Quality is much more important than quantity
This means Higher level knowledge. You may need to link together knowledge from different topics you have studied throughout Years 10 and 11, and possibly some of the work you studied at Key Stage 3 which is not repeated in your GCSE course.
To gain the highest marks it is important that you select relevant information and actually use it to produce your plan. Just listing down everything you can find about a topic will not gain you high marks.
You are expected to identify all the variables, not just those that have a big effect. You only need to investigate the effect of changing one of the variables. If you investigate two variables you will not gain any more marks than if you investigate one variable well. Unless it is a particularly difficult experiment always choose continuous variables to investigate.
You must clearly show how all the control variables are controlled (they must stay the same) and measured.
The evidence must be to a suitable accuracy, in general the higher the better. Equipment and methods must be chosen to give the required accuracy.
If anyone, anywhere, anytime, can carry out your plan and get the same results (within your specified accuracy) then your evidence is reliable. Your results must be taken over a reasonable range of values, with at least five steps. For quick experiments aim for at least 8 or 10 steps. Your results must of course be repeated a suitable number of times to show they are reliable.
Always make a prediction.
A prediction is a scientific model to explain what will happen. You must use Higher level knowledge to explain why it happens. (To explain why something happens is to justify it.) Use scientific diagrams to explain your scientific model clearly.
In general you should be looking to produce a quantitative prediction.
e.g. variable A is inversely proportional to variable B because ...
For all laboratory based investigations you should be able to carry out preliminary work. If you know what you are doing you should be able to quickly try out all the important variables. You can then decide which variable is going to be best to investigate. You then need to find out what range of values you can investigate, and the best values for control variables to be fixed at. In your plan you should summarise relevant trial results and explain how you used them to decide on the values chosen in your final plan.
Text books, data books, experts, the Internet ...
Find as much information as possible, then sort through and pick out the bits
that are useful. Don't forget to make a note of where you got the information
from!
NOTE: You should already know what will happen before you write your prediction because you have carried out trial experiments. The purpose of your investigation is to compare your scientific model with reality. Some parts may agree well while other parts may not. Once you have obtained accurate, reliable evidence you can analyse it and then try to explain any differences.