📚 Practical and Analytical Skills | AQA A Level 物理实践与分析技能
Practical and analytical skills are assessed throughout AQA A Level Physics, especially in Paper 3 and in the practical endorsement. You need to understand how to measure quantities, estimate uncertainties, process data, plot graphs and evaluate the quality of an experiment. This article brings together the key ideas you must be confident with, from instrument resolution to combining uncertainties and drawing valid conclusions.
在 AQA A Level 物理中，实践与分析技能贯穿整个课程，尤其是在 Paper 3 和实践认证中考查。你需要理解如何测量物理量、估计不确定度、处理数据、绘制图像以及评价实验质量。本文汇总了你必须熟练掌握的关键知识点，从仪器分辨率到组合不确定度，再到得出可靠结论。
1. Measurement and Resolution | 测量与仪器分辨率
Every measurement is limited by the resolution of the instrument. Resolution is the smallest change in the quantity being measured that the instrument can show clearly. It is not the same as accuracy, because a high-resolution instrument can still give inaccurate readings if it is poorly calibrated.
每个测量值都受仪器分辨率的限制。分辨率是仪器能够清晰显示的被测量量的最小变化。分辨率与准确度不同，因为一台高分辨率仪器如果校准不良，仍然可能给出不准确的读数。
states otherwise, so a digital ammeter reading 0.42 A has a reading uncertainty of ±0.01 A.
对于模拟刻度，读数不确定度通常取最小分度值的一半。例如，分度值为 1 °C 的温度计读数不确定度为 ±0.5 °C。对于数字仪表，不确定度通常为最后一位显示数字的 ±1，除非制造商另有说明，因此读数为 0.42 A 的数字电流表读数不确定度为 ±0.01 A。
analogue reading uncertainty = ½ × smallest scale division
digital reading uncertainty = ± resolution of last digit
Parallax error occurs when your eye is not perpendicular to the scale, and zero error occurs when the instrument does not read zero at the true zero position. Both can be reduced or corrected with careful technique.
当你的视线不与刻度垂直时会产生视差误差；当仪器在真实零点位置不指零时会产生零误差。这两种误差都可以通过仔细操作来减小或修正。
2. Systematic and Random Errors | 系统误差与随机误差
A systematic error causes readings to differ from the true value by the same amount or in the same direction on every repeat. It is not reduced by taking repeats. Common causes include zero error, calibration drift, or a consistent experimental flaw such as measuring length from the wrong end of a rule.
系统误差使每次重复测量的读数与真实值相差相同的量或朝相同方向偏离。重复测量不能减小系统误差。常见原因包括零误差、校准漂移，或持续的实验缺陷，例如从直尺错误的一端开始测量长度。
A random error causes readings to scatter unpredictably around the true value. Random errors can be caused by reaction time, fluctuations in the environment, or judgement between scale divisions. Taking many repeats and using the mean reduces the effect of random errors, because positive and negative deviations partially cancel.
随机误差使读数在真实值附近不可预测地散布。随机误差可能由反应时间、环境波动或刻度分度之间的判断引起。多次重复并取平均值可以减小随机误差的影响，因为正负偏差会部分抵消。
It is important to identify which type of error dominates in an experiment. If repeats are very close but the mean is far from the accepted value, a systematic error is likely. If repeats are widely spread, random error is significant.
识别实验中哪种误差占主导很重要。如果重复测量非常接近，但平均值远离公认值，则可能存在系统误差。如果重复测量分散程度很大，则随机误差显著。
3. Accuracy, Precision, Repeatability and Reproducibility | 准确度、精密度、重复性与再现性
Accuracy describes how close a measured value is to the true or accepted value. Precision describes how close repeated measurements are to each other. A set of readings can be precise but inaccurate if a systematic error shifts all values away from the true value.
准确度描述测量值接近真实值或公认值的程度。精密度描述重复测量值彼此接近的程度。如果系统误差使所有值偏离真实值，一组读数可能精密但不准确。
Repeatability means the same person, using the same instrument and method under the same conditions, obtains close results on repeated trials. Reproducibility means different people, using different instruments or methods, can obtain close results for the same measurement.
重复性是指同一个人使用相同的仪器和方法在相同条件下重复试验获得接近的结果。再现性是指不同人使用不同仪器或方法对同一测量也能获得接近的结果。
.02 s and the accepted value is 1.50 s, the result is accurate within uncertainty and the small spread shows good precision.
在实践工作中，你应该同时评价准确度和精密度。例如，如果平均周期为 1.52 ± 0.02 s，而公认值为 1.50 s，则结果在不确定度范围内是准确的，且小范围散布表明精密度良好。
4. Absolute, Fractional and Percentage Uncertainties | 绝对不确定度、相对不确定度与百分比不确定度
h as ΔT = ±0.2 s. It represents the interval within which the true value is likely to lie.
