PURPOSIVE SAMPLING

Purposive Sampling This is also called judgment sampling. The investigator exercises his discretion in selecting sample from the universe. 
Consequently, there is an element of bias in the selection. From the standpoint of the investigator, the sample thus chosen may be a true representative of the universe. But the units in the universe do not enjoy equal chance of getting included in the sample. So it cannot be considered as a probability sampling technique. We come across this type of sampling mostly in social science research. The results drawn on the basis of such a sample cannot be applied to the universe as it violates the principle of the law of statistical regularity. 

This sampling technique only helps in making case studies without any consideration for generalized application. Even though the results of such samples are close to the population results on statistical grounds, the possibility of applying such a technique is ruled out. In general, the sample estimates differ widely from population parameters leaving room for large sampling errors. Further, the sample estimates also vary from one purposive sample to the other, thus, making it difficult to draw useful inferences.

Merits :Although, the principles of statistical theory are not followed, this technique is liberally applied in solving many economic and business problems. The justification for its application is provided on several grounds.
1. In a relatively small universe, if sampling techniques of this type are used, it is possible to give adequate representation to all the characters of the universe by careful examination. Application of scientific techniques in such cases would result in omitting of essential characters of the universe. 
2. In most cases, statistical studies are expected to give quick results for policy formulation. The purpose of deliberate sampling has become a practical method in dealing with economic or practical problems. 
3. In studies where the level of accuracy can marginally vary from the prescribed norms, rigid application of statistical techniques becomes unnecessary 
4. In social science research, one is concerned more about the probable results and at different stages the researcher has to resort to approximation. As a result, there is bound to be an error on account of approximations. Therefore, the application of deliberate sampling technique may not significantly disturb the results.
 
Demerits1. Fundamentally, this is not considered a scientific approach as it allows for bias or personal prejudice. 2. The investigator may start with a preconceived idea and draw samples such that the units selected will be subject to specific judgment of the enumerator. 3. The results of the sample cannot be extended to that of the population. Various samples drawn from a single universe give diverse results and in the absence of scientific application of statistical tools, it becomes difficult to establish the significance of statistical error.