Whenever you are conducting a survey, there are several important factors to consider.
First, you must have a sample that is large enough to make the survey valid. So, you will need to reach as many people as possible. If you fail in this area, the survey may not be valid.
Second, you must have a sample of random people. You would not want only to survey one section of the population if you want to gauge the whole population. Hence, when you generate your survey, it must be random, so that you get a healthy cross section of the population.
Third, you will have to decide how you will get the surveys to people. In other words, what means will you use? Will you use phone calls, face to face meetings, written surveys, or something else?
Finally, you must make the survey in such a way that you do not lead people to answer in a certain way. In other words, the formatting and content of the survey must be as free from bias as possible.
If you follow these four points, your survey will have a better chance of being accurate.