Most of the information on Wikipedia comes from a tiny proportion of users. Now social scientists are collecting data in a similar way, allowing participants to design surveys as they contribute.
Gathering data about human preferences and activities is the bread-and-butter of much research in the social sciences. But just how best to gather this data has long been the subject of fierce debate.
Social scientists essentially have two choices. On the one hand, there are public opinion surveys based on a set of multiple choice questions, a so-called closed approach. On the other, there are open approaches in the form of free ranging interviews in which respondents are free to speak their mind. There are clearly important advantages and disadvantages of each method.
Today, Matthew Salganik at Princeton University in New Jersey and Karen Levy at New York University outline an entirely new way of gathering data inspired by a new generation of information aggregation systems such as Wikipedia. Just as Wikipedia evolves over time based on contributions from participants, we envision an evolving survey driven by contributions from respondents, they say.
They say the new approach can yield insights that would be difficult to obtain with other methods. But it also presents challenges for social scientists, particularly when it comes to analyzing the data collected in this way.
Projects like Wikipedia are the result of user-generated content on a massive scale. The question that Salganik and Levy ask is whether surveys could also be constructed by respondents themselves, at least in part.
To find out, these guys have developed a new type of data collection mechanism that they call a wiki survey. This starts with a set of seed questions but allows respondents to add their own questions as the survey involves.
This wiki survey takes a particular form in which respondents are asked to choose between two options: do they prefer Item A or Item B, for example. But crucially, they can also add a new item that will be presented to future participants. So as time goes on, the number of items to choose from increases as respondents suggest their own ideas.
This kind of pairwise survey has a number of advantages. Salganik and Levy point out that this format allows participants to respond to as many choices as they wish. They call this property greediness.
Read the original here:
Inspired by Wikipedia, Social Scientists Create a Revolution in Online Surveys