id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
work_sbe5dfwq65fplp7p7p74lgkjmy	Sepideh Modrek	The Spark of the #MeTooMovement: Text Analysis of the Early Twitter Conversation (Preprint)	2019	2	.pdf	application/pdf	534	33	59	The Twitter Premium API has a maximum retrieval limit of 500 tweets per data request and each This process led to the loss of some request with non-overlap times to fill in missing tweets. to both some duplicate Tweets and brief periods with missing data. After data cleaning we had 11,935 non-duplicates tweets but based on the Twitter API counts the number of non-deleted tweets at the time of query. available via data endpoint due to user compliance actions." We queried the counts before Supplemental Methods 2: Description of the Demographics Pro Algorithm Demographics Pro then provided us with a distribution of gender, age and race/ethnicity for the handles we shared and similar data for a random sample of 20,000 Twitter users in the US as a comparison. Demographics Pro uses a series of proprietary machine learning algorithms to estimate or infer likely 	./cache/work_sbe5dfwq65fplp7p7p74lgkjmy.pdf	./txt/work_sbe5dfwq65fplp7p7p74lgkjmy.txt