Stateology: State-Level Interactive Charting of Language, Feelings, and Values

TL;DR

This paper introduces Stateology, a web-based tool that visualizes state-level demographic, linguistic, and psychological traits derived from analyzing nearly 200,000 U.S. bloggers' language data from over two billion words.

Contribution

It presents a novel interactive platform for mapping and exploring psychological and demographic dimensions based on social media language use at the state level.

Findings

Geographic distribution of language and feelings across U.S. states

Demographic and psychological insights from social media data

A scalable web tool for visualizing large-scale linguistic data

Abstract

People's personality and motivations are manifest in their everyday language usage. With the emergence of social media, ample examples of such usage are procurable. In this paper, we aim to analyze the vocabulary used by close to 200,000 Blogger users in the U.S. with the purpose of geographically portraying various demographic, linguistic, and psychological dimensions at the state level. We give a description of a web-based tool for viewing maps that depict various characteristics of the social media users as derived from this large blog dataset of over two billion words.