Target population is the entire group of people or animals or objects to which the researcher wants to generalize the findings of the experiment, e.g., all females with major depression disorder, all children below 10 years with a learning disability, all males above 65 years with dementia. Target population should be well-known and access to the individuals in the target population should be planned. If the population is spatially widespread and not easily accessible, it may cost a great deal to collect data with a large sample size.
Statistical Attributes of Sample Size
There are a few important statistical parameters in calculating sample size, such as statistical power, significance level, effect size, and standard deviation. The statistical power and significance level are fixed by convention, but the effect size and standard deviation need to be computed from previous studies. These parameters are described in the following subsections.
The statistical power (sensitivity) is the probability that a statistical test will detect a difference when a true difference exists. In other words, it is the probability of correctly rejecting a false null hypothesis. In experimental studies, sometimes the researchers fail to detect a difference when actually there is a difference (false negative). This false negative rate is referred in statistics by the letter в and known as type-II error. Statistically, power is equal to 1 - в. If the power of an experimental study is low, then there is a chance that the study will fail to detect true difference. The acceptable figure of power in statistics is 80%. However, above 80% power is a good study design.