In the Research 101 blog series, we have discussed how clinical trials are designed and the strengths and weaknesses of these designs. One factor that can derail or strengthen any study is Sample Size.
So, what is Sample Size? Very simply stated, sample size is the number of people participating in a study. The more people that participate, the better the study is. Having a large number of participants reduces the risk of accidently having extreme, or biased, groups – such as having all adults or all children in a study that should have equal numbers of adults and children.
Let’s find out why:
Pretend that we have a jar with blue and white marbles. The jar represents our entire population. Half of the marbles are White and half are Blue.
White marbles = adults = ½ of the population is adults
Blue marbles = children = ½ of the population is children
We want a representative sample of people in our study – so ideally we want half of the participants to be adults and half to be children. Unfortunately, we decide that we only need 3 people in our study
We pick 3 people at random (sample size = 3)
There are 4 possible color combinations.
We need to come up with a better plan
After looking at the odds of randomly picking an extreme, non-representative group, we decide that we need a much bigger sample size. This time, we are going to choose 50 people to be in our study (sample size = 50). With a sample size of 50, there are 51 possible combinations of Blue and White marbles.
When 50 people are randomly chosen, there is only a 3.9% chance of having an extreme group composed of either all adults (White) or all children (Blue). Therefore, we are likely to pick a group that is more representative of our population.
This is a simple example illustrating one-way sample size is important for research studies. In real-life situations, many issues associated with extreme, or biased, groups can be avoided by increasing the number of people in each group.