Text As Data.
I ended up scraping the idea of using the NYT API for the Ukraine research question analyzing how coverage of Neo-Nazis in the country has changed throughout the years. I wasn’t really sure on how to analyze the research question in a meaningful way and had too many issues dealing with the API.
Instead I will be conducting a sentiment analysis on the 2019 political crisis that happened in Bolivia, where the incumbent was ousted after an election audit by the Organization of American States.
I used the package rtweet to collect tweets from November 2019 to October 2020. After finally getting approved for Academic access, I ended up going over the request limit. I thought there was only a max on the amount of tweets you can collect but came to find out there’s also a limit of request calls to the API. What sent me over the limit was my issue with retrieving tweets from two lists. I wanted the search query to include two of the keywords from this list: bolivia, coup, boliviacoup, democracy, oas, evo, anez, morales, jeanineanez, anez, evomorales. To get this code right, I was only collecting 10-100 tweets to see if it executed properly. Unfortunately by the time I think I got my search string correct, I was hit with the error that I’ve hit the limit.
Correct(?) code below: