Data Journalism in Indonesia: Practice, challenges, and barriers of doing data journalism in Indonesia 1 ABSTRACT This research tried to look further at the data journalism practice among Indonesian journalists, to see how far they are practising data journalism in their daily work. The population of the survey is not only data journalists, but all journalists in Indonesia no matter which sector they are covering. This research has found that the lack of knowledge in data analysis and access to the data became two main barriers which were mentioned by the respondents. It is understandable since data analysis never been taught to journalism students at the universities in the country. 1 2 INTRODUCTION Howard (2014) defines data journalism work as gathering, cleaning, organising, analysing, visualising, and publishing data to support the creation of acts of journalism. One way to understand how data journalism has grown in a country is by knowing how the journalists are working with data. The idea of this research came from my personal need. In November 2017, I considered starting an organisation or a community in Indonesia that allows the members to learn data journalism. In the process of starting the organisation, I tried to find some researches about data journalism in Indonesia, to know how the Indonesian journalists have worked with data. Unfortunately, I could not find one. Later, I thought researching it might be useful, not only for the organisation that I wanted to build but also for other people interested in the practice of data journalism in Indonesia or in the development of knowledge of data journalism, such as media organisations, universities, or even the organisations that focus on journalism training. In an interview with Bahareh Heravi 1, Mar Cabra, Head of Data and Research Unit at International Consortium of Investigative Journalism said: “There is no way to do good journalism today without looking at electronic records, and many of those electronic records end up being lines of data.” Ideally, every single journalist should know the essential skill on how to work with data, such as doing the calculation with a spreadsheet. In this ‘big data’ era, they need to enhance their technical skill into more advanced data analysis. Paul Bradshaw, a data journalist and an MA Data Journalism course leader at Birmingham City University believes that Data journalism is for almost every journalist (Uskali &Kuutti, 2015). “If you are a journalist, you have to deal with all types of information then you have to work with that. Speed and accuracy are the two key assets for a journalist. And if you deal with data, you have to do with that,” Bradshaw said in an interview with Uskali & Kuutti. For that reason, this research tried to look further at the data journalism practice among Indonesian journalists in general, not only those who are labelled as a data journalist. The questions addressed in this research are; (A) (B) (C) (D) (E) How do the Indonesian journalist work with data? What are their data sources? What technical skills do they use? What are their barriers to working with data? What kind of skills do they need to be developed? 1 http://datadrivenjournalism.net/news_and_analysis/from_zero_to_hero_how_data_journalism_helped_estab lish_the_icij 2 3 RELATED WORK In 2011, the European Journalism Centre (EJC) circulated a survey on training needs for data journalism. The survey asked how journalists use data, their level of expertise, and the barriers they faced2. It reached over 200 respondents from 40 countries in Europe, North and South America, Australasia, Asia and Africa. According to the survey, there was a significant willingness to get out of the comfort zone of traditional journalism and to invest time to master the new skills. The results from the survey depict that journalists need support to cut through the initial problems keeping them from working with data (Laurens, 2011). After 2011, several studies and researches on the practice of data journalism in numerous countries have been published. Some of them are: Computational Journalism in Norwegian Newsrooms (Karlsen and Stavelin, 2014), Data Journalism in Sweden (Appelgren and Nygren, 2014), Data journalism in the UK: a preliminary analysis of form and content (Knight, 2015), and Data Journalism in the United States (Fink and Anderson, 2015). Karlsen and Stavelin (2014) interviewed nine computational journalists in Norway to see how they work in the newsroom. In the United States, Fink and Anderson (2015) also interviewed 23 data journalists across various newsrooms to find out the process of producing data stories and how data journalism fits into the work of their perspective organisations. In Sweden, Appelgren and Nygren (2014) did