Sampling error and non-sampling error

In the collection, processing, analysis and interpretation of the data some errors or inaccuracies are arises. These errors are broadly classified under the following heads:
1) Sampling errors
In a sample survey, we study a small portion or part of the population. So naturally there is a certain amount of inaccuracy in the information collected, the result obtained from it has a chance of being differ from the census survey. This type of errors is known as sampling error. Sampling  error is due to the fact that only a subset of the population i.e. sample has been used to estimate the population parameter and draw inference about the population.

Sampling errors occur randomly and the magnitude of the sampling error depends upon the nature of the population; the more homogeneous the population, the smaller the sampling error.

The main sources of sampling error are:
  • Improper choice of sampling technique: Improper choice of sampling technique causes the error in a survey. The error can be minimized by selecting a proper sampling technique for the purpose of study.
  • Improper substitution: If some difficulties arise in enumerating a particular sampling unit included in the samples, the investigator usually substitute a convenient member of the population. Such substitution leads to some error.
  • Improper choice of sampling units: Result of the sampling study are mostly depend upon the sampling unit. So, proper method should be used while selecting a sampling units.
  • Improper choice of the statistic in estimation: Sampling method consists in estimating the parameters of the population using appropriate statistics computed from the sample. Improper choice of the estimation techniques might introduce the error.
  • Variability of the population: Sampling error also depends on the variability of the population. Lesser the variability, lesser the error and vice versa.
2. Non-sampling errors
The errors arising at the stages of ascertainment i.e. responses or observation, processing and analysis of the data are termed as non-sampling errors. This type of errors are present both in census surveys and in sample surveys. The non-sampling error is likely to increase with increase in sample size, while sampling error decreases with increase in sample size.

The main sources of non-sampling errors are:
  • Faulty planning, include vague and faulty definition of the population or the statistical units to be used.
  • Observational errors due to defective measurement technique.
  • Errors introduced in editing, coding and tabulating the results of the study.
  • Personal bias of the investigator in the different stages of the study.
  • Non-response bias of the respondent.
  • Lack of trained and qualified investigators
  • Publication errors etc.

No comments:

Post a Comment