📚 Mastering Experimental Investigation in A-Level Physics | 掌握A-Level物理实验探究
Experimental investigation lies at the heart of A‑Level Physics. Whether you are testing a theoretical relationship or uncovering new patterns in data, the skills you develop in planning, measuring, analyzing and evaluating are essential not only for your practical endorsement but also for the written papers. This article, inspired by the International A‑Level Science Fundamental Skills Booklet for Physics, will guide you through the key competences every A‑Level physicist must master. We will explore how to design a rigorous experiment, record data with meaningful uncertainties, present results graphically, draw reliable conclusions and critically evaluate your work.
实验探究是A‑Level物理的核心。无论你是在验证理论关系还是在数据中发现新的规律，你在计划、测量、分析和评估方面培养的技能不仅对实践考核至关重要，对笔试也同样重要。本文受国际A‑Level科学基础技能手册（物理分册）的启发，将带你掌握每一位A‑Level物理学习者都必须具备的关键能力。我们将探讨如何设计严谨的实验、记录带有意义不确定度的数据、用图形展示结果、得出可靠的结论并对你的工作进行批判性评估。
1. Planning an Experiment | 规划实验
A successful investigation starts with a clear plan. You must identify the independent variable (the one you change), the dependent variable (the one you measure) and the control variables (those you keep constant). Formulate a focused research question that links these variables, for example: “How does the length of a pendulum affect its period?” A well‑stated hypothesis can guide your selection of apparatus and method.
成功的探究始于清晰的计划。你必须明确自变量（你改变的变量）、因变量（你测量的变量）和控制变量（你保持不变的变量）。提出一个将这些变量联系起来的聚焦研究问题，例如：“单摆的长度如何影响其周期？”一个表述清晰的假设能指导你选择仪器和方法。
Your plan should include a risk assessment, a list of equipment with estimated uncertainties, and a step‑by‑step procedure. Always consider how you will obtain a suitable range of data – at least six readings spread evenly across the range – and how you will repeat measurements to check reliability. Think about whether the experiment is a fair test and how you will handle potential confounding factors.
你的计划应包括风险评估、带有预估不确定度的设备清单和分步操作流程。始终考虑如何获取合适的数据范围——至少六个均匀分布在范围内的读数——以及如何重复测量以检查可靠性。思考实验是否是一次公平测试，以及你将如何处理潜在的混杂因素。
2. Measurements and Uncertainties | 测量与不确定度
on of the instrument. For example, a metre ruler with millimetre markings gives an uncertainty of ±0.5 half the range (max − min) ÷ number of values, or more rigorously the standard deviation.
每一次测量都带有不确定度。单次读数的绝对不确定度通常取仪器最小刻度的一半。例如，一把有毫米刻度的米尺，其不确定度为±0.5 mm。当你进行重复读数时，不确定度可以估算为极差的一半（最大值−最小值）÷测定次数，或者更严格地使用标准偏差。
| Instrument | 仪器 | Typical resolution | 典型分辨率 | Common uncertainty | 常见不确定度 |
|---|---|---|
| Metre ruler | 米尺 | 1 mm | ±0.5 mm |
| Vernier caliper | 游标卡尺 | 0.01 mm | ±0.01 mm |
| Micrometer screw gauge | 螺旋测微器 | 0.01 mm | ±0.005 mm |
| Digital stopwatch | 数字秒表 | 0.01 s | ±0.2 s (human reaction time) | ±0.2 s (人的反应时间) |
| Analogue ammeter | 指针式电流表 | 0.1 A | ±0.05 A |
ion. Percentage uncertainty is calculated as (absolute uncertainty ÷ measured value) × 100 % rly useful when comparing the significance of uncertainties across different measurements.
始终按照仪器分辨率的位数记录原始数据。百分数不确定度计算为（绝对不确定度÷测量值）×100 %。在比较不同测量中不确定度的影响时，这尤其有用。
3. Types of Errors | 误差的类型
Errors in physics experiments are not ‘mistakes’ – they are the inevitable variation in measurements. Understanding the difference between random and systematic errors is critical. Random errors cause readings to be scattered about the true value and can be reduced by taking many repeats and averaging. Systematic errors, such as a zero error or a calibration issue, shift all readings consistently in one direction and cannot be eliminated by repetition.
