Hi,

I am planning to do a web-based survey with a group of people that are all internet users, and I want the results to be representative of this group.

The reason why I am doing a web-based survey is because I expect a higher response rate than a paper-based survey.

However, I don't have a sampling frame (i.e. a list of all email addresses of these groups). I was hoping to distribute the link to the survey via snowball sampling through contacts, adverts in magazines, on social media etc etc

I was wondering whether this would be OK to do and whether this would have an impact on the validity of the results, especially in relation to the representatives of the results?

social media would be a good tool to use to find people, u can at least verify the sample

It depends entirely on what you're trying to achieve, naturally. Web ethnography is a fairly well established method now, I know somebody who did this at Southampton University.