Maths and Stats Support for Research Projects - in development

Planning the Project - thinking about your own objectives

There are often different possible project plans, even when investigating similar scientific problems. Your optimum plan will depend on the laboratory time and facilities available and the detailed objectives of your particular investigation.
Planning your own project is a process of iteration – linking analytical techniques to possible experimental data and with realistic objectives which match the time and facilities available.

The steps below will take you through a series of steps which will build up a provisional picture of your intended project. As you progress you may need to modify the details, but it is important to start with a good overall plan.
There are links at each step which provide
• detailed actions that you should take plus some theoretical backup,
• a range of illustrative examples which will help focus your own proposals, and
• video 'question and answer' tutorial.

Step 1:
The first step is to define the purpose of your research by setting out your overall aim and the factors involved in your study.
This will help you begin to focus in on the details of your research
Action and Theory - Examples of aims and factors - QVA Tutorial

Step 2:
Identify all the response and factor variables that you will record in your experiment, and decide on the type of each of those variables.
This will help you clarify exactly what is involved in your experiments
Action and Theory - Examples of variables - QVA Tutorial

Step 3:
Identify the source of variability in your data. For example, do you expect your data values to be normally distributed.
This will be essential for selecting possible methods for data analysis or transformation of data.
Action and Theory - Examples of distributions - QVA Tutorial

Step 4:
Select appropriate methods for describing your data, based on the variables involved and the analyses required.
It is often useful to visualise the data that you have recorded, either in graphical form or as summary results.
Action and Theory  - Examples of data descriptions - QVA Tutorial

Step 5:
If required, select appropriate data analysis methods for managing your data.
Identify the need for common techniques, e.g. straight line or exponential calculations, data transformation, handling uncertainties, etc.
Action and Theory - Examples of data analysis techniques - QVA Tutorial

Step 6:
If required, select appropriate methods for modelling your results to describe the system your are investigating.
It is possible that this step may come before or after using any hypothesis tests selected in step 8.
Action and Theory - Examples of data modelling - QVA Tutorial

Step 7:
If required, select appropriate hypothesis tests, based on the experimental design and the types of variables involved.
It is possible that more than one analytical method may be available, and the options may suggest a revision of the experimental design.
Action and Theory - Examples of hypothesis tests - QVA Tutorial

Step 8:
If required, select appropriate methods for handling and analysing multivariate data.
Where you have more than one response variable, it is useful to consider ways of describing and analysing them together.
Action and Theory - Examples of hypothesis tests - QVA Tutorial

Step 9:
Review the practicalities of your initial design, considering wider factors (e.g. laboratory time) and using ‘dummy’ data values.
It is important to consider the whole experimental process and estimate possible data values to check that your plans are realistic.
Action and Theory - Examples of planning - QVA Tutorial

Step 10:
Consider the presentation of data and reporting of results that will be required in the final project report.
It is useful, while still collecting results, to have an awareness of how the data might be presented graphically or in text.
Action and Theory - Examples of data presentation - QVA Tutorial