绝对不确定度写作 ± 一个与测量值单位相同的数值，例如 ΔT = ±0.2 s。它表示真实值可能落入的区间。
The fractional uncertainty is the absolute uncertainty divided by the measured value. The percentage uncertainty is the fractional uncertainty multiplied by 100%. These relative measures are useful when comparing uncertainties of different quantities.
相对不确定度是绝对不确定度除以测量值。百分比不确定度是相对不确定度乘以 100%。这些相对量度在比较不同物理量的不确定度时很有用。
fractional uncertainty = Δx / x
percentage uncertainty = (Δx / x) × 100%
if a length is measured as 25.0 ± 0.5 cm, the percentage uncertainty is (0.5 / 25.0) × 100% = 2.0% ties from several contributing instruments must be combined, as shown in the next section.
例如，若测得长度为 25.0 ± 0.5 cm，则百分比不确定度为 (0.5 / 25.0) × 100% = 2.0%。来自多个仪器的读数不确定度必须进行组合，如下一节所示。
5. Combining Uncertainties | 组合不确定度
When quantities are added or subtracted, the absolute uncertainties add. When quantities are multiplied or divided, the percentage uncertainties add. When a quantity is raised to a power, the percentage uncertainty is multiplied by that power.
当物理量相加或相减时，绝对不确定度相加。当物理量相乘或相除时，百分比不确定度相加。当物理量被乘方时，百分比不确定度乘以该指数。
if R = A + B or R = A − B, then ΔR = ΔA + ΔB
if R = A × B or R = A / B, then %ΔR = %ΔA + %ΔB
if R = Aⁿ, then %ΔR = n × %ΔA
R = V / I, where V = 5.0 ± 0.1 V and I = 2.00 ± 0.05 A. The percentage uncertainties are 2% for voltage and 2.5% for current, so the percentage uncertainty in resistance is 4.5% s 2.5 Ω, giving an absolute uncertainty of about 0.11 Ω, which should be quoted as ±0.1 Ω.
考虑由 R = V / I 计算电阻，其中 V = 5.0 ± 0.1 V，I = 2.00 ± 0.05 A。电压的百分比不确定度为 2%，电流为 2.5%，因此电阻的百分比不确定度为 4.5%。计算得到电阻为 2.5 Ω，绝对不确定度约为 0.11 Ω，应表示为 ±0.1 Ω。
| Operation | 运算 | Uncertainty rule | 不确定度规则 |
| Addition / subtraction | Add absolute uncertainties |
| Multiplication / division | Add percentage uncertainties |
| Power Aⁿ | Multiply percentage uncertainty by n |
6. Significant Figures and Rounding | 有效数字与修约
s. It is not meaningful to write 1.523 ± 0.1 s because the extra digits are not justified.
最终答案的精度应与不确定度一致。如果不确定度为 ±0.1 s，则平均值应保留一位小数，例如 1.5 ± 0.1 s。写成 1.523 ± 0.1 s 没有意义，因为多余的位数没有依据。
Uncertainties are usually quoted to one significant figure, or sometimes two significant figures when the first digit is 1 or 2. Percent uncertainties should also be rounded sensibly, typically to one or two significant figures.
不确定度通常保留一位有效数字，有时当首位数字是 1 或 2 时保留两位有效数字。百分比不确定度也应根据情况合理修约，通常保留一到两位有效数字。
For intermediate calculations, keep extra significant figures to avoid rounding errors, then round only the final answer. This is especially important when squaring values or using logarithms.
在中间计算过程中，应保留额外的有效数字以避免修约误差，只在最后答案时修约。这一点在平方或使用对数时尤其重要。
7. Graphical Analysis: Error Bars and Best and Worst Fit | 图像分析：误差棒与最佳、最差拟合线
Plotting a graph is a core practical skill. The independent variable goes on the x-axis and the dependent variable on the y-axis. You should use sensible scales, label axes with quantity and unit, and plot points with crosses or small circles.
绘制图像是一项核心实践技能。自变量放在 x 轴，因变量放在 y 轴。应使用合理的标度，标注坐标轴名称与单位，并用十字或小圆圈描点。
Error bars show the uncertainty in each plotted value. The vertical error bar extends Δy above and below each point, and the horizontal error bar extends Δx left and right. If the uncertainty is too small to show clearly, state that error bars are within the plotted points.
