# Two Important Probability Sampling Techniques

Sample, as we know is a part of the population under study. The sample must be the 'true' representative of the population in terms of characteristics possessed by the population elements. It is difficult, actually impossible to draw a true sample from a population. The survey research pro will always try to draw a sample which is as close to 'true' as possible. There are two fundamental types of sampling technique - probability and non-probability sampling techniques. This article tries to illustrate two of the more important probability sampling techniques.

1. Random Sampling - when all the elements have an equal chance of getting selected in the sample, it is termed as random sampling. It works like a lottery system where each ticket has an equal chance of winning. In a way, this method is the most perfect method. However, it is difficult to apply this method for two major reasons - availability of sampling frame (list of population elements) and scatter over the geographical area if the sample is small. Consider for example a population of one million car owners and survey research pro has to do a sample of 100 from this population. Imagine this sample of 100 from the whole population scattered across the country. On the other hand, if a survey research involving readership of a local newspaper is concerned (and the list is available), this method would be the most useful and easy to deploy.

2. Systematic Sampling - as the name suggests this method applies a very systematic approach to draw a sample from the population. Suppose a survey research has a population of 10,000 and the survey research pro would like to draw a sample of 200 from this population. First of all divide the population size by the desired sample size to result is what is known as sampling quotient. In our illustration, the sampling quotient will be 50 (10,000 divided by 200). Now select any arbitrary number between 1 and 10. Let's say 4 is that number. Now add sampling quotient to this number. The result will be 54 (4 + 50). Now add 50 again to this number to get 104 and so on. One should keep on adding sampling quotient to the resultant number. Thus the sample will consist of person / group / organization number 4, 54, 104 and so on. Main problem faced with this method are availability of sampling frame (list of population elements) and its constitution. Constitution will make a huge difference to the final sample listing. For example, if names of all the females in a list have a salutation Ms., they will be clustered together. On the other hand if they are scattered across the list (A to Z), their probability of getting selected in the sample will be different.

For more information about questionnaire design, data entry, survey research, survey analysis and report preparation, visit here: http://www.migindia.biz

Source: www.a1articles.com