1.2.4Development of Practical Skills in Physics

Evaluating Results and Drawing Conclusions

5 min read 9 parts Lesson 11 of 29 1 quiz Virtual lab

0%read

Lesson summary

1.2.4 Evaluating Results and Drawing Conclusions — OCR A Level Physics A notes: Judge how reliable an experiment's results are, compare them with accepted values, and write a justified conclusion with sensible improvements.Evaluating an experiment asks: how much can the result be…

Random errors
Systematic errors
specific and linked to a named problem
Worked example 1
Worked example 2
Worked example 3

Section 1 of 9What evaluation means

Evaluating an experiment asks: how much can the result be trusted, and what does it show? A good evaluation uses numbers (percentage uncertainty, percentage difference, scatter about the line) rather than opinions such as “it was quite accurate”. The conclusion must be supported by the data, answer the original aim and mention the limits of what was found.

Section 2 of 9Comparing with a theory or accepted value

If a quantity can be compared with a known value, calculate the percentage difference. Then compare it with the percentage uncertainty in the result:

Inside

If the accepted value lies inside the range (result ± uncertainty), the result is consistent with it.

Systematic effects, if any, are smaller than the random uncertainty.

Outside

If the accepted value lies outside the range, there is evidence of an unaccounted systematic error, or the uncertainties were underestimated.

When two measured values should agree (for example the resistance found by two methods) their ranges should overlap.

Section 3 of 9Random and systematic effects

Random errors cause readings to scatter either side of the true value. They show up as spread in repeats and as points scattered about the best-fit line. They are reduced by repeating and averaging.

Systematic errors shift every reading in the same direction by a similar amount, for example a zero error on a meter, a stopwatch that runs fast, or heat loss in a calorimetry experiment.

Random and systematic effects

They cannot be reduced by averaging; they show up as a wrong intercept or a result that is consistently too high or too low.

They are removed by calibration, correcting for zero error or changing the method.

Do not confuse either with a mistake, such as writing a reading down wrongly, or with “human error”, which is too vague to earn credit.

Section 4 of 9Judging validity of the method

An experiment is valid if it measures what it claims to measure.

Judging validity of the method

Common threats are uncontrolled variables (temperature changing during a resistance experiment), too few data points or too narrow a range,

and a method that cannot detect the effect being investigated.

To test a proportional relationship you need at least six well-spread values, and ideally a range spanning a factor of ten.

Ask these questions

Which measurement has the largest percentage uncertainty, and so limits the result?
Were control variables kept constant?
Are there anomalies?
Does the graph pass through the expected intercept?
Is there any trend in the scatter, which suggests a systematic issue, rather than random fluctuation?

Section 5 of 9Writing the conclusion and improvements

A strong conclusion has three parts(1) the result with uncertainty (“g = 9.6 ± 0.3 m s⁻²”), (2) its comparison with theory (“consistent with 9.8 m s⁻²”), and (3) a statement of confidence based on the evidence.

1Specific and linked to a named problem

Improvements must be specific and linked to a named problem: not “use better equipment”

but “time 20 oscillations rather than 1 to cut the reaction-time uncertainty from 6% to 0.3%”.

These techniques are developed further in Methods to Increase Accuracy (1.2.11).

Section 6 of 9A worked checklist for evaluation answers

In extended answers, structure the evaluation in four short paragraphs.

State the result and its uncertainty in numbers.

Compare with the accepted value or with theory, using percentage difference against percentage uncertainty.

Identify the largest source of uncertainty by finding the biggest percentage contribution, and say whether random or systematic effects dominate,

giving the evidence such as scatter about the line or a non-zero intercept.

Finally suggest two improvements, each tied to a named cause and each with an estimate of how much it would improve the result.

This format is quick to write and covers the points an examiner looks for.

If time is short, always include at least one number; qualitative comments alone rarely gain full credit.

Remember also that a conclusion should be proportionate to the evidence.

A worked checklist for evaluation answers

Six points that lie close to a straight line over one decade support a linear law over that range only; they say nothing about what happens at much larger or smaller values.

If a relationship was found for one material, one temperature or one apparatus, say so, and avoid generalising without further experiments.

Where two competing explanations remain, describe the extra measurement that would distinguish them.

Section 7 of 9Worked examples

A student finds the specific heat capacity of water as 4050 J kg⁻¹ K⁻¹ with a percentage uncertainty of 3%. The accepted value is 4180 J kg⁻¹ K⁻¹. Is the result consistent?
Two methods give a resistance: 47 ± 2 Ω (ohmmeter) and 52 ± 3 Ω (graph). Do they agree?
A time of 1.0 s is measured for a trolley run with a hand-held stopwatch of reaction-time uncertainty 0.2 s. Calculate the percentage uncertainty and suggest an improvement.

Section 8 of 9Key terms

Term	Meaning
Valid	An experiment that measures what it claims to measure with variables controlled.
Systematic error	An error that shifts all readings by a consistent amount in one direction.
Random error	Unpredictable scatter in repeated readings.
Consistent	Two values whose uncertainty ranges overlap.
Reliable	Giving similar results when the measurement is repeated.
Calibration	Checking an instrument against a known standard to find and correct errors.

Section 9 of 9Exam technique

Writing “human error” or “the equipment was not accurate” without naming the specific problem.

Claiming agreement with theory without comparing the difference to the uncertainty.

Suggesting improvements such as “take more care” that do not change the method.

Believing averaging removes systematic errors.

Drawing a conclusion that goes beyond the range of data collected.

Try it in the virtual labCircular motion: centripetal force \n

Section 1 of 9What evaluation means

Section 2 of 9Comparing with a theory or accepted value

Section 3 of 9Random and systematic effects

Section 4 of 9Judging validity of the method

Section 5 of 9Writing the conclusion and improvements

Section 6 of 9A worked checklist for evaluation answers

Section 7 of 9Worked examples

A student finds the specific heat capacity of water as 4050 J kg−1 K−1 with a percentage uncertainty of 3%. The accepted value is 4180 J kg−1 K−1. Is the result consistent?

Two methods give a resistance: 47 ± 2 Ω (ohmmeter) and 52 ± 3 Ω (graph). Do they agree?

A time of 1.0 s is measured for a trolley run with a hand-held stopwatch of reaction-time uncertainty 0.2 s. Calculate the percentage uncertainty and suggest an improvement.

Section 8 of 9Key terms

Section 9 of 9Exam technique

Marked practiceQuizzes for this lesson

A student finds the specific heat capacity of water as 4050 J kg⁻¹ K⁻¹ with a percentage uncertainty of 3%. The accepted value is 4180 J kg⁻¹ K⁻¹. Is the result consistent?