Sampling techniques are crucial skills for researchers to master. In research, samples can be drawn from a population, but the selected samples must accurately represent the observed population. Therefore, proper sampling techniques must be applied per scientific principles.

Researchers can consider several sampling techniques before collecting data in the field. Researchers can choose between probability sampling and non-probability sampling, depending on the target population’s characteristics.

The key distinction between these approaches is that probability sampling techniques offer an equal opportunity for each member of the population to be selected as a sample. In contrast, with non-probability sampling, each member of the population does not have an equal chance of being chosen as a sample.

In this article, Kanda Data will discuss probability sampling. The most commonly used sampling techniques within probability sampling are simple random sampling and stratified random sampling.

**The Difference Between Simple Random Sampling and Stratified Random Sampling**

Distinguishing between these two sampling techniques is quite straightforward. A fundamental principle that researchers must understand is that the data should be homogeneous when utilizing random sampling. If the data is heterogeneous, this sampling technique cannot be applied.

Nevertheless, there are differences in the application of simple random sampling and stratified random sampling. In simple random sampling, the entire population’s members must be homogeneous.

Therefore, researchers can randomly select samples from the entire population. Researchers can define a minimum sample size that represents the observed population.

On the other hand, in stratified random sampling, the population is divided into strata. However, it’s important to emphasize that the members must be homogeneous within each population stratum.

Consequently, researchers can randomly sample within each stratum. Furthermore, researchers can determine sample sizes proportionally within each stratum. This sampling technique is known as proportional stratified random sampling.

**Assumptions in the Simple Random Sampling Technique**

In the technique of simple random sampling, there are two main assumptions that researchers must fulfill. The first assumption is that the data from the population members must be homogeneous. The second assumption is the availability of a sampling frame from the observed population.

Understanding these two assumptions can be illustrated by a researcher analyzing rice farming practices in XYZ village. If the researcher gathers data from the entire population of rice farmers in XYZ village, the observed population members are considered homogeneous.

On the other hand, if the researcher collects data from rice farmers, corn farmers, and cattle ranchers in XYZ village, the collected population data falls under the heterogeneity category.

Once the researcher has homogenous data, as, in the first example, the next step is to ensure a sampling frame, the term “sampling frame” refers to the researcher having complete data on the population under study. This comprehensive data should be accessible to the researcher, including the names and addresses of population members in XYZ village.

Subsequently, the researcher can determine the minimum sample size representing the observed population. The researcher can utilize Slovin’s formula to determine this minimal sample size.

If the researcher meets both prerequisites, as discussed in the preceding paragraph, then the researcher can employ the simple random sampling technique.

**Assumptions in the Stratified Random Sampling Technique**

The assumptions for stratified random sampling are nearly identical to those in the simple random sampling technique. The difference lies in the presence of stratification or strata within the population.

For instance, if we consider a population divided by land ownership categories, let’s assume there are three strata. In the observed population, land can be categorized into three groups: the first category is less than 5,000 square meters, the second is between 5,000 and 10,000 square meters, and the third is more than 10,000 square meters.

The researcher can establish criteria for small, medium, and large land areas based on these categories. It’s important to note that each member of the population within each stratum must be homogeneous. Subsequently, the researcher can randomly sample from each stratum and determine the proportions to be sampled from each stratum.

After considering the differences and assumptions outlined for simple and stratified random sampling, it’s hoped that this will enhance readers’ understanding of the research methodology. Researchers can select the appropriate sampling technique based on the characteristics of the observed population.

If the observed population has a sampling frame and all its members are homogeneous, researchers may use the simple random sampling technique. Conversely, if there’s a sampling frame available, the population is stratified, and each stratum’s members are homogeneous, researchers can opt for the stratified random sampling technique.

Well, this is the article I can write for now. Hopefully, it’s beneficial and provides new insights for all of you preparing to conduct field research. Stay tuned for updates from Kanda Data in the upcoming week. Thank you!