Hello there,
In my research, can I use two sampling techniques? is it possible for me to use Stratified random sampling as well as Snowball sampling?
Is it possible?
Thanks in advance.
Yes, it would be possible, but snowball sampling does not provide probabilistic estimation.
Thank you Mr Mariano and Mr. Hisham for your views.
@ mr.Mariano: I agree that snowball is a non-probabilistic sampling, but it will help me to identify more consumers for my research which I on my own may not find that easily. Is it ok to use two sampling techniques, one being probabilistic and the other being non-probabilistic?
The following papers may be helpful to your topic:
- Biernacki, P. and Waldorf, D. (1981) Snowball Sampling: Problems and Techniques of Chain Referral Sampling, Sociological Methods and Research, 10, 2, pp. 141-163.
- Onwuegbuzie, A. J. and Collins , K. M. T. (2007) A Typology of Mixed Methods Sampling Designs in Social Science Research, The Qualitative Report 12, 2, pp. 281-316
- Teddlie, C. and Yu, F. (2007) Mixed Methods Sampling: A Typology With Examples, Journal of Mixed Methods Research, 1, 1, pp. 77-100.
Hi Prof.
I developed interest in your article titled ESTIMATION OF POPULATION VARIANCE USING KNOWN COEFFICIENT OF VARIATION OF AN AUXILIARY VARIABLE IN SAMPLE SURVEYS. Very goog one.
Can I get a similar research with UNKNOWN CORRELATION COEFFICIENT and/or UNKWOWN COEFFICIENT OF VARIATION?
Expect to hear from you Prof.
Best wishes.
Peter from Nigeria
The nature of your data/population determines which method to use. Two Phase sampling is necessary if your data has auxiliary attributes which must be highly corrected with the study variable. Stratified sampling is necessary if the population can be sub/divided into homogenous subgroups. However, two phas sampling is expexted to perform better than stratified random sampling because of the presence of auxiliary attributes.
I feel concerned if your conbine probabilistic sampling with non probabilistic sampling. It would be difficult to estimate for the variance and/or bias in the non probabilistic sampling.
I strongly recommmend using only probabilistic sampling and estimation methods.
Yes, and it should be based on each aim of the research and should be appropriate for the situation of the problem.
- Badges
- Science topic
- Similar topics
- Microbiology
- Scientific Research