物理实验中的误差并不是“错误”——它们是测量中不可避免的差异。理解随机误差和系统误差的区别至关重要。随机误差导致读数在真值周围散布，可以通过多次重复取平均值来减小。系统误差，如零点误差或校准问题，会使所有读数一致地向一个方向偏移，且无法通过重复来消除。
When you design an experiment, identify potential sources of systematic error and try to eliminate or compensate for them (e.g. by subtracting a zero offset). In your evaluation, discuss how these errors might have affected the gradient or intercept of your graph. Remember: precision relates to the spread of repeat readings, while accuracy describes how close the mean is to the accepted value.
当你设计实验时，要识别潜在的系统误差来源，并尝试消除或补偿它们（例如减去零点偏移）。在评估中，讨论这些误差如何影响图形的斜率或截距。记住：精密度与重复读数的离散程度有关，而准确度描述平均值接近公认值的程度。
4. Recording Data Effectively | 有效记录数据
Good data tables are the foundation of reliable analysis. Every table should have a descriptive title, and each column must be headed with the quantity and its unit, separated by a slash (e.g. Length, l / cm). The first column usually contains the independent variable, and subsequent columns contain the dependent variable(s). Include a column for calculated quantities if needed.
好的数据表是可靠分析的基础。每个表格都应有描述性的标题，每一列必须标有物理量及其单位，用斜线分隔（例如 长度, l / cm）。第一列通常为自变量，后续列为因变量。必要时可包含计算量栏。
n adjacent columns and show a column for the mean and one for the uncertainty (e.g. ± half‑ Never change raw data; if a reading appears anomalous, mark it clearly but still record it – you may later justify its exclusion with a reasoned argument.
将所有原始数据直接记录在表格中，不要写在草稿纸上。如果你进行重复测量，将它们列在相邻列中，并显示平均值列和不确定度列（例如 ±半极差）。永远不要更改原始数据；如果一个读数看起来异常，清楚地标记它但仍要记录——你可以在后续用有根据的论证说明其排除理由。
5. Graphical Analysis: Best‑Fit Lines | 图形分析：最佳拟合线
Plotting data on a graph allows you to see trends, identify anomalies, and determine constants. Use graph paper or software that gives you full control over axis scaling. The independent variable goes on the x‑axis, the dependent on the y‑axis. Label axes with the quantity and unit, e.g. ‘T² / s²’. Choose scales that make the plotted points occupy at least half the page in both directions; avoid awkward scales like multiples of 3.
在图形上绘制数据可以让你观察趋势、识别异常值并确定常数。使用坐标纸或可以完全控制坐标轴刻度的软件。自变量放在x轴，因变量放在y轴。用物理量和单位标记坐标轴，例如“T² / s²”。选择能使绘图点至少在两个方向上都占据半页纸的刻度；避免像3的倍数这样不便读取的刻度。
Draw a best‑fit line that passes through as many error bars as possible and has a roughly equal number of points on either side. Never force the line through the origin unless there is a theoretical justification. Use a transparent ruler to help position the line. The gradient of this line can be calculated by taking two widely spaced points on the line (not data points) and using Δy/Δx.
绘制一条通过尽可能多误差棒的直线，并使直线两侧的数据点数目大致相等。除非有理论依据，否则绝不要强制直线通过原点。使用透明直尺帮助定位直线。该直线的斜率可以通过在直线上（而非数据点）取两个相距较远的点，利用Δy/Δx计算得到。
6. Linearizing Non‑Linear Relationships | 线性化非线性关系
Many physics relationships are non‑linear, e.g. the period T of a simple pendulum varies with length l according to T = 2π√l/g /g) l, so plotting T² against l yields a straight line through the origin with gradient 4π²/g.
许多物理关系是非线性的，例如单摆的周期T随长度l的变化关系为T = 2π√l/g。为了验证这样的关系，我们对数据进行变换，使图形成为一条直线。在此例中，两边平方得到T² = (4π²/g) l，因此将T²对l作图会得到一条通过原点的直线，斜率为4π²/g。
common linearizations include: taking natural logs of both sides for exponential or power‑law relationships (ln y = ln k + n ln x), or plotting y² against x for a square‑ lationship. Always label the transformed axes with the appropriate quantity and unit, e.g. ‘ln(T / s)’ or ‘ adient and intercept with the theoretical expression, you can determine unknown constants.
其他常见的线性化方法包括：对于指数或幂律关系，对两边取自然对数（ln y = ln k + n ln x）；或者对于平方根关系，绘制y²对x的图形。始终用相应的物理量和单位标记变换后的坐标轴，例如“ln(T / s)”或“1/d² / cm⁻²”。通过将斜率和截距与理论表达式进行比较，你可以确定未知常数。
7. Calculating Uncertainties in Gradients and Intercepts | 计算斜率与截距的不确定度
A best‑fit line alone does not tell you how reliable the gradient is. To estimate the uncertainty, you can draw a steepest and a shallowest possible line that still fit the data reasonably (often called ‘worst‑fit’ lines). The absolute uncertainty in the gradient is then given by (gradient of steepest line − gradient of shallowest line) ÷ 2.