not only interview editors in seven newsrooms, but also distributed an online survey and got 194 respondents. The purposes of their research are similar, knowing how the data journalist work in the newsroom, including the practice and the challenges. Content analysis also becomes a method to see the progress of data journalism in one country. Knight (2015) did this in the United Kingdom. She analysed 112 newspapers from 15 organisations and found that most of the data that had been published by the newspapers came directly from a third party such government press releases or data from NGOs. Those data came holy wrapped as it were. In Indonesia, there is no research about the practice of data journalism yet. This paper aims to depicts what knowledge the Indonesian journalists have in that field and how to improve that. 4 METHODOLOGY This study was conducted by designing an online survey consisted of 24 questions in two sections. The online survey was launched on 10th July 2018 and remained open to all Indonesian journalists until the 9th August 2018. One hundred and five (105) participants from 61 news organisations in 23 cities in Indonesia participated in the survey. After a verification, two of 105 responses were not included in the analysis due to incomplete answers. In result, only 103 responses were calculated in data analysis. Due to time and geographical limitations, the survey was carried out with a non-probability sampling method; consequently, it does not represent all Indonesian journalists in 33 provinces. 2 http://datadrivenjournalism.net/news_and_analysis/data_journalism_survey_analysis 3 Notwithstanding, to ensure the diversity regarding location and media organisation types among the participants, the survey was shared with journalists in several cities from all medium; print, TV, radio, online. Email, chat application, and social media were used to distribute the questionnaire. 4.1 RESPONDENTS’ FIGURES Here are the figures of 103 respondents: 1. More than 50% or around 67 respondents are based in Jakarta. The rest of them are in the other 22 cities. We can see that the number of respondents in Jakarta are very dominant compared to the other cities. Likewise, the number of media organisations based in Jakarta is also much more significant than the other cities because all national publications (print, online, and television) are found in this city. 2. The respondents have diverse working experience, in less than one year to more than ten years. 4 3. The number of male respondents in this survey is more prominent than the female, but the gap is not that high. 58% respondents identified themselves as male, 41% identified as female, and 1% prefer not to answer. 4. The respondents represent all news medium, online, print, radio, TV, news agency, and freelancer. 5 5 FINDINGS 5.1 THE FREQUENCY OF USING QUANTITATIVE DATA Not all respondents usually work with quantitative data in their daily journalistic work. Among the 103 respondents, only 17% said that they always work with data. About 42% said that they often use data, 40% said they rarely use data in their work, and 2% never use quantitative data. 5.2 GOVERNMENT’S OFFICIAL DATA BECOME THE PRIMARY SOURCE OF DATA Where does the data come from? All respondents answered that the government’s official data is their main resource. Apart from it, scientific research and data which is published by the nongovernment institutions also become their primary sources of data. Not many respondents ever created their own databases, only 21% said that they ever make a database. Web scrapping also is not widespread to do by Indonesian journalist, only one out of five journalists said they ever did that. Sources of data Government's official data 103 Scientific research 82 Data which is published by non-government institutions 80 From company's official website 64 From a press realease 55 Creating a database 22 Web scrapping 20 Others 7 0 20 40 60 80 100 120 6 5.3 LACK OF KNOWLEDGE IN DATA ANALYSIS When it comes to data analysis, more than half of the respondents said that they never analyse the data since the data was ready to be published. Only four respondents know how to use R and just one respondent ever use Python. However, 43 respondents (42%) understand how to analyse data in a spreadsheet. Which data analysis tools do you usually use? Stata 1 AMOS 1 E-Views 1 Python 1 SPSS 2 Never work with data 4 R 4 Data analysis is done by another team 9 Spreadsheet 43 Manually count or using calculator 52 No need to analyse because the data is ready to be published 57 0 10 20 30 40 50 60 5.4 DATA VISUALISATION USUALLY IS DONE BY GRAPHIC DESIGNER The majority of the respondents, about 73%, never visualised their data. Only 9% of them did the data visualisation