This is the English translation of my research paper published in 2018, “Intanetto riyo wa hitobito no haigai-ishiki wo takameru ka” (「インターネット利用は人びとの排外意識を高めるか」), Soshioroji, 192, pp.3-20.

Does use of the Internet make people more racist?
A causal analysis based on a synchronous effects model using the instrumental variable method

1. The problem and its Background
2. Racism and the Internet: Literature review
3. Survey Outline and Measures of Racist/Anti-Racist Attitudes
4. Method of Analysis: Synchronous Effects Model
5. Variables Used in the Analysis
6. Results
7. Conclusions and Discussions

» download the full article (PDF)

Abstract

In Japan, the rapid and widespread dissemination of online xenophobic discourse has been increasingly apparent since around 2000; among the serious resulting problems are the activities of organized hate groups that have emerged out of the online right such as Zaitokukai. Previous studies have confirmed the Internet’s role as a catalyst that connects racists to one another and promotes their activities. However, it remains an open question whether Internet use exacerbates racist attitudes not only among such activist groups but also among the general populace. In addition to the near absence of studies analyzing this question based on social survey data, those that exist merely establish a positive correlation between Internet use and racist attitudes, leaving the direction of causality unclear.

Therefore, by means of a synchronous effects model using an instrumental variable method based on data from a web-based questionnaire survey conducted in November 2017, I analyzed the direction of causality regarding (i) whether Internet use exacerbated racist attitudes and (ii) whether Internet use was more prevalent among those exhibiting stronger racist attitudes. As a result, the directional path for (i) showed significant coefficient values, confirming a positive causal effect; meanwhile, the directional path for (ii) was not significant, meaning that causal effects were not supported.

Interestingly, it was also found that Internet use simultaneously had the causal effect of heightening anti-racist attitudes. From the results of additional analyses, I suggest that this simultaneous and reciprocal bi-directional effect may be engendered by selective exposure to information and peers—in other words, those who are predisposed to racist attitudes will be more strongly influenced by more frequent contact with racist information and peers, while the opposite will be true in the case of those with anti-racist views.