9 Practicalities of the Experimental Design in development

 

You have already developed a general idea of what your are going to do in your research project and how you might analyse your results.

You now need to plan in detail the design of your experimental procedures, and whether this plan is feasible within your time and resource constraints.

 

The details of experimental design will differ widely between different types of research, from a simple choice of the number of replicates for a single sample of laboratory measurements to complex multi-factorial field or clinical trials.

 

ACTION: You now need to

In this section we only include the most common elements in experiment design, appropriate to smaller scale experiments.

Large scale experiments might require a more in-depth study of experiment design using alternative sources.

 

Sampling process

 

Choosing sample size

Selecting subjects (making measurements) at random

Allocation of subjects

 

Controlling for confounding factors

 

Repeated measures and pairing (related data)

Blocking

Handling covariance

 

Checking the practicalities

 

Availability of equipment, samples, technical support.

Time taken to perform all experimental work.

Other resource implications, cost, etc.

 

Performing a dummy analysis

 

Produce a layout (e.g. in Excel) into which you can enter your data as you perform your experiments.

It is also useful to generate some data that may represent, very approximately, the type of results that you might expect to get, and then to check that you can perform the mathematical/statistical analysis as expected.
This is a very helpful in refining your experimental design, e.g.

> for pointing out any values that you might forget to record,

> highlighting where you put the greatest effort in reducing uncertainties in your measurements, or

> ensuring that you don't throw away data that you don't think is relevant.