A voluntary response sampling is a sampling in which people volunteer to participate in a survey. This is a form of biased sampling. It is impossible to get random sample using this sampling method.
Voluntary response sampling is heavily biased because it focuses on volunteered survey answers rather than a random sampling. What makes this biased is because the only people who volunteer are those who have a particular interest in the topic of the survey. For example, if the survey is about feelings surrounding a particular religion, only those who feel strongly about it volunteer. This means that the resulting samples have two extreme sides with few or no middle-of-the-road responses to mellow out the results. This results in a sample that is not representative of the population. The bias often skews negative, as well.
Voluntary response sampling is useful for supporting a biased stance. For instance, if a website is anti-war and puts out a poll to its users, its users are also likely to be anti-war. The results of the poll, then, would have an anti-war view point, which the website can then cite as evidence that a population supports their views. In this way, voluntary response sampling can help research that has an agenda.