I have witnessed that most of the recent quantitative social media research studies which published in the top world ranked journals were employing statistical technique as simple as regression analysis and as advanced as SEM to a greater extent. However, almost every research is using convenient sampling. But still, most of the academia being as panel members of research viva voce are strictly expecting the sampling method to be random to perform such analysis techniques. I want expert scholarly insights on above issue.
The non-probability technique of convenient sampling may be applicable in some circumstances, for example, to get a quick glimpse of an opinion poll using social media. For example, we have seen polls posted on Twitter, but this would be restricted to only those respondents who follow that particular researcher or who he/she follows.
However, where the sample has to be truly representative and the results from the sample are to be used to generalize to the population from which the sample was selected, then probability sampling is merited.
Although probabilistic sampling is always desirable in order to generalize the findings of any study to the population under study, the contextual circumstances should be given consideration when examining any research work. For instance, in a study of online shoppers, given the sophistication involved in shopping online, a random sample might end up in a large number of respondents not relevant to the study. In such cases, convenience sampling turns out to be the saviour.
I hope I have addressed the question.
Regards
Non-probability sampling is often required when the population of interest is hidden or difficult to locate. Ex-offenders, gay adoptive parents, illegal substance users, and undocumented immigrants are examples of large population groups from which a random selection is simply not possible except in a very limited sense. For example, illegal substance users being treated might be randomly selected, but then the results could only be generalized to the population of substance users being treated at that facility.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques