Guide to Essential BioStatistics VIII: Designing and implementing experiments – Randomization and Experimental Design
In this eighth article in the LabCoat Guide to BioStatistics series, we learn about Randomization and Experimental Design.
In the previous articles in this series, we explored the Scientific Method and Proposing Hypotheses and Type-I and Type-II errors, Designing and implementing experiments (Significance, Power, Effect, Variance, Replication, Experimental Degrees of Freedom and Randomization), Critically evaluating experimental data (Q-test; SD, SE and 95%CI), and Concluding whether to accept or reject the hypothesis (F- and T-tests, Chi-square, ANOVA and post-ANOVA testing).
In biological trials, randomization is the process of randomly allocating experimental units (e.g., pots comprising treatments, references, untreated controls, as well as application timings and -doses) across the treatment groups.
Randomization provides a basis for the statistical methods used in generating and analyzing data by reducing bias, such as selection bias (our tendency to see patterns when evaluating experiments).
Scientific controls are part of the scientific method. Untreated controls are always included when designing experiments to confirm that observed effects are in fact due to the treatment.
To control for confounding variables, such as the physical effects of spraying plants, untreated controls in crop protection trials are typically sprayed with water, or with a solution of formulation compounds from which the active ingredient has been eliminated.
Types of design
For crop protection trials, the principal randomized designs in common use are the Completely Randomized Design (CRD) and the Randomized Complete Block Design (RCBD).
The Completely Randomized Design (CRD)
In a Completely Randomized Design (CRD) treatments are distributed at random. Excel, as well as commercial statistical packages, may be used to randomize trials.
For a trial with four treatments (one untreated control, UTC and three herbicide samples, H1, H2 and H3), we can see from the previous section that we need six replicates to attain 15df. Carried out on a greenhouse table, the pots could be randomly placed as follows:
Figure 1: Completely Randomized Design (CRD) for four treatments (one untreated control, UTC and three herbicide samples, H1, H2 and H3)
The Randomized Complete Block Design (RCBD)
Use of the Randomized Complete Block Design (RCBD) assumes that conditions across the greenhouse table (or field) are equivalent. In reality this is rarely the case, and we can expect variation in temperature, light and humidity depending on the experimental units (pots) proximity to windows, insulated walls, etc.
These differences may be reduced by optimal greenhouse design, but their presence and influence on plant growth and response may also be addressed statistically by restricting the randomization process to form blocks.
These blocks, containing all of the treatments, treatment levels or combinations (called complete blocks) are typically set up at right angles to variance gradients.
In the following, we take a temperature gradient (due to proximity to uninsulated glass panels and insulated interior walls) into account by placing blocks at right angles to the gradient.
Each treatment appears in each (complete) block, and are assigned at random within the block, and for 6 replicates we have six blocks:
Figure 2: Randomized Complete Block Design (CRD) for four treatments (one untreated control, UTC and three herbicide samples, H1, H2 and H3)
It is common practice in crop protection trials to ensure (for demonstration purposes) that a single block retains the treatment order as described in the experimental protocol.
With this information on Experimental Design, and the information on the Scientific Method and Hypotheses as well as Experimental Parameters presented in previous articles – we are ready to initiate our experiment and move on to the next article in this series: Data and Descriptive Statistics.
Thanks for reading – please feel free to read and share my other articles in this series!
For more information, visit BIOSCIENCE SOLUTIONS – Strategic R&D Management Consultancy.
The first two books in the LABCOAT GUIDE TO CROP PROTECTION series are now published and available in eBook and Print formats!
Aimed at students, professionals, and others wishing to understand basic aspects of Pesticide and Biopesticide Mode Of Action & Formulation and Strategic R&D Management, this series is an easily accessible introduction to essential principles of Crop Protection Development and Research Management.
A little about myself
I am a Plant Scientist with a background in Molecular Plant Biology and Crop Protection.
20 years ago, I worked at Copenhagen University and the University of Adelaide on plant responses to biotic and abiotic stress in crops.
At that time, biology-based crop protection strategies had not taken off commercially, so I transitioned to conventional (chemical) crop protection R&D at Cheminova, later FMC.
During this period, public opinion, as well as increasing regulatory requirements, gradually closed the door of opportunity for conventional crop protection strategies, while the biological crop protection technology I had contributed to earlier began to reach commercial viability.