Sample sizes reduce expenses and time by allowing researchers to estimate information about a whole population without having to survey each member of the population.
Before any clinical studies or polls are taken, statisticians usually determine how many individuals, or what sample size, should be sufficient for conclusive results. Different formulas help to determine this number, represented by "n," depending upon the type of estimator needed.
The larger the sample size, the more accurate the results will be. For example, if we wanted to determine the average weight of men and measured seven men, we might get an estimate of 236 pounds, inflated by one obese man in the mix. If, however, we measured 300 men, our extremes would likely balance each other out and produce an estimate closer to the true average, which is around 180 pounds.
Standard error is the measure of how inaccurate an estimate could be. Standard error depends, in part, on sample size because the larger the sample size, the smaller the likelihood of error. If you measure 90 percent of a population, you are more likely to know the truth than if you measure 15 percent and generalize from there. That is, your likelihood of error lessens as your sample size grows.
Sometimes a sample size can be very small. For instance, in a 2010 clinical study on the effects of whole grain wheat cereal on the cholesterol levels of obese adults, researchers used only 402 men and women to generalize about a population of more than 250,000 Americans. That sample size is less than 0.5 percent of the actual population. Sample sizes can be even smaller. For example, a study published in the "Nutrition Journal" based its results--that the omega-3 fat in fish oil helps children with ADHD--on a sample size of only nine children. In some cases, even very small sample sizes can provide enough evidence for statisticians to draw conclusions, but larger sample sizes definitely help.