How to write a good physics IA?

Internal Assessments in IB Physics can very easily boost your exam grade. And it’s not particularly hard to write one that scores 5/6 or 6/6 for each section. Today, let us first focus on the Evaluation Component, which IB examiners report to be the most difficult section.


As a general rule, you should include at least 3 weakness and 3 improvements. And most importantly, you need to include the EFFECT of the weakness (how does it affect your results). And you need to discuss the EFFECT of your improvements (what will this particular improvement do for your results?). Also, make your improvements realistic – you can NOT conduct your experiment in zero gravity in space!

So basically, you need to structure it like this:

Sources of random error and possible improvements
Source 1:
Improvement 1:
Source 2:
Improvement 2:
Sources of systematic error and possible improvements
Source 1:
Improvement 1:
Source 2:
Improvement 2:
N.B. It may also be a good idea to include subheadings to make it very clear to the examiner that you know what is required.
Below is an example of one of my conclusion and evaluations that scored 6/6: (Note: Don’t plagerise this. It’s on the internet already)
This practical is about measuring the effect of changing the area of a wire and the length of a wire on the resistance. (Again, don’t try to plagiarize this)


The line of best fit of the Length v Resistance graph
shows a straight line going through origin. Let the slope of the line be k.

R = k L

And thus, R α L (R is directly proportional to L)

The line of best fit of the graph of Cross Sectional Area v 1/Resistance shows a straight line going through origin. Let the slope of the
line be k. Therefore:

R = k 1/A

And thus, R α 1/A
(R is proportional to 1/A)

to the textbook:

R α L

R α 1/A

(Source: Physics for the IB Diploma, K. A. Tsokos, 2008)


Precision and accuracy:
All of the points of the graph lie very close to the line of best fit. Thus the data is fairly precise. Also, the uncertainties involved are all very small, except for the relative uncertainty for the cross sectional area of the wire. This is because the wire is not of uniform thickness. Nonetheless, it is still fairly precise.

Furthermore, the line of best fit of the data points for each graph show linear relationships that go through origin. This is accurate as it is the same as that of the theory. The value of k (i.e. the slope) however, does not have a text book value (nor was it given by the manufacturer) as it is different depending on the wire, thus it is difficult to ascertain
whether or not the k value in the experiment is accurate.

of random error and possible improvements:

Source 1: It was noticed that the readings on the Ammeter and the Voltmeter flickered between values. It was the hundredth digit (i.e. 0.01, 0.02, etc.) that flickered, often between 2 to 4 values. This created some uncertainty in the readings.

1: The uncertainty caused by this is difficult to minimize without changing to a more precise Ammeter/Voltmeter, perhaps one that can measure Voltage and Current to 4 decimal places. The readings may still fluctuate, but it will be a less significant digit that is fluctuating. And since this was the cause of most of the error involved with the Voltage and Current, it would significantly reduce the uncertainty. However, the uncertainty is already fairly low and the results of the experiment verified the theory, so, for the purposes of this practical, the improvement would not be completely necessary.

Source 2:

It was noticed that the wires would heat up after a couple of seconds of running a current through them. This means that the resistance of the wires will increase. This will cause some uncertainty in the readings of current and voltage.

Improvement 2:
To minimize this problem, time should allowed for the wires to cool after each experiment. Also, the recording should be done very quickly after the power source is turned on, to minimize the time the wire has to heat up. This will help to ensure that the temperature is similar for each reading. During this experiment however, there was insufficient time, which made this difficult to implement.

Source 3:
The thickness of the wire was not constant throughout the length of the wire. This means that different sections of the wire will create different amounts of resistance. As a result, this would effectively mean that the experiment would be conducted on a slightly different wire each time. This was the main cause of uncertainty in the cross-sectional area of the wires.

Improvement 3:
Since the wires are very thin, it would be very difficult to alter the thickness using the equipment available (i.e. school equipment). Furthermore, there are no wires that are exactly uniform in thickness throughout the whole length of the wire (microscopically,
this would not be feasible). Thus, this is a very difficult error to fix. A possible improvement may be to use thicker wires, where the relative uncertainty caused by the varying thicknesses will be smaller (i.e. the variations in thickness would account for a smaller percentage of the whole thickness). This would also make it easier for the alligator clips to clip onto the wire. This improvement could significantly help to reduce the uncertainty in the practical.

Improvement 4:
Random error can often be reduced by increasing the number of results taken. This however, requires more time, so extra time available to conduct the experiment next time may help. This will allow more sets of data to be collected. However, increasing the number of results taken can only reduce random error to a certain extent.

Sources of systematic error and possible improvements:

Source 1:
The measurement of the wires was difficult, because the wires were not completely straight, i.e. they had to be stretched out in order to be measured. Furthermore, the cutting process, using the scissors was also difficult, as it was hard to get the cut exact. In addition to this, it was difficult to clip the alligator clips perfectly onto the ends of the wire. Consequently, the wires became effectively shorter.

Improvement 1:
One possible improvement may be to clip the clips onto the wires before measuring out the length. That way, the length lost due to the clips would be taken into account. The only problem with this is that the wires used were too thin and clipping was difficult. Another solution may be to mark out the length, and leave 1 centimeter either side and then adjust the clips. This however, would still have an uncertainty involved, because the clips also take up space, i.e. the exact point where the clips join the wire is uncertain.

So there you have it! Remember the key elements:
– 3 weaknesses and 3 improvements
– The effect of the weaknesses and improvements

Good Luck! And Wishing You Success!

Written by Owen Yang