Scientists Accurately Predict Individuals’ Income Solely off Social Media Posts

Phone Money

A large-scale study using 2.6 million Nextdoor posts reveals that users’ online activities can predict socioeconomic status, with wealthier neighborhoods showing more positive but crime-sensitive posts. The study also highlights regional differences in crime discussions, with the wealthiest areas in both the US and the UK talking about crime more, even with lower actual crime rates, and discussions in the US more skewed towards weapons and violent crimes.

A study conducted by scientists at the Queen Mary University of London involved the analysis of 2.6 million posts on the well-known social networking site, Nextdoor. Surprisingly, they were able to predict a user’s income accurately, using only the content of their posts. The researchers noted prominent disparities in the type of content posted by residents of more affluent neighborhoods compared to those in less affluent areas. This discovery suggests that our overall online activities, not just on Nextdoor, might provide insights into our socioeconomic standing, enabling the potential for user profiling.

Such knowledge of users’ income could pave the way for social media platforms to suggest content that aligns with a user’s income level. Additionally, advertisers and e-commerce businesses could take advantage of this economic profiling to target consumers more effectively, promoting specific products at varying prices corresponding to the user’s income level.

The findings of the study also show that people who live in wealthier neighborhoods were more likely to share positive posts but would discuss crime more, even if the actual crime rates are lower than in poorer neighborhoods.

Dr. Ignacio Castro, lead researcher and Lecturer in Data Analytics at Queen Mary University of London, said: “Our study shows that the text posted by users in poor neighborhoods is distinguishable from the text generated in wealthier neighborhoods. Online users’ content reveals socioeconomic factors: in wealthier neighborhoods, there is more crime-sensitive posting activity, but overall, more positive sentiment in the posts.”

This is the first large-scale study of Nextdoor, published on 2 June in the Proceedings of the International AAAI Conference on Web and Social Media, that shows how income levels and income inequality within neighborhoods manifest online. Researchers collected and analyzed 2.6 million posts from 64,283 neighborhoods in the United States and 3,325 neighborhoods in the United Kingdom, shared on Nextdoor between November 2020 and September 2021. With 10 million users, the platform allows verified residents to share posts on forums that are dedicated to their neighborhoods. The study results are consistent in both countries.

Residents who live in richer neighborhoods are more concerned about crime with the 20% richest neighborhoods discussing crime approximately 1.5 times more compared to the poorest neighborhoods. This happens even though crime levels are 1.3 times higher in those poorer neighborhoods. People who live in wealthy neighborhoods with less inequality discuss crime more than anyone else.

Regarding the type of crimes talked about, non-violent crimes are discussed more than violent crimes. Most user content trends are very similar between the US and the UK, with one notable distinction when it comes to weapons and violent crimes, which are discussed more in the US than in the UK for the richer neighborhoods. This is not the case for middle-income neighborhoods, as UK residents tend to post about this type of crime more than their US counterparts.

Reference: “Lady and the Tramp Nextdoor: Online Manifestations of Real-World Inequalities in the Nextdoor Social Network” by Waleed Iqbal, Vahid Ghafouri, Gareth Tyson, Guillermo Suarez-Tangil and Ignacio Castro, 2 June 2023, Proceedings of the Seventeenth International AAAI Conference on Web and Social Media.

Be the first to comment on "Scientists Accurately Predict Individuals’ Income Solely off Social Media Posts"

Leave a comment

Email address is optional. If provided, your email will not be published or shared.