Discuss and compare and contrast all the sampling methods used for data collection.

The primary data collection demands the researcher to collect real data or first-hand data for analysis. In most of the prior data collection events, it becomes rarely possible for the researcher to collect data from each and every person in the group, or rather population. The sample is the small collection of people who becomes the representatives of the population and who actually participate in research. Hence for accurate representation, it becomes necessary that we chose the sample with great care. For the same, the researcher requires to administer sampling methods depending on the characteristics of the research and the population selected for the study. Broadly, we have two types of sampling:

1. Probability sampling
2. Non-probability sampling

Probability sampling is the random selection of representatives or participants, which will allow the researcher to use statistical inferences on the group. Here every member of the population has equal chances of being selected as a sample, i.e., it follows the law of probability.

Non-probability sampling is the non-random or convenience-based selection, which will make the data collection easy. Here the chances of getting selected as a sample cannot be predicted.

• TYPES OF PROBABILITY SAMPLING

1. Simple random sampling: Here, the sample is selected in a random manner. The advantage of this method is that it is highly effective if all subjects participate. But it comes with the disadvantage of a high level of sampling error, especially when the sample size is small.
2. Stratified: Here, the selection of the sample is based on subgroups or specific strata. It comes with the advantage of preciseness in case of homogeneity and heterogeneity within the strata. But the disadvantage is that it is difficult to apply in practical levels and will not work in the absence of knowledge regarding the different strata.
3. Systematic: Here, the sampling is done in such a way that every Nth member in the population is included in the sample. The main advantage is that it is time-efficient and cost-efficient but fails to work if periodicity exists, leading to sampling biases.
4. Cluster: Here, samples are identified as clusters. This method also has the advantage of being time-efficient and cost-efficient. But this method fails to work when group-level information is less known. It also generates a high level of sampling errors compared to other methods.

• TYPES OF NON-PROBABILITY SAMPLING

1. Convenience sampling: This method selects samples based on their accessibility, i.e., the most accessible persons to the researcher are chosen. The main advantage of this method is that it is easy to administer and inexpensive. But it has the drawback of not able to decide whether the sample is an actual representation of the population, so difficult to produce generalized results.
2. Purposive sampling: As the name suggests here, the researcher uses their judgment to select the sample based on the purpose of the research. This method is mostly used for qualitative researches, where the researcher is studying a specific phenomenon. But to administer this method, the researcher needs to have clear criteria and rationale for inclusion.
3. Volunteer sampling: This is similar to convenience sampling since it is also based on ease of accessibility. But here, instead of the researcher approaching the sample, the sample volunteer to be a part of the study. It has the same advantage of convenience sampling, i.e., cost and time-efficient. But the disadvantage is that there may be biases since some people are more likely to volunteer than others.
4. Snowball sampling: This method is used when it is difficult to access the population, and here the participants are selected through other participants. In short, the participants act as "snowballs," and they nominate additional members to participate in the study. This has the advantage of finding the hidden population and recruiting them for study. The main disadvantages found here are the over-representation of a particular network and the reluctance of the participant to nominate additional people for the study.