All
  • All
  • Insight
  • Research Services
  • Preclinical CRO
  • Animal Resource Bank
CN
中文 English
Latest News

Latest News

Focusing on cutting-edge technologies and research advancements,
sharing Aniphe BioLab’s achievements in scientific innovation, Professional insights
and practical experience in the fields of laboratory animals and preclinical research.

Practical Guide | How to Randomize Animals into Groups? How Many Animals per Group? How Should Animals Be Assigned?

Introduction:

 

Many researchers new to animal experimentation often struggle with how to design experimental groups—uncertain about how many groups to create, how to allocate animals, or how many animals should be assigned to each group. Today’s article aims to address these questions for you. 。 How many groups?

 

As everyone knows, animal experiments typically involve the establishment of numerous experimental groups. The design of these groups and the number of animals assigned to each group directly affect the reliability of the final results and the validity of statistical analyses. Moreover, once data are collected, it is often difficult to implement corrective measures. Therefore, the allocation of experimental animals into groups is a critically important step in the entire experimental process.
 
 
 
 

How do I set up groups?

 
 
 
Method 1: General Grouping Method     
They are divided into: a self-control group and a parallel control group. The parallel control group is further subdivided into a positive control group and a negative control group (blank control group).
Self-control group: The relevant monitoring data before and after the animal experiments will be compared.
Positive control group: The same experimental procedures are performed on the control group, but the target factor under investigation (e.g., the therapeutic drug, the method of administration, or the site of administration) is excluded.
Negative control group (Blank control): No treatment is applied.

 

Method 2: Three-group control design     
First Step: Set up a blank control (negative control) group, a model control (positive control) group, and a positive control group.
Model control group: Administer a negative control (solvent or excipient) to disease-model animals;
Positive control group: Administer an effective drug treatment (or other known effective intervention) to disease-model animals.
Step two ： Design three or more measurement gradients.
The final grouping is:

 
Grouping can be flexibly adjusted based on the specific circumstances of your experiment. For example, if a disease model is not required in the experiment, the model control group may be omitted; similarly, if no positive control drug is included in the experimental design, the overall study need not include a positive control group.

 

 
 
 
 

How do you determine the number of animals in each group?

 
 
 

First, we need to know that, A larger sample size is not always better. . A large sample size does not necessarily make the final data more authoritative; on the contrary, it can increase the difficulty of experimental procedures, waste research funds, compromise rigorous quality control, and even violate the principles governing the use of laboratory animals. 3R Principle ”。

Conversely, an insufficient sample size may prevent detectable differences from emerging in the final data, leading to negative findings and unsatisfactory results; moreover, the experimental conclusions would lack statistical significance and persuasiveness.
Method 1: The simplest and most widely applicable     
Under the principles of the 3Rs for laboratory animals and in accordance with statistical requirements, each group shall have a minimum sample size of six.
In each group, there should be at least 10 experimental mice, 6 rats, and for larger animal species, the sample size may be appropriately reduced by 1–2 individuals.

 

Method 2: Estimation of Degrees of Freedom (E)     
The degrees of freedom (E) is an important metric in analysis of variance, and its calculation formula is:
When the value E < 10, it indicates that the sample size is too small;
When 10 ≤ E ≤ 20, the sample size is considered normal.
When 20 < E, it indicates that the sample size is excessively large.

 

Method 3: Rigorous Sample Size Calculation     
If the sample size requirements in an experiment are very stringent, more complex formulas can be used to calculate the sample size.
When the sample size is calculated based on the means of two groups , the calculation formula is:
When the sample size is calculated based on multiple groups of means , the calculation formula is:
Of the three methods mentioned above, the latter two are more commonly employed in clinical research, as they account for individual variability within populations. However, the laboratory animal strains we use exhibit high genetic purity and minimal inter‑individual variation; therefore, in most animal studies, it is not necessary to consider such variability. Instead, one can directly adopt the first approach while consulting relevant literature and seeking guidance from experienced colleagues.
Special Note: If there are losses in the experiment (e.g., animal mortality), it is advisable to add a number of additional samples equal to the expected loss to the original sample size.
 
 
 
 

How should experimental animals be allocated?

 
 
 
To minimize inter-individual variability and ensure the validity of the final experimental data, random allocation is recommended.
The allocation method should be determined according to the specific experimental objectives and study subjects; common methods include the completely randomized design and the randomized block design, among others.
Method 1: Completely Randomized Design     
Number the animals one by one, then use a random number table to assign them to groups.
This method is applicable to animals with relatively small inter-individual differences (e.g., age, height, sex, etc.).
Method 2: Randomized Block Design     
All experimental animals are divided into several blocks (with a recommended number of blocks greater than 3), with the number of animals in each block set at twice or more the number of treatment groups. Subsequently, the animals within each block are randomly assigned to the respective experimental groups according to a completely randomized design.
This method is applicable to animals exhibiting substantial inter-individual variability (e.g., in age, body weight, sex, etc.).
 
That concludes all the key takeaways for this episode. Once you’ve completed all the steps above, congratulations—you’ve successfully finished one of the crucial preparatory tasks for your experiment!

References

Common Methods for Random Grouping of Laboratory Animals, Laboratory Animals, Issue 4, 2009

Randomization Methods in Clinical Trials. Chinese Journal of Traditional Chinese Medicine, 2007, Vol. 48, No. 3.

National Research Council. 2011. Guidance for the Description of Animal Research in Scientific Publications. Washington, DC: National Academies Press. Available online (http://www.ncbi-nlm.nih.gov/books/NBK84205/)

Festing M F, Altman D G. Guidelines for the Design and Statistical Analysis of Experiments Using Laboratory Animals[J]. Ilar Journal, 2002, 43(4): 244-258.

Michael F. W. Festing. Randomized Block Experimental Designs Can Increase the Power and Reproducibility of Laboratory Animal Experiments[J]. ILAR Journal, 2014.

Zhao Wei, Sun Guozhi. Common Methods for Random Allocation in Laboratory Animals, Laboratory Animals, Issue 4, 2009.

Guidance for the Description of Animal Research in Scientific Publications

 

 

Organizer  |  Aniphe BioLab Marketing Department

Supervising Producer  |  Solar wind

Editor-in-chief  |  Lemon

Editor-in-chief  |  Solar Wind, Aaron

Editor  |  Lemon

PREVIOUS:

Mouse Classroom | Do nude mice really have no fur? How should you choose between OB and DB strains?

NEXT:

Literature Review | New Advances in Alzheimer’s Disease Research—Turns Out It May Not Be Related to β-Amyloid?