检验要求正态分布且方差大致相等；卡方检验要求原始计数，而非百分比，且期望值低于5的单元格不超过20%。如果假设被违反，考虑替代检验，如Mann-Whitney U检验（非参数）。

7. Performing a t-test | 进行t检验

The unpaired (independent) t-test is common in biology: you calculate t using the formula below, then compare it with the critical value from a t-table at p = 0.05 and the appropriate degrees of freedom (df = n₁ + n₂ – 2). If the calculated t value exceeds the critical value, you reject the null hypothesis.

非配对（独立）t检验在生物学中很常见：你使用下面公式计算t，然后将其与p = 0.05及适当自由度 (df = n₁ + n₂ – 2) 下的临界值比较。如果计算的t值超过临界值，则拒绝零假设。

t = (x̄₁ – x̄₂) / √[ (SD₁²/n₁) + (SD₂²/n₂) ]

For example, if you compared the mean germination rate of seeds with and without a hormone treatment, you would calculate t, find the critical value for your sample sizes, and state whether the difference is significant. Always use two-tailed tests unless you have a strong directional hypothesis supported by prior research.

例如，如果你比较了有和没有激素处理的种子的平均发芽率，你将计算t，查找你的样本大小对应的临界值，并说明差异是否显著。除非你有先前研究支持的强方向性假设，否则始终使用双尾检验。

8. Performing a Chi-squared Test | 进行卡方检验

The chi-squared test is invaluable for genetics, ecology and any experiment producing count data. The formula is:

X² = Σ (O – E)² / E

where O is the observed frequency and E is the expected frequency. Expected values are derived from the null hypothesis (e.g. a 3:1 Mendelian ratio). Degrees of freedom are (number of categories – 1) for a goodness-of-fit test, or (rows – 1) × (columns – 1) for a contingency table.

卡方检验在遗传学、生态学及任何产生计数数据的实验中极为宝贵。公式为：X² = Σ (O – E)² / E，其中O为观察频率，E为期望频率。期望值源自零假设（如3:1孟德尔比例）。拟合优度检验的自由度为（类别数 – 1），列联表检验的自由度为（行数 – 1）×（列数 – 1）。

Interpreting X²: if the calculated X² is less than the critical value at p = 0.05 and df, you fail to reject the null hypothesis; any deviation is due to chance. If X² exceeds the critical value, you reject the null hypothesis, meaning the difference is statistically significant and may indicate a biological effect. Always link statistical significance back to the biological context.

解释X²：如果计算的X²小于p = 0.05及相应df下的临界值，则不能拒绝零假设；任何偏差都是偶然的。如果X²超过临界值，则拒绝零假设，意味着差异在统计上显著，可能表明生物学效应。始终将统计显著性联系回生物学背景。

9. Identifying Sources of Error | 识别误差来源

Errors can be systematic (e.g. a poorly calibrated thermometer, consistently reading 0.5 °C too high) or random (e.g. slight variations in reaction time when starting a stopwatch). Systematic errors affect accuracy, while random errors affect precision. In your evaluation, clearly identify potential errors and state their type and effect on results.

误差可以是系统性的（例如，校准不良的温度计始终偏高0.5 °C）或随机的（例如，启动秒表时反应时间的微小变化）。系统误差影响准确度，随机误差影响精密度。在你的评价中，清晰地识别潜在的误差，说明其类型及对结果的影响。

Common systematic errors in biology: using a water bath that is at the wrong temperature, incomplete mixing of enzyme and substrate, light leaking into a dark treatment. Random errors: parallax error when reading a meniscus, fluctuation in light intensity from a lamp due to voltage changes. Distinguish between human error (avoidable) and inherent measurement uncertainty (unavoidable).

