Using social media for large-scale studies of gender differences

Social networks capture data about most aspects of the daily lives of millions of people around the world. The analysis of this rich and ready-available source of information can help us better understand the complex dynamics of society. In a recent article published in EPJ Data Science the authors propose the use of location-based social networks to study the activity patterns of different gender groups.

(Photo from Pixabay, CC0 public domain)

Guest post by Willi Müller and Thiago H. Silva

Gender differences have a subjective nature and may vary greatly across cultures, making them challenging to explain. Indeed, over the past decades, this topic has received a lot of attention by researchers, but there is still a long way to reach a consensus on the subject.

Traditional ways to study differences between gender groups depend on surveys, which are costly and do not scale up. Moreover, data produced under such conditions are commonly released after long time intervals (e.g., several years). Therefore, the studies cannot quickly capture changes in the dynamics of societies. Besides, the results from cross-regional gender differences studies are usually available only for large geographic regions, often countries. Thus, even though survey-based studies could be carried out in arbitrarily small regions, such as a city, a neighborhood or even a particular venue (e.g., a university or a mall), information about gender differences at such fine spatial granularities is not easily available.

We present another way to obtain and explore similar data that could help the study of global gender differences. To map individual preferences, we propose using publicly available data from location-based social networks (LBSNs). This is interesting because when specific users of LBSN check into a specific location they express their preference for this type of place. LBSNs are accessible almost everywhere by anyone and thus allow data collection from the entire world at a much lower cost compared to traditional surveys.

We propose a new methodology to quantify the differences between male and female preferences for venues in different regions at different spatial granularities, around the world, based on LBSNs. The aggregation of such differences over multiple venues could then be used, for example, in the construction of an indicator of gender differences in a given region. We illustrate the use of our methodology by extracting user preferences for venues located in different urban regions around the world from check-in data collected from Foursquare. We have found that:

  • Gender and venue preferences may not be independent in specific regions. The level of geographic detail we analyzed ranges from countries to cities, neighborhoods, and even single venues;
  • Our approach might capture some essential aspects of gender differences. Comparing our results with an official gender difference index, we found evidence that motivate the study of new approaches using LBSN data jointly with other datasets in future developments of gender differences indices.

Our methodology could be a promising tool to support large-scale gender preferences for venues studies that require less human effort and time, compared with traditional methods, and can quickly react to changes in the real world because it relies on LBSNs data. With this method it is possible to analyze cultural gender preferences for venues, opening up opportunities for different studies and applications in several areas.

Willi Müller is currently a graduate student at the Hasso-Plattner-Institute in Potsdam, Germany. He was visiting for one year the Federal University of Minas Gerais in 2014, where he started to collaborate with the other authors of this work. Before that, he received a B.Sc. in IT-Systems-Engineering in 2013 from the Hasso-Plattner-Institute.

Thiago H. Silva is a Professor at the Federal University of Technology, Paraná, Brazil. He obtained an M.Sc. (2009) and a Ph.D. (2014) in Computer Science from the Federal University of Minas Gerais. Thiago has strong experience in the industry and academia in the areas of urban and social computing. More info at:

View the latest posts on the SpringerOpen blog homepage