This is the Final Summary Overview of a study that assessed the utility of using data from Twitter, an online platform for sharing thoughts and feelings, in assessing the social climate that contributes to hate crimes in the United States.
The stated goal of the study was to address the lack of reliable knowledge about the nature and prevalence of hate crime in the United States by identifying and analyzing online hate speech and the link between online hate speech and data on hate crimes. The study relied on four types of data: 1) recorded hate-crime data; 2) social media data; 3) census data; and 4) data on hate-crime risk factors. The study used an ecological framework and Poisson regression models in determining the link between hate speech online and actual hate crimes. Risk terrain modeling was used to further assess places at higher risk of hate crimes. The study found weak correlations between tweets that contained hateful language about persons with specific physical characteristics or group associations and the prevalence and characteristics of actual hate crimes against persons or groups; however, the study does not discount the use of online content in measuring the climate for hate crime. It argues that with additional methodological development, online data sources may provide reliable and useful information that can augment traditional police and victim survey sources in assessing and facilitating the development of countermeasures against hate crime. 8 tables and 28 references
810 Seventh Street NW, Washington, DC 20531, United States