误差棒表示每个描点值的不确定度。垂直误差棒在每个点上下延伸 Δy，水平误差棒在左右延伸 Δx。如果不确定度太小无法清楚显示，应说明误差棒包含在描点内。
A best fit line passes through the plotted points as closely as possible, with roughly equal numbers of points above and below the line. A worst fit line may be used to estimate the uncertainty in the gradient: it is the steepest or shallowest line that still passes through all the error bars, or through the most extreme reasonable points.
最佳拟合线尽可能靠近所有描点，使线上方和下方的点数大致相等。最差拟合线可用于估计斜率的不确定度：它是仍能通过所有误差棒或最极端合理点的最陡或最浅的线。
gradient = Δy / Δx
Δgradient = (gradient_max − gradient_min) / 2
The y-intercept should also be interpreted physically only if the model predicts a non-zero intercept. Otherwise a non-zero intercept may suggest a systematic error.
只有当模型预测截距不为零时，y 截距才具有物理意义。否则非零截距可能暗示存在系统误差。
8. Linearising Data and Logarithmic Plots | 数据线性化与对数图像
Many physical relationships are non-linear, such as T = 2π√L/g graph, rearrange the equation into the form y = mx + c and plot the appropriate variables.
许多物理关系是非线性的，例如单摆的周期公式 T = 2π√L/g。为了从直线图像中求未知量，需要将方程整理为 y = mx + c 形式，并绘制相应变量。
aight line through the origin, with gradient 4π²/g. From the gradient, g = 4π² / gradient.
对于单摆，两边平方得到 T² = (4π²/g) L。以 T² 为 y 轴、L 为 x 轴作图，得到一条过原点的直线，斜率为 4π²/g。由斜率可求得 g = 4π² / 斜率。
T² = (4π²/g) L
g = 4π² / gradient
ln k. If y = k e⁻λx, then ln y = ln k − λx, so a graph of ln y against x has gradient −λ.
对于指数或幂律关系，对数图像很有用。若 y = k xⁿ，则 ln y = n ln x + ln k。以 ln y 对 ln x 作图得到直线，斜率为 n，截距为 ln k。若 y = k e⁻λx，则 ln y = ln k − λx，因此以 ln y 对 x 作图的斜率为 −λ。
9. Anomalies and Improving Experiments | 异常值与实验改进
An anomaly is a reading or plotted point that does not fit the overall trend. It may be caused by a mistake in measurement, recording error, or a brief uncontrolled change in conditions. In tables, anomalies should be identified and excluded from the mean, and in graphs they should be labelled and ignored when fitting the line.
异常值是指不符合整体趋势的读数或描点。它可能由测量错误、记录错误或条件的短暂失控变化引起。在表格中，应识别异常值并从平均值中剔除；在图像中，应标注异常值并在拟合直线时忽略它。
To improve an experiment, you should increase the number of repeats, use instruments with higher resolution, control more variables, or change the method to reduce the largest uncertainty. You should also consider calibration of instruments before use.
为了改进实验，应增加重复次数、使用更高分辨率的仪器、控制更多变量，或改变方法以减小最大的不确定度。还应考虑在使用前校准仪器。
The largest percentage uncertainty often comes from the smallest measured quantity, such as a short time or a small temperature change. Focusing on that quantity is usually the most effective improvement.
最大的百分比不确定度通常来自最小的被测量，例如很短的时间或很小的温度变化。针对该量进行改进通常最有效。
10. Drawing Conclusions and Evaluating the Method | 得出结论与评价方法
A conclusion must be supported by the data, and it should state whether the results agree with the accepted value or theoretical prediction within the uncertainties. Use the phrase 'within experimental uncertainty' rather than claiming exact agreement.
结论必须由数据支持，并应说明结果是否在不确定度范围内与公认值或理论预测一致。应使用“在实验不确定度范围内一致”这样的表述，而不是声称完全一致。
percentage difference = (|experimental value − accepted value| / accepted value) × 100%
If the percentage difference is larger than the total percentage uncertainty, systematic error is likely. If it is smaller, the result supports the model or accepted value.
如果百分比差异大于总百分比不确定度，则很可能存在系统误差。如果小于总百分比不确定度，则结果支持该模型或公认值。
Evaluation should identify the main sources of uncertainty, whether they are random or systematic, and suggest realistic improvements. Avoid vague statements like 'be more careful'; instead suggest specific changes such as 'use a light gate to measure time rather than a stopwatch', or 'clamp the rule vertically and read at eye level to reduce parallax'.
评价应指出主要的不确定度来源，说明它们是随机还是系统的，并提出切合实际的改进建议。避免模糊的说法，如“更仔细些”；而要提出具体改变，例如“使用光门代替秒表测时间”，或“竖直固定直尺并在视线水平处读数以减少视差”。
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