Quantifying participation biases on social media
نویسندگان
چکیده
Abstract Around seven-in-ten Americans use social media (SM) to connect and engage, making these platforms excellent sources of information understand human behavior other problems relevant sciences. While the presence a can be detected, it is unclear who or under what circumstances was generated. Despite large sample sizes SM datasets, they almost always come with significant biases, some which have been studied before. Here, we hypothesize largely unrecognized form bias on platforms, called participation , that distinct from selection bias. It defined as skew in demographics participants opt-in discussions topic, compared underlying platform. To infer participant’s demographics, propose novel generative probabilistic framework links surveys data at granularity demographic subgroups (and not individuals). Our method existing approaches elicit such individual level using their profile name, images, metadata, thus infringing upon privacy. We design statistical simulation simulate multiple diverse range topics validate model’s estimates different scenarios. Twitter case study demonstrate topic gun violence delineated by political party affiliation gender. Although Twitter’s user population leans Democratic has an equal number men women according Pew, our point control opposite direction, slightly more Republicans than Democrats, women. cautions rush digital for decision-making understanding public opinions, must account biases inherent how are produced, lest may also arrive biased inferences about public.
منابع مشابه
Quantifying the Economic and Cultural Biases of Social Media through Trending Topics
Online social media has recently irrupted as the last major venue for the propagation of news and cultural content, competing with traditional mass media and allowing citizens to access new sources of information. In this paper, we study collectively filtered news and popular content in Twitter, known as Trending Topics (TTs), to quantify the extent to which they show similar biases known for m...
متن کاملQuantifying Information Overload in Social Media and Its Impact on Social Contagions
Information overload has become an ubiquitous problem in modern society. Social media users and microbloggers receive an endless flow of information, often at a rate far higher than their cognitive abilities to process the information. In this paper, we conduct a large scale quantitative study of information overload and evaluate its impact on information dissemination in the Twitter social med...
متن کاملQuantifying the Language of Schizophrenia in Social Media
Analyzing symptoms of schizophrenia has traditionally been challenging given the low prevalence of the condition, affecting around 1% of the U.S. population. We explore potential linguistic markers of schizophrenia using the tweets1 of self-identified schizophrenia sufferers, and describe several natural language processing (NLP) methods to analyze the language of schizophrenia. We examine how ...
متن کاملQuantifying Mental Health from Social Media with Neural User Embeddings
Mental illnesses adversely affect a significant proportion of the population worldwide. However, the methods traditionally used for estimating and characterizing the prevalence of mental health conditions are time-consuming and expensive. Consequently, best-available estimates concerning the prevalence of mental health conditions are often years out of date. Automated approaches to supplement t...
متن کاملQuantifying Influence in Social Networks and News Media
Massive numbers of users of social networks share various types of information such as opinions, news, and ideas in real time. As a new form of social network, Twitter is a particularly useful information source. Studying influence can help us better understand the role of social networks. The popularity of social networks like Twitter is primarily measured by the number of followers. The numbe...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: EPJ Data Science
سال: 2023
ISSN: ['2193-1127']
DOI: https://doi.org/10.1140/epjds/s13688-023-00405-6