I am doing a study on consumer engagement in Facebook brand pages. I am collecting the data using an online survey that was posted on a certain brand page on Facebook. I posted the link of the online survey on the brand page where i was able to get 591 responses from consumers who are members in that page.
My question is : am I using simple random sampling? Some researchers told me that i am using that sampling technique because every member in the population of the brand community had an equal chance to appear in the sample?. I didnt choose the participants who answered but i sent an invitation asking consumers to answer my survey
I am not sure about the sampling technique that i have used. Please advice me.
Technically speaking, it probably is a convenience sample, but that label has a negative stereotype. So, for any write up that you do, I would simply describe what you did, rather than trying to label it.
there will be an issue with this , because the participants are self selecting themselves when the choose to answer your survey, and may be a certain section of the population (not random), so any inference you make will be valid for that sub-population who chooses to answer surveys (for whatever reason) rather than the full population
No, you are not using simple random sampling. That would require defining a "universe" and then drawing potential participants from the universe in a random fashion.As Vikas Ramachandra points out, you have a self-selected sample, which is no way random.
If your underlying question is whether you can use tests of statistical significance, this is common with non-random samples. But you need to point out the non-random nature of you sample as a limitation in your research.
In line with the above 2 answers, the way that you have sought a sample, has allowed self-selection; which undermines the criterion for a random sample from a defined sample-frame that each member of the frame has an equal chance to be chosen for the random sample.
Thanks Vikas, David and Philips. Do you think my sample is a convenience sample?
Technically speaking, it probably is a convenience sample, but that label has a negative stereotype. So, for any write up that you do, I would simply describe what you did, rather than trying to label it.
It is a case of Non-probability sampling and may be treated as convenience sampling as discussed by other respondents as well.
Hazem -
Perhaps the following, by Mike Brick, might be of interest to you:
"Beyond traditional survey taking: adapting to a changing world
Explorations in Non-Probability Sampling Using the Web," J. Michael Brick
Proceedings of Statistics Canada Symposium 2014
http://www.statcan.gc.ca/sites/default/files/media/14252-eng.pdf
His ResearchGate contributions can be found through his profile:
https://www.researchgate.net/profile/J_Brick
Cheers - Jim
As you said "I didnt choose the participants who answered but i sent an invitation asking consumers to answer my survey" this negate the propability sampling. However, if you chose them through a random table or numbers from a hat, then it is simple random technique. For those who will not answer, you can do sampling following the sampe process until you reach your sample size.
Above we see "For those who will not answer, you can do sampling following the sample process until you reach your sample size." - But actually that would introduce bias. Your population then is not the one you want, but the population of those who are willing to respond, which can be quite different.
It may be helpful to research the term "response propensity" groups.
Cheers.
Any idea about the size of the targeted population from which the self selected sample will be investigated? In addition to the self-selection bias, the response rate and the brand social media biases are also possible limitations to the results of this study.
In the sense that---I assume---you are going to use the sample of 591 that you got from your online survey, you do have a convenience sample; but as David commented above, this label can attract a slightly negative stereotype; but this does not mean that such a sample cannot be useful for research in some respects; albeit, in project write-up, you'd have to be transparent in describing how you got your sample, even if you do not directly label it; and also be aware of the limitations of using such a sample.
Simple random sampling technique gives each element an equal and independent chance of being selected. . where some elements are purposely excluded from the sample ,the resulting sample is not a random sample. You have to include all elements in the sample frame to draw a random sample. First of all enumerate all elements in the population, prepare a list of all elements and then draw sample .
To ensure that you have using random sampling which is a probability sampling then researcher must has a sampling frame. Your question implies that your contact list or friend list or members list is not qualify for a sampling frame. If your target population is only member of brand' page and you have sent invitation to all then no sample is extracted therefore it is more census rather sampling.
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology