Sampling – Meaning, Essentials, Advantages, Sample Size, Sample Errors
What is Sampling?
Sampling is the process of learning about the population, i.e., all the units of the universe, on the basis of a sample drawn from it. It requires selection of the sample, collection and analysis of information and making an inference about the population.
For example, if the objective of the study is to ascertain consumer satisfaction, it is not possible to get the necessary information from all the consumers. The researcher may, therefore, select a sample of 500 or 1000 consumers from the market and proceed to collect the required information from them.
Essentials of sampling
The following are some of the essentials of sampling.
1. The sample selected should be representative of the entire population. This may be achieved by using the random sampling method.
2. The size of the sample must also be adequate. The larger the size of the sample, the greater will be the accuracy of the results.
3. All the units of the universe should have the same chance of getting selected. The researcher should not use his judgement in selecting the sample.
4. There should be no basic difference in the nature of the units of the universe.
Advantages of Sampling
The following are some of the advantages of sampling.
1. A sample study is less time consuming. This is because it considers only a portion of the universe.
2. It is also less expensive as information has to be collected only from a limited number of respondents.
3. The results of the sample study may be more accurate. This is possible because the number of units to be considered is less. The researcher, therefore, can take every possible care to get accurate and complete information from the respondents.
4. It is also possible to get more detailed information from every respondent, as only few respondents need to be interviewed.
5. For certain studies, sample method alone can be used. For example, if a producer of bricks wants to test the breaking strength of the bricks made in his factory, such a test can be done only by using the sample method.
Drawbacks of Sampling
The following are some of the drawbacks of sampling.
1. If the sample is not representative of the population, it will affect the results.
2. The study results will also get affected if the sample size is inadequate.
3. There is chance for the bias element entering the process of sample selection. This again will affect results.
4. Sampling method cannot be used if each unit of the universe is heterogeneous.
Factors determining Sample Size
The following are some of the factors that determine size of sample.
1. If the size of the population is large, the size of the sample should also be large.
2. If the time and money available for the study are limited the researcher may have to settle for a small sample size.
3. If the researcher expects a higher degree of accuracy in results, he may have to go in for a larger sample size.
4. A large sample size may be required if the units of the universe are heterogeneous. On the other hand, if all the population units are homogeneous, a small sample size may be enough.
5. If the nature of the sample is a simple random sample, the sample size needs to be large. On the other hand, if it is stratified random sample, a small sample-size may be enough.
6. For an intensive and more detailed study, small sample is suitable.
Sampling Errors
Sampling errors may fall under two categories:
- Biased Errors and
- Unbiased Errors.
a. Biased Errors
Biased errors may arise due to the bias in the selection of the samples. Example: Deliberate selection of the random sample. The other causes of bias may be stated as follows:
- Deliberate selection of the representative sample.
- Non-existence of randomness.
- Substitution of an item in place of the one intended. For example, in the case of systematic sampling, if it is decided to contact the 1st, 5th, 10th, 15th household and so on in a locality, if the researcher contacts the 9th or the 14th household, as the 10th or the 15th cannot be approached, error creeps in.
b. Unbiased errors
Unbiased errors arise due to chance differences. For example, if a producer of bulbs checks 10 bulbs, in a batch of 100 bulbs, at random and all the 10 are faulty.
Non-Sampling Errors
Such errors occur due to any of the following causes:
1. Faulty Questionnaire or Schedule
2. Improper responses from the respondents.
3. Lack of trained enumerators.
4. Use of incorrect statistical tools for analysis etc.
