Tan cerca, tan lejos. (So far, so close)

Tan cerca, tan lejos. (So far, so close)

Organisation: La silla vacía (Colombia)

Publication Date: 04/14/2016

Size of team/newsroom:small


"So close, so far" is an interactive app that allows readers to understand where they lie on the spectrum of candidates in an election, thus allowing them to compare their thoughts to candidates' and make an informed decision on who to vote for. We set up a questionnaire with the most pressing is issues for each of the 5 largest cities in Colombia. Respondents were expected to answer in a scale from Totally agree to totally disagree. This questionnaire would ask them about the needs of the city or the need to preserve policies from the current mayor at the time. We approached all the candidates in 5 cities and ask them to fill in the questionnaire. With this information at hand we prepared the data app that would allow citizens to respond to the same questionnaire. After they sent their responses they would be redirected to the results page. In this results page, user would see 3 things: - A continuous scale were they would be placed closer or further to the candidates depending on how similar or dissimilar responses the gave to the issues of the city described in the questionnaire. In the scale, candidates themselves would be put closer together depending on how similar their answers were. - The answers each candidate in their city gave for all issues, and weather they gave additional comments on the question. - The possibility for users to compare themselves to each candidate separately. - The possibility for users to see where the average of all other users of the app lie. The story, which is nothing more than an invitation to fill in the questionnaire for Bogota got 20.000 views until the day of the election, of which, more than a 11.000 users filled the questionnaire and interacted with the app. In total the app was used by around 15.000 users the days before the elections that took place on October 25th 2015. This means that 15.000 users were engaged enough to fill an online questionnaire with 12 questions just to understand more clearly which candidates they were closer to by measuring many dimensions at once and showing them in an intuitive way. We can set up the application for any newsroom covering elections in any country and provide the hosting and setup to target the individual context for an election in a country for a flat one time fee.

What makes this project innovative? What was its impact?

In terms of visualization we condense multiple dimensions (the different issues each candidate respond to) into a single intuitive line that let's you see clearly who is closer to (or farther from) whom. This is very different from the way usually this type of information is visualized with bars of pie charts that show the percentage of people that answered the same to a candidate. The app invites people to actually give in their thoughts if they want to get something valuable to make a better informed decision about who to vote for. What's also innovative is that we started integrating methods of data science not usually available in newsrooms. We used machine learning techniques to evaluate responses from qualitative information and process them in a way that is complex enough mathematically but showed them in a simple way to end users.

Technologies used for this project:

An online form service that works really well on mobile. With the use of their API connections to get the data and process them using the R language for statistical programming in a Shiny (R framework) for Web applications to display information using wrappers and connectors to different javascript libraries like d3.js for the graphical scale and a jquery plugin called datatables to visually explore tabular data.
Follow this project

Comments (0)

You have to be connected to contribute

You have to be connected to follow

Leave this project and no longer be informed about this project

By joining this project, you will be informed by email when an update or a new contribution is posted on the website.

Thank you for your active participation !

The GEN Community Team