NON-PROBABILITY

Non-probability Sampling

Non-probability Sampling

Non-probability Sampling

Non-probability sampling comprises a broad category of approaches that are similar primarily in that subjective criteria are used in case selection rather than randomization (Patton, 2002). In the case of non-probability sampling, since we are chosen arbitrarily, there is no way to estimate the probability of any one element be included in the sample. Also, as the method in question provides no assurance that each item has a chance of being included in the sample, one can estimate the sampling variability or to identify bias as possible.

We cannot measure the reliability of a non-probability sampling; the only way to measure the quality of the resulting data is to compare some of the survey results to available information about the population. Again, there is no assurance that the estimates will not exceed an acceptable level of error. Statisticians are reluctant to use non-probability sampling methods, because there is no way to measure the accuracy of resulting samples (Casella and Berger, 2002). While non-probability samples lack the statistical properties that random samples have, there are often compelling reasons to use them, ranging from feasibility to convenience to specific criteria tailored to the research design. Some examples of non-probability sampling are as follows:

Quota sampling

Quota sampling which also sometimes called accidental, usually sits on the basis of a good knowledge of the strata of the population and / or individuals more "representative" or "adequate" for purposes of research. It remains, therefore, similarities with stratified random sampling, but not in the nature of randomness of it.

In this type of sampling is set a "quota" that consist of a number of individuals who meet certain conditions, for example, 20 individuals from 25 to 40 years, females living in Trujillo. Once the quota is elected the first to be found that meet these characteristics. ...