File ITR, invest & save upto
₹46,800 in taxes on the go
0% commission • Earn upto 1.5% extra returns
Reviewed by Nov 11, 2021| Updated on
A stratified sample is one that ensures adequate representation of the subgroups (strata) of a given population within the entire sample population of a research study. The researcher will then randomly pick equal amounts of people from each age group to stratify the study.
Importantly, strata used in this technique should not overlap, because if they did, some individuals would have a higher risk of being picked than others. This would create a skewed sample which would make the research biased and the results invalid. Stratified Random Sampling (SRS) uses the most common strata, such as age, gender, educational attainment, socioeconomic status, and nationality.
There are several cases where researchers should choose stratified random sampling over other sampling types. Next, when the researcher needs to analyse subgroups within a population, this is used. This approach is often used by researchers when they want to examine interactions between two or more subgroups, or when they want to investigate a population's unusual extremes.
With this type of sampling, the researcher is guaranteed to include subjects from each subgroup in the final sample. In contrast, simple random sampling does not ensure that subgroups within the sample are represented equally or proportionately.
Using a stratified sample would often achieve higher accuracy than a simple random sample, provided the strata is chosen such that representatives of the same stratum are as similar as possible in terms of the new characteristic. The bigger the differences between strata, the higher the precision gain.
Administratively, stratifying a sample is always more efficient than choosing a completely random sample. For example, interviewees may be trained on how to best deal with one specific age or ethnic group, while others are trained on how to best deal with another age or ethnic group. In this way, the interviewees can concentrate and develop a specific collection of skills, and it is less timely and costly for the researcher.
One major disadvantage of stratified sampling is that the selection of appropriate strata for a sample may be difficult. A second downside is that arranging and evaluating the results is more difficult compared to a simple random sampling.