Dear All, I am doing research on Homosexuality and have decided to explore their familial and educational status. Since the issue is sensitive particularly in Nepalese context, so i have decided to trace the respondents based on the referral networks.I am kind of confused regarding the no. of respondents that would be ideal for conducting such research. Thanks
The number depends on what data you want to collect and what you want to do with it. For some research questions, one person is enough. I’ve sometimes had problems with samples of several thousand, because it was not enough.
If you are suggesting that you will use snowball sampling to identify potential homosexual respondents in Nepal, you are about to commit a very serious breach of research ethics. you are putting referrers and identified respondents at risk for public exposure, humilitation, and punative acts. You research problem is not one of numbers required, it is to select a method of subject recruitment that provides basic ethical human subjects research protections. Please seek help from a local university institutional review board, ethicist, or researcher who has already published on this topic. I know that they exist in Nepal. AIDS has been researched in Nepal. Perhaps one of those authors can help you select an appropriate method.
Joseph – that’s an interesting angle of response that I didn’t think of. I’d be interested in finding out more about these issues – is there any published work you can point us to that discusses ethical issues in snowball sampling?