Non Probability Sampling:
Convenient sampling or accidental sampling entails the use of the most conveniently available people as subjects in the study. Quota sampling arises from researcher's desired quota or proportion for same population variable of interest. It is similar to stratified random sampling in that the first step involves dividing population into homogenous strata and selecting same elements from each of these strata.
i) Purposive or Judgmental Sampling involves "hand-picking" of subjects. The researcher might decide to purposely select widest possible variety of respondents or might choose subjects who are judged to be typical of population in question or knowledgeable about the issues under study.
Purposive Sampling is non-probability sampling in which the researcher attempts to identify typical cases whatever is being studied. For example, if you are studying cardiac patients, findings "the typical cardiac patient" would be one approach. If the researcher is interested in the play patterns of four year old village boys, then studying a small sample of "typical" four year olds attending the "Anganwadi" or pre-nursery schools, would comprise the purposive study sample. Obviously, this approach is not often a very dependable one, for what is "typical" is often a very subjective opinion. Sampling in this subjective manner provides no objective method for assessing the typicalness of the selected subjects. Purposive sampling should be avoided, particularly, if the population is' heterogeneous, and if they are unavoidable, the data analysis and interpetation should be done carefully.