Online surveys are increasingly common due to the myriad benefits they offer over traditional survey methods. One of the habitual inconveniences of Web surveys is that they are not probabilistic and that it is not possible to measure sampling error. This paper presents a probabilistic Web methodology to carry out a social study in a population distributed in education centers that have “multimedia” classrooms, with computers and an Internet connection. The sampling design is probabilistic and easy and cheap to implement. As an example of the application of the said methodology, we utilize a 2013 survey of 11,276 high school students studying in the district of the University of Valladolid distributed in four provinces of Castilla y León (Spain).