Twitter "data library"
Idea: The content of the Twitter "data library” to provide data about topics in high demand among researchers and enables to assess a variety of subtopics.
Example: I’m studying about tweets of an election to understand the population and presidential candidate sentiment on green card and public assistance. My first stab API query would be “election”, “green card”, and “welfare”. The larger topic here would be the US Presidential Election. Instead of posting the specific API request, the Twitter "data library” (hypothetically) would already have a substantially sized dataset of all the tweets related to the US Presidential Election.
Even if the Twitter "data library” did not satisfy the exact inquiry, it is still helpful to have similar content. The researcher can do an exploratory analysis of the available data, evaluate if the content is likely available or worth pursuing, and refine the research question.
Example: Initial research question asks “Do Social Media Influencers encourage recycling?”. The “Twitter data library” has a similar dataset on replies/engagement tweets of Social Influencers that post about climate change initiatives. The researcher evaluates the available dataset and finds that most replies don’t engage with the subject and only compliment the looks of the Social Influencer. With this preliminary insight, the researcher can improve the question.
Last, researchers also think “with this available dataset, what are the questions that I can solve?”. Given limited availability and quality of data, researchers would often use the same dataset.