仅凭一条最佳拟合线无法告诉你斜率的可靠性如何。为了估算不确定度，你可以绘制出仍能合理拟合数据的、可能的最陡和最浅直线（通常称为“最差拟合线”）。斜率的绝对不确定度则由（最陡线斜率−最浅线斜率）÷2给出。
Similarly, the uncertainty in the y‑ n the form: value ± uncertainty, with the appropriate unit. For example, g = 9.78 ± 0.23 m rcentage uncertainty in the gradient can be used to discuss the quality of the experiment.
同样，y轴截距的不确定度可以从这两条极端线的截距之差求得。始终以“值±不确定度”的形式表达你的最终结果，并附上合适的单位。例如，g = 9.78 ± 0.23 m s⁻²。斜率的百分数不确定度可用于讨论实验的质量。
8. Evaluating Results and Drawing Conclusions | 评估结果与得出结论
Your conclusion must be directly linked to the original research question. State whether the data support the hypothesis and refer to the graphical or numerical evidence you obtained. Compare your measured value with the accepted value (if known) by calculating the percentage difference. If the ranges overlap within uncertainties, your result is consistent with the accepted value.
你的结论必须与最初的研究问题直接相关。陈述数据是否支持假设，并引用你所获得的图形或数值证据。通过计算百分差，将你的测量值与公认值（若已知）进行比较。如果在不确定度范围内两者重叠，则你的结果与公认值一致。
An excellent evaluation goes beyond a simple ‘it worked’. Discuss sources of systematic and random error in detail, suggest realistic improvements to the method or apparatus, and comment on the reliability of your data. Could you have taken more repeats? Would a different sensor reduce delays? Always support your evaluation with specific examples from your data.
优秀的评估不仅仅是简单的“实验成功了”。要详细讨论系统误差和随机误差的来源，对方法或仪器提出切实可行的改进建议，并评价数据的可靠性。你是否可以做更多重复测量？采用不同的传感器能否减少延迟？始终用数据中的具体例子来支撑你的评估。
9. Common Pitfalls in Practical Work | 实验工作中的常见陷阱
Many students lose marks through avoidable mistakes. One pitfall is using a scale that is too small, which makes it impossible to read intercepts or see scatter. Another is forgetting to include units in table headings or axis labels. When repeating measurements, some students discard ‘bad’ readings without justification – this is not scientifically acceptable. Instead, record all data and apply an agreed rule for identifying outliers, such as those more than 2 standard deviations from the mean.
许多学生因可避免的错误而失分。一个陷阱是坐标轴刻度太小，导致无法读取截距或观察数据的离散情况。另一个是忘记在表头或坐标轴标签中写上单位。在进行重复测量时，有些学生毫无理由地丢弃“不好”的读数——这在科学上是不可接受的。应当记录所有数据，并采用一致认可的规则来识别异常值，例如超出平均值±2个标准偏差的点。
Another common error is misusing significant figures. A calculated value should not be quoted to more significant figures than the least precise measurement in the dataset. Respect the precision of your instruments. Finally, when plotting a graph, do not join the dots; draw a smooth straight line or curve of best fit. Error bars are often omitted – always add them to at least the most important points if calibration data permits.
另一个常见错误是误用有效数字。计算值不应比数据集中精度最低的测量值具有更多的有效数字。要尊重仪器的精密度。最后，绘图时不要逐点连线，而应画一条平滑的最佳拟合直线或曲线。误差棒经常被忽略——如果通过校准数据允许，至少要为最重要的数据点添加误差棒。
10. Worked Example: Investigating a Resistor–Capacitor Circuit | 实例分析：探究电阻–电容电路
Let us apply these skills to a common A‑ tor through a resistor. The voltage V across a capacitor decreases exponentially: V = V₀ e−t/RC. To test this, we measure V at different times t after closing the switch. The independent variable is t, the dependent variable is V. Control variables include the initial voltage V₀ and the total resistance R.
让我们将这些技能应用到一个常见的A‑Level实验中：电容器通过电阻放电。电容器两端的电压V呈指数衰减：V = V₀ e−t/RC。为了验证这一点，我们在合上开关后测量不同时刻t的V值。自变量是t，因变量是V。控制变量包括初始电压V₀和总电阻R。
ould yield a straight line with gradient −1/RC and intercept ln V₀. After drawing the best‑fit line and two worst‑ es, we find gradient = −0.0482 ± 0.0015 s⁻¹. Using R = 10.0 ± 0.1 kΩ, we calculate C = 1/(|gradient| × R) = 2.07 ± 0.07 µF, which is consistent with the manufacturer’s value of 2.2 µF within uncertainties.
