An anonymous survey is a survey that does not reveal enough information about respondents for them to be identifiable through routine statistical analysis of the information collected. Survey designers cannot figure out who took the survey on the basis of the information they ask for, and participants cannot identify each other on the basis of their answers. This differentiates from a confidential survey, where such information is available, but the designers agree to maintain confidentiality for the safety of participants. In open or public surveys, participant information is readily available.
Survey designers must exercise care in the creation of a survey, and need to determine whether it will be anonymous, public, or confidential at the start. When a survey is used for academic research, an institutional review board will typically need to examine the methodology and structure to determine if the interests of its human subjects are adequately protected. For example, it can mandate a disclosure at the start of the survey that contains information about the designers and the confidentiality of the information.
Anonymous surveys can have benefits and flaws. Complete anonymity may encourage respondents to be more frank, as they have fewer concerns about being outed or exposed to risks. The anonymity can also, however, make it easier to lie or stuff survey results. In the data analysis process, the survey designers need to be alert for signs of false responses so they can weed those responses out and leave the valid ones behind.
Sampling is also difficult with an anonymous survey. In a confidential survey, researchers can use a variety of techniques to get an evenly distributed and reliable sample of information. This can increase the validity of their results. With anonymous surveys, there is less control over the sample population. The researchers can make the anonymous survey available in as many places as possible and collect some information about their participants to determine the validity of the sample, but they may get imperfect results.
Participants in an anonymous survey with an interest in how the information is used can keep track of the researchers listed on the survey. When they publish papers based on the survey results, it can be possible to read them to see what kind of conclusions were reached as a result of the survey, and how the survey will be applied to research in the future. Clipping services and online people tracking programs can do this automatically for anonymous survey participants who do not have the time to keep track of researchers themselves.