Results should be described transparently

Few animals or events

Studies involving small numbers of animals or people may be inaccurate and could misrepresent the ‘truth’.

When conducting studies, we are always taking a small subset of individuals from a larger population. For this group to be generalisable back to the whole population it needs to be a representative sample. One important factor in ensuring this is to have sample sizes that are large enough.

In all situations, some participants will improve or decline regardless of what treatment they received. It is possible that by chance, there are more of these animals in one group than another. These individuals will influence the results; however, their effects on the data set as a whole are diluted in larger sample sizes.

It is like tossing coins. Imagine you have two coins that are the same and you flip both simultaneously. You would expect to get roughly the same amount of heads vs tails from tossing each coin in the air 100 times. But if you tossed each coin just five times, you might get “heads” four times with one coin and only one time with the other coin, just by chance.

It is important to note that when investigating groups or populations of animals, it is the number of patient groups that is important rather than the total number of individuals. For example, a study conducted with two flocks of birds only has two patient groups regardless of whether the flocks contain 100 birds or 10,000 birds.

REMEMBER: You cannot be sure if a study with small sample sizes is representative of the ‘truth’.