When can you use crowdsourced information (that is, information collected from citizens through online platforms such as Twitter, Facebook, and text messaging) to get trusted
information?Are there any particular identifiable characteristics of crowdsourced info that lends to making the info more or less credible?
These are some of the guiding questions to a groundbreaking research project we are conducting, using the recently concluded Kenyan General Elections as a case study. The study; funded by the International Development Research Centre; runs to July 2013.Our project specifically assesses the following aspects of crowdsourcing (3 Vs of Crowdsourcing):
- Viability: in what situations/events is crowdsourcing a viable venture likely to offer worthwhile results/outcomes?
- Validity: does crowdsourced information offer a true reflection of the reality on the ground?
- Verification: what aspects of crowdsourced information can be verified, and if so, can the verification process apply automated features?
Based on our literature review, we have identified over 10 possible attributes and characteristics of crowdsourced data that might help to prioritize users and content. Interviews with traditional and crowdsourced data users groups (including BBC, Standard Media, Nation Media Group, Capital FM, and Al Jazeera) have revealed that time and resources limit the ability of organizations and individuals to sift through the massive amounts of generated crowdsourced data. Media organizations have stated that if particular characteristics or attributes could be identified as good indicators of reputability and credibility, verification of social media sources would become easier and more usable.
We are currently conducting post-analysis on the information collected during the General Election period (March 3 – April 9) to assess whether crowdsourced data is a true reflection of on-the-ground happenings. To collect and aggregate crowdsourced information around the election, we are using DataSift
; a platform for building applications with insights derived from the most popular social networks and news sources. We have also been assessing what aspects of the information generated during this time can be verified, or more importantly, which aspects are worth verifying.
We are continuing to collect and aggregate information generated, especially on Twitter, based on keywords used in conversations, hashtags and names of key places around the country to capture as much information as possible to conduct the post-analysis.We will also be going to the field at the end of April 2013 to conduct on-the-ground investigative interviews based around discrepancies or inconsistencies found in the data analysis.
For more information, please contact Angela Crandall ([email protected]) or Nanjira Sambuli ([email protected]