A three-day workshop on data journalism took place in Seoul, South Korea, in accordance with the 9th World Conference of Science Journalists (WCSJ). A total of 15 participants attended the activity: 10 participants selected through an open competition from Indonesia, Vietnam, the Philippines, Nepal and Sri Lanka, on top of a project manager, and three trainers.
Approximately 50 applications were received from the region: The management team included one journalist from Sri Lanka and one journalist from Nepal among the participants as an effort to expand SjCOOP Asia’s network. The selection criteria put emphasis on professional quality, and reflected gender and geographic balance.
DAY 1: Technical training on how to use data in reporting science
Professor Charles Seife’s from New York University was the first trainer. The session started with an introduction to the importance of checking the validity and accuracy of data before reporting on a number. He gave numerous examples where unverified numbers were used in news reporting. He advised the trainees to double-check the data collection method before using them.
After the introduction, he moved to the technical part of DAY 1’s curriculum: how to organize data using Excel. It started with introducing basic formula, and moved to developing a comprehensive set of data. The case study used as an example was the list of falsified science research and its authors, the number of their publication and the number of peer-reviews. He introduced how to use Excel e.g., pivot table to ask questions, such as how many research publication violated the code of ethics by institution.
The second part of the tutorial was based on an exercise with an actual database of the 2008 elections in the U.S., which showed missing votes between two ballot counts. The participants were given tasks to answer 1) where are hiding missing votes? 2) what is the pattern of missing votes 3) What data stories can a journalist find with this dataset?
He concluded his session with tips for data journalism: 1) novel dataset is where many newsworthy stories exist. 2) use smart search tools based on search formulas 3) commonness means interesting stories are not there.
DAY 2: Working on the individual data-driven story project
Day 2 was intended to provide a hands-on skills workshop. The goal of the Day 2 was to help the trainees to get started their data-driven story. The trainees’ main interest was to learn how they can apply the learned skills immediately to their journalistic work. However, as it was only a three-hour-long workshop, the goal of the workshop was limited to showing a pathway to the trainees, rather than completing the data journalism project.
Dr. John Bohannon from M.I.T was the second trainer. Mr. Bohannon broke down into four small groups according to core skills required to implement each participant’ data journalism project. His fellow trainers who were invited to the World Conference by the conference organizers kindly offered their help in the Day 2 workshop. In total, three trainers including Prof. Seife from Day 1, and Mr. Justin Mayo, data journalist from Seattle Times, led small groups.
DAY 3: Disaster Big Data case studies, Twitter as data scrapping tool and Google Map as a mapping tool
The 3rd day workshop by Mr. Hirofumi Abe from NHK intended to guide the trainees’ step-by step, using his Disaster Big Data project. After the first two days of workshop, DAY 3 workshop covered more complex data use process and top notch visualization.
By showing disaster data graphics, Mr. Abe simulated the steps he took. He also asked the trainees what type of data the trainees would gather to report on disasters. For the participants, a data validity verification method was among the main interests. Mr. Abe also instructed useful methodologies of big data collection and mapping: How Google Map can be used to build a basis in mapping, e.g., roads and streets, and traffics status. He used MERS cases to demonstrate how a mapping can be easily set up for a data-driven story without using complicated tools. He spotted MERS infection incidences around Seoul, and showed how the participants can develop a storyline about zones at higher infectious risks.
The trainer also taught the participants about how to scrap and process data from Twitter. Use of Google Map and big data scrapping from Twitter were well received by the trainees as they were immediately applicable to their everyday journalistic work.
DAY 4: Plenary session about data journalism tools, and travel to Fukuoka, Japan
The participants attended a plenary session about how to turn data into journalistic stories. Prof. Charles Seife recapitulated fundamental skills in data journalism-finding useful data, importing it into spreadsheet and database software, cleaning it up, joining it with other data, and making the most out of complex datasets. Topics such as Excel, SQL, Google Refine/Open Refine, frequency analysis were covered during discussion.
Mr. Jonathan Stray, data journalist, expanded the scope of data beyond numbers- piles of government records, company disclosures, and email archives. He mainly focused on document mining, with new techniques including visualization and natural language processing. His presentation about new open source tools was especially praised by the trainees.
Mr. Justin Mayo, data journalist, provided an overview of the latest tools for a beginning data journalist. His presentation included extracting data from PDFs, cleaning up and analyzing raw data, and visualizing and sharing results.
DAY 5: Data journalism session and visit to Fukuoka’s top notch research centers
Following the data journalism skill training held in Seoul, the ten trainees participated in a two-day post-conference tour to Kyushu.
Support for Science Journalists in Asia through SjCOOP Asia
Science journalism COOPeration in Asia or SjCOOP Asia aims to develop a regional community of competent science journalists in Asia, particularly in Southeast Asia by strengthening professional capacity and networks of local journalists. These journalists trained and networked internationally will play a key role in educating their audience on science-related issues closely linked to the development of their society.