Non probability sample mehtod does not involve random selection and sampling like probably sampling. So we can say that it is not necessary to respect the population. Unlike probability sampling, it does not depend on rationale of probability theory. With probabilistic technique we can have a close look on population. With non-probability technique it is not obvious to represent the population well and how accurate we are. In research methodology sampling research method is better than nonprobabilistic, which are more accurate.
Sometimes in applied social research it is not practical apply random sampling. Following are the different types of non probability sampling.
In this case sample is not selected by the researcher but the member of the universe offers themselves for selection. Suppose an inquiry is to be made by the people about an issue, an advertisement is published in newspaper. The people come volunteer and particulate in the study and acts as subject of the study. They attend the meeting on their own will and present themselves for selection of sample.
It is the type of non-probability sampling in which data is collected from the specified number of individuals. The group of these individuals is called quota. That quota is specified on the basis of age, sex, education etc. called quota-control factors. For example, quote of male and female, old and young, upper and lower, rural and urban, educated & uneducated etc.
Accidental or Incidental Sampling
Accidental or incidental is that type of sampling in which a researcher pick up data or information’s from those who fall into hand or present at the time of research. It continues the process till the completion of the sample size. It is accidental because it is selected accidentally from all type of people comes to face like, teacher, students, house wife, tailors, workers, etc.
Selecting convenient population units is called convenience sampling. It refers to that part of population being selected on convenience. For example sampling of automobiles registrations and telephone directories but this method is not representative to the whole universe because every person does not possess automobile or telephone.
Sequential sampling in which the selection of sampling size is not fixed in an advance but a researcher select elements until a decision reached and he stop sampling. This method was adopted during 2nd World War for rapid decisions.
Here we have a purpose in mind with some predefined groups. Sometimes you may have noticed some people in the market and streets with a clipboard and asking questions from selective people with a purposive sample in their mind. We might have looking for Asian male between 25 to 30 years, if they notice any one they try to engage them in the process.
Extensive sampling is less or more resemble to census method in which the entire universe is studied and no unit is left out expect the irrelevant items. Suppose if we study the opinion of Pakistan educated person about the future family system, we dropped out the non-Pakistani and uneducated from out sampling and will interview the rest of individuals.
Replicated sampling is a sort of non-probability sampling where two independent samples are taken using the same sample procedure by different investigators. It means that the sample is repeated or relocated by one research from that of the other.