CIB Mango Tree
A Civic Tech DC Project
About Download & Install Data Preparation Test Library
Graphing Post Time Test This test will graph all the posts in your dataset by the time of day that they were made. The test will yield a graph that visualizes the frequency of posting within discrete time windows of a length that you choose across a 24-hour period, allowing you to look for abnormal posting patterns. See Code for Test on GitHub Wang, Austin Horng-En, Mei-chun Lee, Min-hsuan Wu, and Puma Shen. "Influencing overseas Chinese by tweets: text‑images as the key tactic of Chinese propaganda." Journal of Computational Social Science 3, no. 20 (November 2020): 469-486.
Username
Unique Post Number
Timestamp
Post Content

  1. Set Time Window Length: Start by specifying the desired length of your time windows for the 24-hour period (recommended: 30 minutes).
  2. Tally Posts: The test will count the posts made in each time window across your dataset. For example: with a 30-minute window, it will count posts from 12:00 AM to 12:30 AM, 12:30 AM to 1:00 AM, and so on.
  3. Graph Results: Finally, the test will display these frequencies in a csv file and bar chart, with each bar representing a specific time window across the 24-hour period.

Output Format Option 1: .csv File
  • The test will produce a .csv file that provides the numerical tallies of posts for each given time window in the 24-hour period.
Sample CSV output
Output Format Option 2: Offline Dashboard
  • The test will also produce a link to a dashboard that can run offline in your browser (this dashboard does not require an internet connection). The dashboard will render the 24-hour graph of post frequency.
Sample Dashboard output