Presentation of Data
A surefire way of getting data that are unreliable is to jot them down on a scrap piece of paper with no semblance of order. While you may know what the data items refer to when you are taking them, it is very easy to forget. Or the bit of paper gets lost. It is depressing to hear the excuse for an experiment not being handed in, "Sean's got my results".
Data are recorded on a table of results, and each student must have its own set of data. This should be drawn up before you start your experimental work. The picture below shows a neat table of results. While it may not be of an experiment at AS level, it still shows the principle:
Your data must be recorded to the precision of the instruments. The averages from repeat readings must be recorded to the same number of significant figures as the data.
If you process data, the number of significant figures must be no more than the minimum number of significant figures. Let us look at the resistance of a length of resistance wire:
Length (m) 
Current (A) 
Potential difference (V) 
Average (V) 
Resistance (W) 

Reading 1 
Reading 2 
Reading 3 

0.100 
0.25 
2.30 
2.25 
2.32 
2.29 
9.0 
0.200 
0.25 





0.300 
0.25 





0.400 
0.25 





0.500 
0.25 





0.600 
0.25 





0.700 
0.25 





Etc… 






The p.d. readings are consistent with the readout of a digital voltmeter.
The current was kept constant at 0.25 A. The number of significant figures is consistent with the precision of an analogue ammeter.
The average is consistent with the number of significant figures of the p.d. readings.
The resistance is calculated to two significant figures. It calculated to 9.04 W, but is written as 9.0 W to make it consistent with the current which is recorded to two significant figures.
When you draw the graph, remember the rules. Graphs should always:
Be large – they should occupy the whole of a side of A4 paper.
Have a title to tell the reader what the graph is about;
Have sensible calibration of axis, with simple scales;
Have both axes labelled with the quantity and units.
And do them in pencil, please. Also be a good chap and don't produce a graph like this:
Graphical Skills
On
their own numbers do not mean a lot.
A table of numbers can be confusing.
A graph allows us to see a picture of how the numbers relate to each
other.
Always
use a sharp pencil and a ruler.
Draw
the axes
Label
the axes with the quantity and the units
When
you plot Quantity 1 against Quantity 2, you put Quantity
2 on the horizontal axis.
Look
for the highest value in each range.
You calibrate (put numbers on) your axes to the nearest
convenient step above your highest value.
Use
a sensible scale.
Plot
your points with crosses (+ or ×).
Points get lost.
Join
your points with a line, but not dottodot!
It
can be difficult to decide whether a set of results is a straight line or a
curve. If
it’s clearly a straight line, draw your line of best fit with a ruler.
If the graph is a curve, then try to make a smooth curve.
A flexicurve can help you with this.
If
a point is way out from the rest, then it’s probably an anomalous result.
If you can, recheck the data or do that part of the experiment again.
If not, ignore it.
The
table below shows some data to plot:
Voltage
(V) 
Current
(mA) 
0 
0 
1 
20 
2 
30 
3 
65 
4 
98 
5 
174 
6 
280 
This graph is nonsense.
Can you see why?
Although graphs drawn like this are quite useless, they are depressingly
common.
Notice:
Axes
labelled with quantities and units;
Scales
are sensible;
Line
of best fit drawn through the points. No dottodot.
When we read a point off the graph within a range, we are interpolating. When we extrapolate a graph, we are extending it beyond the plotted range, making a reasoned guess as to where the line is going to go.
In
the Exam:
It
is important that you understand that the exam not only tests knowledge with
understanding, but also tests a range of other skills.
One of these is to be able to interpret and transfer data from a variety
of sources such as tables and graphs.
Therefore graphical skills are essential for success and are widely used
to decide whether a candidate gets a grade A or B, or at the other end, an E or
U.
All graphs in physics are line graphs. A set of results is not very easy to use to judge the way quantities are related to each other. The graph shows it instantly.
Graphs should always:
Be large – they should occupy the whole of a side of A4 paper.
Have a title to tell the reader what the graph is about;
Have sensible calibration of axis, with simple scales;
Have both axes labelled with the quantity and units.
You can draw a graph in portrait (long side vertical), or landscape (long side horizontal). Sometimes it’s quite clear which way it should go. Other times it’s a matter of what seems best.
In general, the independent variable goes on the horizontal axis (the xaxis, or ordinate), and the dependent variable on the vertical axis (yaxis or abscissa). When you are told to plot quantity 1 against quantity 2, quantity 2 goes on the horizontal axis.
Sometimes the scales go the other way round, for example voltage (independent variable) against current (dependent variable). This enables us to work out the resistance from the gradient.
A line of best fit needs to be drawn, because the data points will always be slightly out.
Consider whether the origin is a valid data point. For example, at zero volts, we get zero amps. So in a voltagecurrent graph, the origin is a data point in its own right.
A straightline graph that goes through the origin shows that the two quantities are directly proportional.
In this case:
If the force is doubled, the extension is doubled.
If there is zero force, there is zero extension.
Descriptions like “…if weight increases, the extension increases…” are too vague and will not gain any credit.
This graph does NOT go through the origin:
A student draws a graph as below and joins the points with a “line of best fit” as shown.
(a) The title of his graph is _________________ against _____________________. (b) What is wrong with the line of best fit? (c) Draw the correct line of best fit. (d) In the exam, he writes that Quantity A is directly proportional to Quantity B. Discuss how you would mark his answer. 
Sometimes the data form a curve. Draw a smooth curve, not doing dottodot.
Do not force a straight line through the curved progression.
Decide whether the origin is a valid data point. If it is, include it.
Sometimes it is not at all easy to decide whether the graph is a curve or a straight line. In this case, you should take more data points.
You may well get a data point that does not fit in with the rest of the data. This is called an anomalous result.
If possible, check out anomalous results by doing a repeat. Your new data will most likely fit in with the other points, and the anomalous data can be discarded.
The gradient will give a reading that could be used directly, for example resistance is the gradient of the graph of voltage against current.
For other relationships, you may need to do some further processing to get the value that you want.
Consider this graph:
We know the formula for resistivity as R = rl/A. To get the area, we would need to measure the wire with a micrometer.
How would you work out the resistivity from the graph of resistance against length? 
First thing to do is to draw a line at 90 degrees to the curve, a normal line. Then draw the tangent at 90 degrees to the normal. Then work out the rise and run of the tangent, making sure that your triangle is at least 8 cm in its shortest dimension (of course).