Try to achieve 100%.Failing that means that the survey responses that you get are about a subpopulation that responds to surveys. The response from this population may be different from the entire population. Depending on the goals, you might need to look into assessing if your responses are not an accurate representation of the total population. There can be other issues like some portion of the surveyed population finds your questions (or at least one question) offensive.A different approach would be to ask if there were enough favorable responses to enable your product to be successful despite other factors. At least in the political arena, some "surveys" seem to be more advertisement than data gathering tools.
What is special about the third condition that so many respondents drop out?A low response rate is a major problem in survey design. Typically one then tries to encourage respondents to respond and records the effort needed to get a response. The fast responders are then compared to the encouraged responders to identify potential differences. If differences exist then there is a bias in the survey and that needs to be accounted for in trying to interpret the results. You need to keep track of this as it is a key element in survey reporting.
There is stratified random sampling where you identify specific groups and then randomly sample within each group.There is purposeful sampling. Depending on the project goals you might need to worry about bias, but you should worry about this anyway.You could increase sample size until minor groups have a sufficient sample size. You have more analysis options with this approach. There is no requirement that you use all of the data collected, though how you selectively analyze the data can also lead to bias. One solution is to be open about how you did the analysis and make the raw data available. If someone disagrees with you they can take your raw data and do the analysis in the way they think is right. If they get a different conclusion then you have lively debate.
Rocky Horror is the movie that comes to mind. H.P. Lovecraft is an author that comes to mind, though there are many more if I stop to think about it.
Already have an account? Login
Enter your username or e-mail address. We'll send you an e-mail with instructions to reset your password.
Sorry, we're still checking this file's contents to make sure it's safe to download. Please try again in a few minutes.
Sorry, our virus scanner detected that this file isn't safe to download.