In this second article in his new series on method comparisons, Stephen MacDonald moves on from his initial general introduction to consider experimental design and analysis, the sources of samples and the number required, and framework design.
Why bother with study design? Can’t we just run a few samples through and do some correlations? Well maybe, but more often than not this leads to trouble. It is better to run a smaller, limited experiment properly than to run endless, laborious and expensive experiments, only to find at the end the data we require are not available.
Design of experiments is an entire field of science of its own. Hopefully, what this article will demonstrate is that if we design our experiments well at the beginning, a lot of pain is taken out of the process. A well-designed experiment should comprise 80% planning and designing, 10% performing the experiment, and 10% analysing and reporting the experiment. Performing ultra-modern statistical tests on poorly collected data cannot rescue a badly designed experiment. If we begin with the analysis in mind, life is that little bit easier.
Awareness of regulatory requirements
Log in or register FREE to read the rest
This story is Premium Content and is only available to registered users. Please log in at the top of the page to view the full text.
If you don't already have an account, please register with us completely free of charge.