1.2.5Development of Practical Skills in Physics

Observations and Measurements

5 min read 9 parts Lesson 12 of 29 1 quiz Virtual lab

0%read

Lesson summary

1.2.5 Observations and Measurements — OCR A Level Physics A notes: Choose suitable instruments, use them correctly to take readings with sensible resolution, and record observations without bias.An observation is something noticed, often qualitative (the wire glows red, the pattern spreads out).

Zero error
Meniscus and curves
Fiducial marker
Choosing meter range
Worked example 1
Worked example 2

Section 1 of 9Observing and measuring

An observation is something noticed, often qualitative (the wire glows red, the pattern spreads out). A measurement is quantitative: a number with a unit and an uncertainty. Good practical work records both, but physics conclusions rest on measurements. Observations should be recorded at the time, stating what actually happened rather than what was expected.

Section 2 of 9Choosing an instrument

The right instrument has a resolution (smallest change it can show) that is small compared with the quantity being measured. A useful rule of thumb is that the percentage uncertainty from the instrument should be well below the percentage you hope to achieve overall.

Quantity	Typical instrument	Typical resolution
Length of a few cm to 1 m	Metre rule	1 mm
Diameter of wire or thin sheet	Micrometer screw gauge	0.01 mm
Diameter of a tube, small block	Vernier callipers	0.1 mm or 0.02 mm
Time interval	Stopwatch / light gate and timer	0.01 s / 0.001 s
Mass	Electronic balance	0.1 g or 0.01 g
Current, p.d.	Digital meters	last digit shown

A 1 mm ruler is fine for a 50 cm length (0.2%) but poor for a wire diameter of 0.5 mm, where the micrometer is needed.

Section 3 of 9Reading instruments correctly

Parallax

Parallax view an analogue scale or pointer at 90° so that the pointer and scale coincide. Mirrors behind pointers help.

Zero error

Zero error check the reading with nothing being measured. For a micrometer, closed jaws might read +0.02 mm; subtract this from every reading. A digital balance should be zeroed (tared) with the container on it.

Ratchet

Ratchet use the ratchet on a micrometer so the force applied is consistent and the wire is not squashed.

Meniscus and curves

Meniscus and curves read a liquid level at the bottom of the meniscus at eye level.

Fiducial marker

Fiducial marker when timing oscillations, start and stop as the object passes a fixed mark at the centre of the swing, where its speed is greatest and the timing error smallest.

Choosing meter range

Choosing meter range use the lowest range that does not overload so the reading has as many significant figures as possible.

Section 4 of 9Uncertainty in a single reading

Even a careful single reading has an uncertainty.

Uncertainty in a single reading

A sensible approach is: for a digital instrument use ± one in the last digit (unless a manufacturer's value is given); for an analogue scale use ± half a division, but increase this when you must judge two edges, as when measuring a length with a rule where both ends are read.

Timing with a hand stopwatch is limited by reaction time, about 0.2 s, far larger than the 0.01 s resolution, so use the larger value.

Measuring a quantity that is the difference of two readings, such as a length found as x₂ − x₁, doubles the absolute uncertainty (see Analysing Quantitative Data, 1.2.2).

Section 5 of 9Good measurement habits

Record every reading immediately with its unit.

Take enough readings: at least six different values of the independent variable, over as wide a range as is safe.

Repeat each reading at least twice, and more if the scatter is large.

Keep control variables fixed (for example the same starting position, the same temperature).

Avoid measuring small quantities directly where a larger one can be measured and divided, for example the thickness of 50 sheets of paper rather than one, which leads on to Methods to Increase Accuracy (1.2.11).

Section 6 of 9Planning a set of measurements

Before taking readings, decide the range and spacing.

For a relationship such as T against l, choose lengths from 0.20 m to 1.00 m at equal steps of 0.10 m or 0.20 m, giving six points, so that a graph can be drawn and curvature detected.

Do a trial run first to check that the range is safe and the meter will not overload.

Note room conditions if relevant, such as temperature in a resistance experiment.

A short trial often reveals practical problems, like a clamp that slips, before you have spent time collecting poor data.

Section 7 of 9Worked examples

A micrometer reads +0.03 mm when the jaws are closed. A wire reads 0.89 mm. Find the true diameter and the percentage uncertainty if the reading uncertainty is ±0.01 mm.
The thickness of a stack of 40 identical sheets of card is 6.8 mm, measured with a rule, with an uncertainty of ±1 mm for the stack. Find the thickness of one sheet and its uncertainty.
A pendulum bob's 25 swings take 41.5 s measured with a hand stopwatch (reaction uncertainty 0.2 s). Find the period, its percentage uncertainty and its absolute uncertainty.

Section 8 of 9Key terms

Term	Meaning
Resolution	The smallest change in a quantity that an instrument can detect or display.
Zero error	A non-zero reading when the true value is zero; it is subtracted from all readings.
Parallax error	A misreading caused by viewing a scale from an angle.
Fiducial marker	A fixed reference mark used to judge the start and end of a timing.
Qualitative	Descriptive, not numerical.
Quantitative	Expressed as a number with a unit.

Section 9 of 9Exam technique

Not checking for zero error before using a micrometer, callipers or balance.

Reading from the side of an analogue scale (parallax).

Quoting an uncertainty as the instrument's resolution without considering human reaction time.

Measuring one very small object when a multiple (a stack, many oscillations) would give a smaller percentage uncertainty.

Recording observations that describe what should happen rather than what was seen.

Try it in the virtual labMeasuring with precision \n