Sampling

Sampling

Sampling is a process of selecting units from a population or universe of interest. Mostly, researcher selects only few items from universe for the study. If the researcher select whole universe for his/her study that is known as Census which is not possible to all study. If the universe is very large it is not possible to researcher to collect data from them so they select either probability sampling or non-probability sampling to select respondents to collect the data for the study.

There are certain conditions which should have to conduct probability sampling those conditions are:

• There should be complete list of subjects
• The size of the universe should be known
• All elements should have equal chance to get selected

Probability sampling

There all universes have equal opportunity to get selected which we cannot see in non-probability sampling. The researcher may use

1. Simple random sampling

This is the simplest form of random sampling. The main objective of the Simple random Sampling is to select sample size units out of Population size. This sampling method give equal opportunity to responds to the researcher. This type of sampling is known as Chance Sampling or Probability sampling where each and every item in the population has an equal chance to inclusion in the sample.

Steps follow to achieve a Simple Random Sampling

• Assign number to each population unit
• Decide the sample size
• Required sampling unit can be selected either by lottery method or using random number table.

2. Systematic random sampling

This is a method of selecting a sample when a list of the units of the population to be sampled is available. There are the general steps needed to follow in order to achieve a systematic random sampling:

• Assign number to each population unit
• Decide the sample size
• Obtain sample interval
• Randomly select first integer between interval size

3. Stratified random sampling

This technique is applied if the population from which a sample is to be drawn does not constitute a homogenous group. In this techniques population stratified into a number of non-overlapping sub-population or strata and sample items are drawn from each stratum adopting simple random sample of fraction. If the item selected from each stratum is based on simple random sampling the entire procedure, First stratification and then simple random sampling.

4. Cluster random sampling

Cluster or area sampling is used when the researcher does not have complete information of population but has information about cluster. It is the grouping of population and then selecting the cluster/groups rather than individual elements/units for inclusion in the sample.

Steps followed in Cluster sampling

1. Divide population into cluster
2. Randomly select the cluster into non-overlapping area
3. Measures all units within selected clusters
4. Multi-Stage Random sampling

This technique is applied for a big inquiries extending to a considerably large geographical area. Under this sampling first researcher select large primary sampling units such as regions, then zones, then districts, then VDCs then Town then individual. By combining different methods researcher is able to achieve a rich variety of probabilistic sampling methods that can be used in a wide range of social research context.

Non-Probability Sampling

Non-Probability Sampling methods can be useful when descriptive comments about the sample itself are desired. They are quick, inexpensive, and convenient. There are also other circumstances, such as in applied social research, when it is unfeasible or impractical to conduct probability sampling. The following are the techniques to use Non-Probability sampling:

1. Convenience Sampling

A sample is selected from a readily available list like telephone directory, voters’ list; however do not follow random selection procedures. The researcher studies all sampling items which are most conveniently available. Most of the pilot tests and pre-tests use connivances sampling. This sampling is easy and less expensive compared with other methods.

2. Purposive sampling

In this sampling, the researcher sample with a purpose in mind. And usually would have one or more specific predefined groups that the researcher is seeking. Purposive sampling can be very useful for situation where the researcher need to reach a target sample quickly and where samplings for proportionally is not the major concern. This sample method is considered desirable when the universe happens to be small and known characteristics.

3. Quota sampling

The quota sampling method is the non-probabilistic analogue of stratified random sampling. However, it differs in how the units are selected. In Quota sampling the units are selected by the interviewers. This result in selection bias. Thus, quota sampling is often used by market researcher.

4. Volunteer sampling

This method is highly applicable when the sample has pleasant measurement procedures such that probability sampling methods cannot be used.

5. Snowball Sampling

In snowball sampling, the researcher may begin by identifying someone who meets the criteria for inclusion in the selected research problems. The researcher then asks them to recommend others who they may know and also meet the set criteria. Although, this method would hardly lead to representative samples, there are times when it may be the best method available. The researcher begins the research with few respondents whom are known and available. This method of sampling is especially useful when members of universe cannot easily be located or hard to find by probabilistic procedures. An important bias in this sampling is that the more other who know a given person the more the likely the person is to be selected in the sample.