Q. Can I give CrowdTangle access to my students?
A. Currently, we only allow staff, faculty, and PhD students at your university to access your CrowdTangle account, and unfortunately, this does not include students unless they’re your Research Assistant. We’re still exploring policies for the Academics and Researchers vertical, and it’s possible this will change in the future. There are many ways to keep students looped into your work without adding them to your account, such as the Chrome Extension, creating public Live Displays, and subscribing your students to Notifications.
Q. How can I provide CrowdTangle access to someone outside of my university?
A. If the person is at a different university, and is a collaborator on your specific research project, you're welcome to add them to your dashboard. If they are working on a separate project, please have them apply for access at crowdtangle.com/academics.
Q. What if my collaborator is at a government institution or NGO?
A. You may not provide CrowdTangle access to collaborators who are in government affiliated institutions, such as national research institutions. If you are working on a government-affiliated research project, you may show the results of your research to government partners, but you may not share the raw data.
If the NGO already has CrowdTangle access, you may add them to your dashboard for the specific project, but they must use their own CrowdTangle account for other projects.
Q. If I move to a different institution, can I retain CrowdTangle access?
A. Yes, but we need to move your dashboards into a new account for that new institution. Please email your CrowdTangle POC to request this move.
Q. What data does CrowdTangle track?
A. Read this article for information on the data that CrowdTangle tracks.
Q. Does CrowdTangle include any deleted content?
Q. How do I add a new page or group to the CrowdTangle database?
Q. How should I cite CrowdTangle data, or describe it in an article I’m writing?
A. This article contains information about how to cite and describe CrowdTangle data.
Q. I added a new page to CrowdTangle -- why can't I see the historical data yet?
A. It can take up to 48 hours to pull in all the historical data for a page or group that was just added to the system. Please come back to your query in 48 hours and check again.
Q. How can I access Reddit data?
A. While academics don't have access to Reddit dashboards, you are able to view and download the Reddit data that already in our system using CrowdTangle Search.
Q. Why can I see metrics on posts from the time a Page was created, but can't see Page Likes from that same time period?
A. CrowdTangle can only track Page Likes starting when the Page was added to our system. You might see Post metrics that go further back than Page Likes for this reason.
Q. How can I share my data with other researchers for validation or replication?
A. Please do not share CrowdTangle data with people outside your university research team (or NGO, if you are in a research NGO). If you want to make information publicly accessible, you're welcome to post the Page or Group IDs of the Pages and Groups that you studied, so that others can recreate your List in CrowdTangle and pull the same dataset.
Q. If an account is deleted from Facebook, what happens to its interactions on posts in CrowdTangle?
A. Once an account is deleted from Facebook, interactions from the account will be deleted as well, and any posts with which the account interacted will be updated in CrowdTangle to reflect that. Any posts made by the account will be deleted.
Q. How can I track interactions over time?
A. Intelligence will let you track the number of interactions on Posts over time. This means that we group the interactions by the date the Post was made -- NOT the date the interaction was made. To track interactions on the dates that they were made, either use the Post Time-Series data, or, view a graph by week in the List Leaderboard.
Q. What languages can I search within images text?
A. The following languages are supported for image text search:
Q. How can I see incremental change in a post’s interactions over time?
A. To see how a post’s interactions have changed over time, click the inverted carat on the upper-right-hand corner of the post, and click “download CSV”. This will show you the timesteps and corresponding change in interaction rates in an Excel CSV file. Note that we can only track this data once the page/group has been added to CrowdTangle -- we can't track time-series growth for posts made before we started tracking the account.
Q. How can I see how many times a keyword has been mentioned in the CrowdTangle database?
A. To see how many times a keyword has been mentioned, set up a Saved Search with your parameters. Then, navigate to the leaderboard in the Saved Search. This will show you how many posts each source has made with that keyword within the time period your search is set for.
Q. How often is the data on Posts refreshed?
A. We refresh the data for Facebook and Instagram Posts about once every 15 minutes. You can see the exact breakdown of timesteps in the Codebook.
Q. Is there a limit to how many Posts I can download in a CSV?
A. You can download a maximum of 10,000 Posts in a CSV from a Dashboard. If you export a CSV from the Historical Data tool (accessible via Dashboards), you can get up to 300,000 posts.
Q. Can I download the entire text of a Post in a CSV?
A. The text of the Post will truncate at about 5000 characters.
Q. Why is my Search for URLs with a query string truncating after the question mark?
A. We don't automatically track query URLs. This article explains how to ensure that query URLs will surface within Search, but unfortunately, this won't work using Link Search in Chrome Extension. You can now force query URLs to be supported in the /links API using the SearchField parameter.
Q. How can I convert a Saved Search, or a result from the new Search surface, into a List?
A. To convert any search results into a List, download a CSV of the results, and Batch Upload the Facebook IDs of the accounts as a new List in the Dashboard. If you get any errors from the Batch Upload, just manually add the URLs for those pages into the list.
Read the API cheat sheet for an overview of endpoints and API best practices.
Q. Why don’t I have access to the Posts/Search API even though I have API access enabled?
A. Posts/Search is the only API endpoint we don’t activate by default, because it’s a heavier call than all the others. In order to activate it, please fill out this form and ask for it to be activated. Once you provide us with your use case, we’ll activate it and you’ll be good to go.
Q. Can I increase my API call limit beyond 6 calls/min (Posts, Search, Leaderboard, Lists) and 2 calls/min (Links)?
A. Please fill out this form and request the increase that you need. Once the tech team approves the request and adjusts the rate limit in our system, you’ll be good to go.
Q. Why is it taking so long for the API to load data, and how can I make sure that CrowdTangle doesn’t crash for everyone while I’m using the API?
A. If you are not setting start and end date parameters, it takes our system much longer, and takes more energy, to query the database. A call should take about 10 seconds for very long queries, and with data parameters, should only take in the hundreds of milliseconds. Set date parameters to speed up your call, and to make sure CrowdTangle stays available for everyone!
Q. Can I do a weighted search with the Posts/Search endpoint?
A. No. The Search endpoint operates similarly to the Historical Data download in the UI, wherein you can’t weight or set very granular parameters. If you want to pull in data from a weighted search using the API, create the search using Saved Search in the UI, and then pull from that search by putting the search ID into the /Posts endpoint.
Q. What is the maximum number of Posts that the API will return per query?
A. All API endpoints will time out after 10,000 posts. We recommend slicing your query by time to make sure that you don't miss any Posts.
Q. Can I get time-series data for a single post in the API?
A. Yes. Just use the includeHistory parameter, set to "true."