生物学中常见的系统误差：使用温度不正确的水浴、酶和底物混合不均匀、光漏入黑暗处理。随机误差：读取弯月面时的视差、因电压变化导致灯光强度的波动。区分人为错误（可避免）和固有的测量不确定性（不可避免）。

10. Analyzing Anomalous Results | 分析异常结果

An anomaly is a data point that does not fit the overall pattern. Do not automatically discard it. Investigate whether a mistake occurred in the procedure (e.g. misreading a syringe, incorrect dilution). If a procedural error is confirmed, you can exclude the result and repeat that replicate. If no cause is found, retain the datum but acknowledge its presence in the evaluation.

异常值是一个不符合总体模式的数据点。不要自动丢弃它。调查程序中是否发生了错误（例如，错误读取注射器、稀释不正确）。如果确认了程序错误，你可以排除该结果并重复该重复。如果未找到原因，保留该数据但在评价中承认其存在。

Anomalous results can sometimes reveal new biology – for instance, a seed that germinated far earlier might indicate a rare genetic variant. However, in the context of A-Level assessment, you are expected to follow the principle that removing a data point requires justification. Where appropriate, calculate the mean with and without the anomaly to assess its impact.

异常结果有时能揭示新的生物学——例如，一粒发芽极早的种子可能指示一个罕见的遗传变异。然而，在A-Level评价的背景下，应遵循移除数据点需有正当理由的原则。在适当时，计算包含和排除异常值的平均值，以评估其影响。

11. Evaluation and Continuous Improvement | 评价与持续改进

Evaluation is not simply listing weaknesses; it demands a critical analysis of the whole experimental process. Discuss how limitations in design, sample size, and control of variables affected the validity and reliability of the conclusions. Suggest specific, realistic improvements, such as using a data logger to reduce human error, increasing the number of replicates to improve statistical power, or standardising the age of organisms used.

评价不仅仅是列出缺点；它要求对整个实验过程进行批判性分析。讨论设计、样本量和变量控制的局限性如何影响结论的有效性和可靠性。提出具体、现实的改进措施，例如使用数据记录器减少人为错误、增加重复次数以提高统计功效，或标准化所用生物的年龄。

Always link suggestions back to the biology. For example, ‘The enzyme activity could have been affected by slight pH changes; next time, a buffer solution at pH 7.0 should be used throughout.’ Also reflect on whether the original hypothesis is supported. Remember that a ‘non-significant’ result is not a failed investigation – it is still valid data. Use this reflective thinking to show evaluative skill.

始终将建议联系回生物学。例如，“酶的活性可能受到轻微pH变化的影响；下次，应全程使用pH 7.0的缓冲溶液。”同时反思原始假设是否得到支持。记住，“不显著”的结果并非失败的调查——它仍然是有效的数据。利用这种反思性思维展示评价技能。

12. Linking Planning, Analysis and Evaluation | 计划、分析与评价的衔接

These three skills are not isolated. A well-planned experiment yields cleaner data, making analysis more straightforward and evaluation more meaningful. For instance, if you fail to randomise the position of pots in a greenhouse, you introduce a confounding variable (light gradient) that will complicate analysis and weaken your evaluation. Anticipating sources of error at the planning stage is a hallmark of excellent practical work.

这三项技能并非孤立的。规划良好的实验产生更干净的数据，使分析更直接，评价更有意义。例如，如果你未能随机化温室中花盆的位置，你就引入了一个混杂变量（光照梯度），这将使分析复杂化并削弱你的评价。在规划阶段预见误差来源是卓越实验工作的标志。

In the write-up, ensure your analysis section includes justification for the statistical test chosen, clear citations of critical values, and a correct interpretation of p-values. Then, in evaluation, examine whether the analysis truly answers the biological question. This circular reflection mirrors genuine scientific practice and will secure top marks in the Planning, Analysis and Evaluation paper.

在书面报告中，确保分析部分包含所选统计检验的理由、清晰引用的临界值以及对p值的正确解释。然后，在评价中，审视分析是否真正回答了生物学问题。这种循环式的反思反映了真实的科学实践，并将在计划、分析与评价试卷中确保获得高分。

Published by TutorHao | Biology Revision Series | aleveler.com

更多咨询请联系16621398022（同微信）

Planning, Analysis and Evaluation | 计划、分析与评价