websites in the white supremacist world that have played a role in encouraging these attacks( see“ White Supremacists Encourage the Online Harassment of Jewish Journalists” section).
METHODOLOGY
This report covers the time period of August 2015 through July 2016.
To capture the vast sweep of anti-Semitic Tweets directed at journalists, ADL utilized the latest in“ big data” techniques. There were four phases to the report.
Phase one: ADL interviewed journalists impacted by the anti-Semitic harassment and they provided critical background information and described their experiences as targets of harassment on Twitter. They also described the effect the attacks had on their work and personal sense of safety.
Phase two: ADL conducted a search of tweets using a broad set keywords( and keyword combinations) designed by ADL to capture anti-Semitic language. These keywords did not include any terms associated directly with the 2016 presidential campaign. This yielded 2.6 million results.
Phase three: We focused our search on tweets received by a list of 50,000 journalists and compared those with the 2.6 million results. tweets to journalists
based on keywords correlating with anti-Semitism
19,253 anti-semitic tweets
at US journalists from a pool of 50,000 journalist Twitter handles
45,000,000
Number of estimated impressions generated
Phase four: We manually reviewed each of these tweets and narrowed the results to 19,253 overtly anti-Semitic tweets, which we found were directed at 800 journalists.
Note 1: One can never include all of the words that might be used in an anti-Semitic attack, and you can’ t predict the ways in which anti- Semites will create“ codes” to avoid censure and potential exclusion by social media platforms.( In October 2016, for example, after this classes of tweets to US journalists
60 % 29 % 11
Replies Re-Tweets Regular Tweets
analysis was complete, white supremacists attempted to avoid tech-based approaches to isolate online harassment. To do so they assigned tech-oriented code words to their favorite slurs, referring to“ kikes” as“ Skypes,” among many others).
Note 2: It is impossible to capture all of the anti-Semitic tweets or identify all of the anti-Semitic Twitter users, and because 21 percent of the accounts responsible for tweets containing anti-Semitic language have been deleted( either by Twitter or by the users), there is reason to conclude that the numbers in this report – especially the number of anti-Semitic Tweets received by individual journalists – are conservative.
WHY TWITTER?
This report is focused on Twitter because it is the primary social media platform used to perpetrate these attacks on journalist, according to the journalists themselves
4