Top Ten #ddj: This Week’s Top Data Journalism

What’s the #ddj community tweeting about? Our NodeXL mapping from May 22 to 28 includes what it’s like to code as a journalist from @CJR, artists and designers visualizing Google search data from @FastCoDesign, the Data Journalism Awards 2017 shortlist from @mberzosa and Austria’s changing job market from @derStandardat.

Can We Make Algorithms More Accountable?

At the recent European Investigative Journalism and Dataharvest conference, Nicholas Diakopoulos spoke about the emergence of algorithmic accountability reporting, and how it applies the core journalistic functions of watchdogging and investigative reporting to algorithms.

How Not To Be Wrong

While your story can be true, it might be wrong. That’s what New York Times data journalist Robert Gebeloff explained at the recent European Investigative Journalism Conference & Dataharvest. Check out some of Gebleoff’s tips, and a data journalism checklist.

Top Ten #ddj: The Week’s Top Data Journalism Items

What has the #ddj community been tweeting about? Our NodeXL mapping from May 15-21 includes North Korea’s targets (@PostGraphics), Wannacry’s ransomware spoils in real time (@TA_Interaktiv), Berlin’s space use in a gorgeous map (@tagesspiegel) and the White House guests (@Poynter & @Politico).

Top Ten #ddj: The Week’s Most Popular Data Journalism Links

Here are top data journalism tweets for May 1-7, per our NodeXL mapping: delayed cherry blossoms (@TheEconomist); dataviz how-to (@albertocairo); 20 million starving (@washingtonpost); Munich route planner code (@Munichrocker); FB German filter bubble (@SZ); Open Data Index (@pinardag); and more.

Top Ten #ddj: The Week’s Most Popular Data Journalism Links

Here are top data journalism tweets for Apr 24-30, per our NodeXL mapping: draw it yourself interactives (@br_data); visualizing Google searches (@albertocairo); opinion polls analysis (@SZ); Politibot’s graphics (@politibot); French election (@BIQdata); ProPublica jobs (@ProPublica); and more.

Así investigó Univision Data las debilidades de la defensa pública en Costa Rica

Tras cuatro años de trabajo y 8.000 sentencias judiciales procesadas, el equipo de Univisión Data reveló que en Costa Rica una persona tiene más probabilidad de ser condenada en un juicio si tiene un defensor público que si tuviese uno privado. La metodología incluyó “web scraping”, R regresión logística, un método estadístico de uso muy frecuente en las ciencias sociales pero prácticamente inexplorado por las salas de redacción.