The question of the unreliability of the information resulting from the analysis of our data on social networks is reversed by two American university professors.
Statistics, borrowing methods even from epidemiology, have developed methods to correct some mistakes, but they are not all corrected, Derek Ruths and Jürgen Pfeffer from universities McGill and Carnegie Mellon respectively in their article in the scientific review Science.
Real data of people who participate in a number of online social networks, available to the public, are currently being used by researchers in various studies. In fact, many times the data that is now generously available on social networks is used to extract information that will play a critical role in a decision, even at government level or for investment.
However, information based on this data should be treated with caution, especially if the researcher has not taken into account the following, said IT professors:
1. What is the sample profile in the social network chosen? For example, if the survey was done with Pinterest data, it should be taken into account that the average user profile in this network is 25 to 34 years, as otherwise the population projection will produce false results.
2. Data from social networks exploited by researchers is - usually - publicly available. However, the researcher may not know the whole of the user's data or, especially, when and how they filter the network.
3. Designing any social network can affect user behavior. For example, on Facebook, a study of what "dislikes" is more difficult to do, in the absence of a Dislike, as opposed to a study based on the actual as such, Like.
4. The study may include data from "robots" or automated procedures to disperse infamous information (bot or spam), "polluting" research with unreliable data.
[wdca_ad id = ”78638 ″]
In their study entitled "Social Media for Large Studies of Behavior" (Ruths and Pfeffer, Science, Nov. 28, 2014), academics note that science has found ways to address such problems, but should first be recognized as such. .
The issue of the reliability of the so-called "big data" - large data sets that are collected, stripped of their identity and subjected to processing - is the subject of an ongoing debate today.
In October of 2013, Kate Crawford, a researcher team at Microsoft Research and a visiting professor at the MIT Center for Civic Media and the University of Australia, UNSW, called for laws to protect the right of the user to refute what is accused of by the opaque analysis of big data. Crawford has even explained how this analysis can affect us, reminding us of the growing demands of eminent individuals on the Internet to get our data from the Web.
In July, the revelation that Facebook was playing with the feelings of its members in a psychological experiment was of particular concern. Despite the general outcry, Facebook Inc. did not prohibit their conduct, but changed the rules to be made public.