Research data for paper: Racist way of victim blaming in the aftermath of the Grenfell Tower fire

This dataset for a forthcoming publication involves racist tweets against Grenfell Tower fire victims. We aimed to understand how tweeters used racist language to delegitimise the victims and their supporters.
Data was collected in January 2021, using the Advanced Search option within Twitter and by downloading tweets using tweepy.org, developer.twitter.com, and kaggle.com. The data was processed using Pandas Package (pandas.pydata.org).

The dataset consists of four files of tweet ID numbers of racist Twitter posts against Grenfell Tower victims, survivors and bereaved families. In total we collected 26,653 tweets that involved #Grenfell #GrenfellTower #GrenfellTowerfire hashtags in four different time periods (see below for more information), and separated the ones that used hostile and racist language against Grenfell Tower fire victims and bereaved families. Tweets were also analysed which were in themselves not using racist language, but where replies included racist language.This dataset represents 267 of the 416 tweets that were analysed, in the following documents:

1. ‘2017_Tweets_Grenfelll Public Inquiry began’ represents data that were collected between September 7 and 21 in 2017. This centred on the formal opening of the Grenfell Tower Inquiry on 14th September 2017. We collected 93 racist tweets out of 6,348 Grenfell Tower fire-related tweets from this period.

2. ‘2018_Tweets_Petition’ represents tweets from the second data collection time period that was during the time of the petition that Grenfell supporters created to demand a debate in Parliament to include survivors and bereaved families in the inquiry process. The petition began on 14th May 2018 and it ended on 30th May 2018. We collected 112 racist tweets out of 7,578 Grenfell Tower fire-related tweets between those dates.

3. 2019_Tweets_Inquiry Report’ has tweets that were collected between 23rd October and 7th November in 2019. This included the date of the first inquiry report on 30th October. We collected 47 racist tweets out of 11,718 Grenfell Tower fire-related tweets during this period.

4. ‘2020_Tweets_COVID cancelled SW’ involves the last data collection time period that happened between 7th and 31st March 2020. This included the first Silent Walk that was cancelled because of the COVID-19 outbreak. We collected 33 (13 tweets with hashtags) racist tweets out of 1,010 Grenfell Tower fire-related tweets.
The Readme file includes further details.
The files include TweetIDs as captured in January 2021. These can be rehydrated using resources such as Twarc (https://github.com/DocNow/twarc) or Hydrator (https://github.com/DocNow/hydrator) in order to retrieve the Tweets as they currently appear on Twitter. Tweets which have been deleted since data capture will not be retrieved.

本数据集用于一篇即将发表的学术论文，内容包含针对格伦费尔大厦火灾遇难者的种族主义推文。本研究旨在探析推文发布者如何使用种族主义言论，以否定遇难者及其支持者的合法性。
本数据集于2021年1月采集，通过Twitter高级搜索功能，借助tweepy.org、developer.twitter.com及kaggle.com平台下载推文，并使用Pandas库（pandas.pydata.org）完成数据处理。
本数据集包含四个文件，均为针对格伦费尔大厦遇难者、幸存者及遇难者家属的种族主义推文的ID编号（Tweet ID）。本次研究共采集到包含#Grenfell、#GrenfellTower、#GrenfellTowerfire话题标签的相关推文共计26653条，覆盖四个不同时间段（详见下文），并从中筛选出针对格伦费尔大厦火灾遇难者及遇难者家属使用敌意性、种族主义言论的推文。此外，本研究还对未直接使用种族主义语言，但回复中包含种族主义言论的推文进行了分析。本数据集对应其中267条已分析推文，详情如下：
1. '2017_Tweets_Grenfell Public Inquiry began'（原文文件名存在笔误，原拼写为Grenfelll）对应2017年9月7日至21日采集的数据，该时段聚焦于2017年9月14日格伦费尔大厦火灾调查公开听证会的正式启动。本次采集的6348条格伦费尔大厦火灾相关推文中，共筛选出93条种族主义推文。
2. '2018_Tweets_Petition'对应第二次数据采集时段，即格伦费尔支持者发起请愿活动期间。该请愿旨在要求议会就将幸存者与遇难者家属纳入调查流程一事展开辩论，发起于2018年5月14日，结束于5月30日。本次采集的7578条相关推文中，共筛选出112条种族主义推文。
3. '2019_Tweets_Inquiry Report'对应2019年10月23日至11月7日采集的数据，该时段包含首份调查报告发布日期（2019年10月30日）。本次采集的11718条相关推文中，共筛选出47条种族主义推文。
4. '2020_Tweets_COVID cancelled SW'对应最后一次数据采集时段，即2020年3月7日至31日，该时段包含因COVID-19疫情取消的首次静默游行活动。本次采集的1010条相关推文中，共筛选出33条种族主义推文（其中13条带有相关话题标签）。
本数据集的详细说明可参阅Readme文件。
所有文件包含2021年1月采集时留存的推文ID（Tweet ID），可通过Twarc（https://github.com/DocNow/twarc）或Hydrator（https://github.com/DocNow/hydrator）等工具进行推文还原，以获取当前Twitter平台上的推文原貌。自数据采集以来已被删除的推文将无法被还原。