Randomization
True randomization is achieved when you are unaware which experimental group a study subject has been assigned to.
Click here to download a fact sheet on Randomization at the bench.
What is randomization?
Randomization is a method often used to eliminate ‘selection bias’ bias in experimental group assignments. Adequate randomization allows researchers to fairly compare groups. is comparable between groups. In other words, randomization allows researchers to fairly compare animal subjects that received Treatment A with animals that received Treatment B.
Who can randomize?
Both preclinical and clinical researchers can randomize study subjects (animals, trial participants). Although this technique is very popular in clinical research, it remains important in the preclinical field as well (even when animals are genetically homogenous – read on!).
When should we randomize?
Randomization should be performed before the study begins (i.e. before the intervention is introduced). Once the study groups are organized, animals are randomly allocated to each study group and then the respective intervention is introduced.
Why randomize?
Randomization eliminates selection bias. Selection bias happens when animals are purposely assigned to treatment groups based on their characteristics. For example, a researcher may be unconsciously more inclined to assign a very sick looking rat to the one treatment group rather than another. Due to this selection bias the different treatment groups are no longer considered representative of the intended population and it is unfair to compare them.
How can we randomize?
Below are two case studies illustrating proper randomization and something that can be confused for proper randomization.
Optimal Practices
You have sixteen rats that you need to assign to receive either Treatment A or Treatment B.
You number each rat 1 through 16 using indistinguishable ear tags. You then input numbers 1-16 into a website/software and randomly assign each subject ID to receive either Treatment A or Treatment B.
Using a website or software to randomly assign subject IDs to treatment groups preserves the unpredictability of each assignment.
In other words, you don’t know what treatment the next animal is going to be assigned to so you can’t pick a sick animal knowing that it is going to be assigned to the treatment group.
Suboptimal Practices
You have sixteen rats that you need to assign to receive either Treatment A or Treatment B.
You number each rat 1 through 16 using indistinguishable ear tags and assign the odd number rats to Treatment A and the even number rats to Treatment B.
Although it may not be obvious, sequentially allocating animals to groups is not true randomization because the assignment sequence is known and predictable.
Being able to predict the sequence can introduce bias (i.e. how the animals are handled, how outcomes are analyzed).
Resources and Tools
Tools to facilitate proper randomization:
Publications outlining importance:
- Bebarta, V., Luyten, D. and Heard, K. (2003), Emergency Medicine Animal Research: Does Use of Randomization and Blinding Affect the Results?. Academic Emergency Medicine, 10: 684-687.
- Hirst, J. A., Howick, J., Aronson, J. K., Roberts, N., Perera, R., Koshiaris, C., & Heneghan, C. (2014). The need for randomization in animal trials: an overview of systematic reviews. PloS one, 9(6), e98856.
- H. Bart van der Worp, Emily S. Sena, Geoffrey A. Donnan, David W. Howells, Malcolm R. Macleod, Hypothermia in animal models of acute ischaemic stroke: a systematic review and meta-analysis, Brain, Volume 130, Issue 12, December 2007, Pages 3063–3074
- Jerndal, M., Forsberg, K., Sena, E. S., Macleod, M. R., O’Collins, V. E., Linden, T., Nilsson, M., & Howells, D. W. (2010). A systematic review and meta-analysis of erythropoietin in experimental stroke. Journal of cerebral blood flow and metabolism : official journal of the International Society of Cerebral Blood Flow and Metabolism, 30(5), 961–968.
- Macleod, M. R., van der Worp, H. B., Sena, E. S., Howells, D. W., Dirnagl, U., & Donnan, G. A. (2008). Evidence for the Efficacy of NXY-059 in Experimental Focal Cerebral Ischaemia Is Confounded by Study Quality. Stroke, 39(10), 2824–2829.