Answer to Question #159253 in Management for Kamal

Question

Differentiate between Sampling and non sampling errors?( please give at least 10 points of difference)

Answer 1

A sampling error refers to a statistical error where the sample analyzed does not represent the entire population of data. A non-sampling error refers to an error that arises during data collection making the data to differ from true values.

Sampling error can arise even when no mistakes were made whereas non-sampling error is specific to a particular study design.

A sampling error can be reduced or eliminated by increasing the sample size or by randomizing the sample. It's hard to correct a non-sampling error and most times impossible to eliminate.

A sampling error can be measured mathematically whereas it is very difficult to detect a non-sampling error in a sample and hard to quantify.

Sampling error is a random type of error whereas non-sampling error can either be random e.g processing errors or non-random e.g. biased sample

Sampling error is caused by internal factors whereas non-sampling error is caused by external factors beyond the study or survey.

Sampling error is used to measure reliability rather than validity of the sample while non-sampling error affects both validity and reliability of the results. In case of systematic non-sampling error, the whole data is scraped off.

Sampling error is not fully avoidable because samples are analyzed on population parameters that are unknown and heterogeneous. Non-sampling error can be avoided by maintaining high standards in the measurement and processing systems.

The amount of sample error is inversely proportional to sample size while it becomes otherwise in the case of non-sampling error.

While a sampling error is not indicative of any bias, when a non-sampling error occurs, the rate of bias in a population parameter increases.

Sampling errors are categorized as either population specification error where the analyzer does not understand the specific type of respondents to be included or selection error where only a small sample is selected. Non-sampling error is categorized as processing, measurement, non-response, adjustment or interviewer bias error.