原始数据记录在表格中，列标题为t/s和V/V。为了线性化，取自然对数：ln V = ln V₀ − (1/RC) t。将ln(V)对t作图应得出一条直线，斜率为−1/RC，截距为ln V₀。绘制最佳拟合线和两条最差拟合线后，求得斜率 = −0.0482 ± 0.0015 s⁻¹。利用R = 10.0 ± 0.1 kΩ，计算出C = 1/(|斜率| × R) = 2.07 ± 0.07 µF，这在不确定度范围内与制造商给出的2.2 µF一致。
01 V) is small relative to the scatter of points, suggesting another source of random error, possibly due to imperfect switching or dielectric absorption. Systematic error could arise from the input impedance of the voltmeter itself. To improve, we could use a data logger that captures V and t simultaneously with higher time resolution, and shield the circuit from stray capacitance.
在评估中，我们注意到电压表的读数不确定度（±0.01 V）相对于点的离散程度较小，这表明存在其他随机误差来源，可能源于不完美的开关动作或介质吸收。系统误差可能来自电压表本身的输入阻抗。为了改进，我们可以使用数据采集器以更高的时间分辨率同时捕获V和t，并对电路进行屏蔽以防杂散电容的影响。
11. Using Error Bars Correctly | 正确使用误差棒
bsolute uncertainty in the dependent variable, while horizontal bars indicate uncertainties in the independent variable. When both are small, they may be omitted, but you should still state the uncertainty. The size of error bars helps you judge whether a straight line or curve is a good fit, because you expect about 68 % of points to fall within their error bar from the best‑fit line if uncertainties are correctly estimated and random errors dominate.
误差棒是数据点不确定度的视觉呈现。垂直误差棒表示因变量的±绝对不确定度，水平误差棒表示自变量的不确定度。当两者都较小时，可以省略，但仍应说明不确定度。误差棒的大小有助于判断直线或曲线是否拟合良好，因为如果不确定度估算正确且随机误差占主导，你预期大约68%的点会落在从最佳拟合线起算的一个误差棒范围内。
Draw error bars accurately using a sharp pencil or software. If your error bars are very large relative to the scale, it suggests either poor precision in measurement or that important variables were not controlled. In such a case, discuss in your evaluation how the experiment could be redesigned to reduce these uncertainties.
使用削尖的铅笔或软件准确地绘制误差棒。如果你的误差棒相对刻度非常大，这表明要么测量精度差，要么重要的变量未得到控制。在此情况下，你应在评估中讨论如何重新设计实验以减小这些不确定度。
12. Mastering the Written Practical Paper | 掌握实验笔试题
A‑Level physics examinations often contain questions that ask you to describe how you would carry out a practical investigation, or to analyze a given set of data including uncertainties. When answering such questions, be systematic: specify the apparatus you would use (giving typical instrument uncertainties), outline the method step‑by‑step, specify what measurements you would take and how you would control variables, explain how you would process the data (including any linearization), and indicate how you would find a constant and its uncertainty.
A‑Level物理考试中经常包含一些题目，要求你描述将如何进行一项实验探究，或分析一组包含不确定度的给定数据。在回答这类问题时，要有条理：明确你将使用的仪器（给出典型的仪器不确定度），逐步概述方法，说明你将进行哪些测量以及如何控制变量，解释你将如何处理数据（包括任何线性化步骤），并说明你将如何求得常数及其不确定度。
Practice past paper questions until the structure becomes second nature. Pay particular attention to phrasing like “describe a method to determine …” or “comment on the reliability of the data”. Use the fundamental skills from this booklet as a mental checklist: did you include a risk assessment? Did you discuss how to reduce systematic errors? Did you state the final result with the correct number of significant figures and an uncertainty? Mastering these details will set your answers apart.
反复练习过往试卷题目，直到这种结构成为你的第二天性。特别注意诸如“描述测定……的方法”或“评论数据的可靠性”之类的措辞。将本文中的基础技能用作大脑中的检查清单：你是否包括了风险评估？你是否讨论了如何减少系统误差？你是否以正确的有效数字和不确定度陈述了最终结果？掌握这些细节将使你的答案与众不同。
Published by TutorHao | Physics Revision Series | aleveler.com
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