Helping Research writing for student & professional researchers Research Writing
Stratified random sampling is a type of random sampling or probability sampling. Random sampling ensures that every member of the population has an equal and independent chances of selection. In a population where there is low heterogeneity simple random sampling is very effective. In case the population has elements that does not share common characteristics the researcher should use another sampling method.
A very effective method is to use stratified sampling instead of simple random sampling. It helps reduce and manage the heterogeneity of the population. In stratified sampling each strata is mutually exclusive and collectively exhaustive. Also, each element in one strata share a common basis, heterogeneous elements of the population are easy to manage. The shared attributes may include income, education, age, gender, race, social status, or common ailments etc.
Procedure for stratified random sampling
In stratified sampling the researcher stratifies the population in clearly identifiable strata based on some common characteristics. The process of classifying the elements in separate strata is called stratification. Within each strata the researcher selects the sample randomly.
Benefits of stratified random sampling
- It assures that all groups in the population have equal chances of representation in the sample. This makes the sample highly generalizable and representative of the real characteristics of the population.
- Since the researcher classifies the population in strata therefore, it is easier to make comparisons between different groups of the population. It also makes it easier to estimate the population characteristics.
- The variability of the population is easy t reduce in this way.
Drawbacks of stratified sampling
- Stratified random sampling requires accurate information abut the population characteristics in order to classify it in strata.
- Sometimes it is expensive to conduct stratified random sampling especially when the population is large and widespread.
Example
An example of stratified sampling can be to study the prevalence of hypertension in a specific population of elderly people. So, to study this we may need a sample that includes both genders, participants from different socioeconomic backgrounds, or different occupations etc. The variation of this population is hard to control in a simple random sampling but it is much easier in stratified random sampling.