SoylentNews is people
SoylentNews
The internet dominates our world and each one of us is leaving a larger digital footprint as more time passes. Those footprints are ripe for studying, experts say.
In a recently published paper, a group of Stanford sociology experts encourage other sociologists and social psychologists to focus on developing online research studies with the help of big data in order to advance the theories of social interaction and structure.
[...] In the new study, the researchers make a case for "online field experiments" that could be embedded within the structure of existing communities on the internet.
The researchers differentiate online field experiments from online lab experiments, which create a controlled online situation instead of using preexisting environments that have engaged participants.
"The internet is not just another mechanism for recruiting more subjects," Parigi said. "There is now space for what we call computational social sciences that lies at the intersection of sociology, psychology, computer science and other technical sciences, through which we can try to understand human behavior as it is shaped and illuminated by online platforms."
As part of this type of experiment, researchers would utilize online platforms to take advantage of big data and predictive algorithms. Recruiting and retaining participants for such field studies is therefore more challenging and time-consuming because of the need for a close partnership with the platforms.
But online field experiments allow researchers to gain an enhanced look at certain human behaviors that cannot be replicated in a laboratory environment, the researchers said.
For example, theories about how and why people trust each other can be better examined in the online environments, the researchers said, because the context of different complex social relationships is recorded. In laboratory experiments, researchers can only isolate the type of trust that occurs between strangers, which is called "thin" trust.
Is Big Data the path to respectability for the social sciences?
More information: Paolo Parigi et al. Online Field Experiments, Social Psychology Quarterly (2017). DOI: 10.1177/0190272516680842
(Score: 1, Insightful) by Anonymous Coward on Wednesday April 05 2017, @03:11AM
Big data will not fix the problem with sociology and other social sciences. If anything it could greatly exasperate it. The whole problem is the complete conflation of causation and correlation. And big data opens up way more room for bastardization there. If you give an individual access to enough information about e.g. murderers, you can find things that they all have in common that most others do not. This does not mean it's causal, or even that it would result in a higher probability of somebody else who shares these characteristics of being a murderer.
You can even see them doing as much with the small amount of data these groups have access to. I was involved in a research project that was trying to link the ratio of the length between two of your fingers to your sexuality and was eventually published in a major journal. The thing is - given enough pieces of data you can find patterns that don't actually exist. Imagine you put 7.125 billion balls into a quite large urn. But you link each of these balls to a number where you generate a number of quantifiable variables comparable in size to the number of variables you could formulate about humans - which is to say practically infinite. You are going to create links between some of these variables that in this case are clearly just complete noise, yet social scientists would suggest are indicative of causality. You start drawing out balls. And now notice random variable 1,435,256,231 (identified sexuality) correlates against random variable 3,452,357 (digit ratio). It's totally random, but now comes the "scientific method 2.0" used by social sciences:
- Observe a pattern in a sample
- Formulate a hypothesis
- Sample another group.
- Assuming the pattern is still there, claim hypothesis is verified.
What you should note in the above is that the hypothesis is in no way used. Social scientists are smart enough to make smart sounding hypothesis, but the hypothesis plays no role whatsoever. In the finger ratio to sexuality correlation, the hypothesis has to do with suggesting that prenatal hormonal levels affect the ratio (another completely arbitrary correlation of dubious worth) and so by measuring the ratio you're actually measuring prenatal hormonal levels. It's literally a correlation built upon a correlation laying claim to causation. It sounds smart so nobody really says anything, but it could just as well be 1+1=3. And I think this sort of science is really starting to drag down the entire field. You can find numerous examples of people beginning to suggest that the replication crisis present primarily in social and other correlation based sciences is indicative of failure of science as a whole. Correlation-only science needs to be separated from more principled science.