Toolkit for Digital Methods
Page tree
Skip to end of metadata
Go to start of metadata

2.1. Twitter

Twitter is a micro-blogging service and one of the most popular social media services in the world. The user base, however, is often strongly inclined to societal elites, journalists and politicians, which is important to keep in mind when studying Twitter (see e.g. Vainikka & Huhtamäki 2015 for Finland, Blank 2017 for the US).

Here we introduce some tools that can be used for gathering information from Twitter. The tools have been divided into easy-to-use-tools and tools that require coding skills. There are also tools that are subject to a fee. In any case, querying for Twitter data usually requires some advance planning regarding the strategy for filtering the data,, i.e. searching with a keyword, hashtag or perhaps a user name. A good way to test the results is to experiment with the Twitter search on a browser.

While Twitter is most open of all platforms, the options to collect historical Twitter data are limited. Both the search view in a browser as well has the Twitter API start "forgetting" tweets after approximately one week of their publication. The logic by with Twitter chooses what to show in a search or in response in an API call is a mystery. Therefore, historical data sets obtained this way are always a sample of the actual set of tweets published. This doesn't mean they couldn't be used for research, but it has to be done with considerations—for example when studying Finnish language tweets, the frequency of tweeting is generally not so high as globally, which means search results seem to be more comprehensive. A detailed explanation of the limitations with Twitter data and API's offered by prof. Cornelius Puschmann here.

Further, in the wake of the Cambridge Analytica Facebook scandal back in 2018, Twitter has added restrictions to its API. Each user who wishes to have access to the API must apply for developer rights and explain the purpose of using the API, and agree to comply with its terms of service for developers. Consequently, it has become more difficult to get access to Twitter's API.

In January 2021, Twitter announced they will give academics "[f]ree access to the full history of public conversation via the full-archive search endpoint, which was previously limited to paid premium or enterprise customers". To receive an api key, the user needs to apply for access and using the endpoint requires some coding skills or TAGS sheet or alike (see below).

However, not all research approaches require the API. Consider also in what ways you wish to analyze the data. Data from the API will be obtained in json format, which can be converted to csv or excel. In many cases analyzing the textual content from a reasonable number of tweets can also be done by querying the data manually with the Twitter Advanced Search and exporting it from the browser by printing, saving or copying (for Chrome, check the GoFullPage extension). Also, if you are interested in for example images, videos, gifs or profile pictures, they are a bit more difficult to analyze from the json/csv.

  • Here is a blog post about Twitter's Developer Policies from the point of view of research.
  • An overview blog post of social media research tools for Twitter can be accessed here.

Easy-to-use tools for collecting Twitter data

  1. TAGS
    • Easy Twitter monitoring from within Google sheets. Still requires an API key.
    • Added: 18.11.2021. Verified: ?
  2. Postman
    • A desktop and web application that allows you to make requests to an API from a graphical user interface, offered by Twitter.
    • Added: 18.11.2021. Verified: ?
  3. Atlas.TI
    • The popular qualitative analysis software also offers options to make queries to Twitter Search API.
    • Added: 18.11.2021. Verified: ?
  4. DMI-TCAT 
    • Retrieve and collect tweets from Twitter and analyse them in various ways. The software requires API access and a server where to install it.
    • Added: 18.11.2021. Verified: ?
  5. Mecodify
    • Similar to T-CAT: tweet collection, simple analyses and visualization. Own server and API access required, also instructions for Amazon AWS free service provided.
    • Added: 18.11.2021. Verified: 14.12.2022.
  6. Netlytic
    • Netlytic is especially made for pedagogical and learning purposes, but some research papers have been published utilizing Netlytic.
    • Capture publicly available posts from social media sites or other sources (Twitter, YouTube, RSS Feed or text/csv file)
    • Discover popular topics
    • Find and explore emerging themes of discussions
    • Build, visualize and analyse online communication networks using social network analysis
    • Map geo-coded social media data
    • Added: 18.11.2021. Verified: ?
  7. Socioviz
    • Scrape and collect
    • Analyse any topic, term or hashtag
    • Identify key influencers, opinions and contents
    • Export data to your favorite Social Network Analysis tool
    • Tutorials and instructions for Socioviz are found on their webpage.
    • Added: 18.11.2021. Verified: ?
  8. NodeXL
    • Excel-based tool for data collection and scraping, network visualizations and key statistics. Basic version is free.
    • Only for Windows.
    • See also Social Media Research Foundation student license
    • Added: 18.11.2021. Verified: ?
  9. Chorus
    • Chorus is a free, evolving, data harvesting and visual analytics suite designed to facilitate and enable social science research using Twitter data.
    • Made for social scientists for research purposes
    • Added: 18.11.2021. Verified: ?
  10. Twitonomy
    • Twitonomy is an app you can use to get information from your own account and from the accounts of others.
    • Added: 18.11.2021. Verified: ?
  11. Osome – Observatory on Social Media Overview
    • Allows you to submit queries to a massive social media database
    • Given a set of Twitter hashtags and a time period, you can count the number of matching tweets, generate a time series, count the number of tweets from each user or just list matching Tweet IDs.
    • You can subsequently utilize Twitter's REST API to retrieve more information about tweets or users of interest.
    • This API is provided on behalf of an academic institution (Indiana University) for the sole purpose of non-commercial research (https://rapidapi.com/truthy/api/osome-observatory-on-social-media-1/details).
    • Added: 18.11.2021. Verified: ?


Tools that require coding skills

  1. Twurl
  2. Poll.emic
    • Data collecting, especially egocentric networks, visualization
    • Added: 18.11.2021. Verified: ?
  3. Python-twitter
    • This library provides a pure Python interface for the Twitter API.
    • Added: 18.11.2021. Verified: ?
  4. Pattern
    • Data mining: Google, Twitter, Wikipedia
    • Tools for data mining (Google, Twitter and Wikipedia API, a web crawler, a HTML DOM parser), natural language processing (part-of-speech taggers, n-gram search, sentiment analysis, WordNet), machine learning (vector space model, clustering, SVM), network analysis and canvas visualization
    • Added: 18.11.2021. Verified: ?
  5. Finnish language Twitter is being continuously collected since April 2020 by the Human Sciences – Computing Interaction (HSCI) research group at the University of Helsinki (kudos to Eetu Mäkelä!). The tweet id's provided can be used to download the full tweets from the Twitter API. Added: 18.11.2021. Verified: ?

  6. The Tweets Sampling Toolkit is a python toolkit contains a set of tools for 1) creating a random sample from massive (100M+) Tweet ID datasets and for 2) performing set operations with Tweet ID datasets .


Subject to a fee

  1. Discovertext
  2. NVIVO
  3. Tweetarchivist
  4. Legentic/Mohawk offers comprehensive access to Finnish language twitter data


  • No labels