as well as the other journalistic tasks, and 18% said sometimes did that. 7 The most popular data visualisation tools among the respondents who do data visualisation by themselves are Photoshop, Tableau, and Canva. Some of them use several free online tools such as Datawrapper, Infogram, and Pictochart. WHICH TOOLS DO YOU USUALLY USE FOR DATA VISUALISATION? ORAD 1 VIZRT 1 SPECIAL TOOLS OWNED BY MY NEWS ORGANISATION 1 PICTOCHART 4 INFOGRAM 5 DATAWRAPPER 5 ADOBE ILUSTRATOR 7 CANVA 9 TABLEAU 10 PHOTOSHOP 11 0 5 10 15 5.5 NOT FAMILIAR WITH AUTOMATION TOOLS In the survey, the respondents were given a list of automation tools that usually help a data journalist work. The tools are Google Alert, Google Trend, IFTTT, Twitter Bot, Slack Bot, Dataminr, and CrowdTangle. More than half of the respondents said they never use one of them. WHICH AUTOMATION TOOLS DID YOU EVER USE? 59 NONE OF THEM 39 GOOGLE TREND 17 GOOGLE ALERT 10 CROWDTANGLE 8 TWITTER BOT 3 DATAMINR 2 SLACK BOT 1 IFTTT 0 10 20 30 40 50 60 70 8 5.6 THE BARRIERS OF DOING DATA JOURNALISM IN INDONESIA Doing data journalism in Indonesia is quite challenging due to the lack of transparency and open data from the government. Four main barriers were mentioned by the respondents are limited access to the data, limited technical skill in data analysis, inaccurate yet old data, and limited time for the journalists due to the daily deadline. What is the barriers of doing data journalism? Data journalism is not the priority of my news organisation There is no internal research team There are many versions of data for the same thing I have to work by myself because of skill limitation among… Finding story in data It is hard to narrating the data Inconsistency of data structure Data visualisation is quite chalenging for me Don't know how to verify the data Limited time to work on data due to daily deadline Inaccurate and old data Limited technical skill in data analysis Limited access to the data 1 1 1 1 3 4 4 4 7 10 14 19 26 0 5 10 15 20 25 5.7 TOPICS IN DATA JOURNALISM THAT THE JOURNALISTS WANT TO LEARN About 94% of respondents said that they are interested in learning data journalism. Around 75.7% said they would join if there is a community where they can learn together about data journalism, and 24.3% said they probably would join. Here are the topics around data journalism that they interested in more. 3 things that you want to learn about data journalism Data cleaning 25 Bot and automation 27 Data analysis with spreadsheet 45 Coding for journalist 46 Statistic for journalist 61 Data visualisation 62 Data gathering 66 Social media analysis 67 0 10 20 30 40 50 60 70 80 9 30 6 DISCUSSION AND CONCLUSION The world has become so complicated with the explosive of information, for that reason, a journalist has to be a database manager, a data processor, and a data analyst (Meyer, 1991). This study is trying to see how far the Indonesian journalists are practising data journalism in their daily work. The population of the survey is not only data journalists, but all journalists in Indonesia no matter which sector they are covering. Among the respondents, two journalists identified themselves as a data journalist. The survey showed that only 17 out of 103 respondents said they always work with data every day. However, 10 of that 17 are covering economy and business—a reporting area that indeed uses quantitative data frequently. Respondents who answered seldom and never use data are mostly covering politics, environment, arts and culture, human rights, law, and IT. Those areas also need data to make the stories more powerful, and the data are available. Using data in journalism is not only for the economy and business journalists. It is something that journalists can explore if they know how to work with data. For politics story, for instance, journalists can create a mapping of the unclear ideology behind each political party in Indonesia or do a fact-checking of some claims from the politicians. Some news organisation proved that politics stories also need data. In May this year, Jaring.id published a story about ten province or city leaders in Indonesia with the biggest corruption3 level. In July, Tirto.id published a story that criticised the mental revolution program by the current government with data about the cost4. This research has found that the lack of knowledge in data analysis and access to the data became two main barriers which were mentioned by the respondents. It is understandable since data analysis never been taught to journalism students at the universities in the country. There is no such a course called data journalism or computational journalism in Indonesia universities. Statistic and public opinion are in the module of journalism course, but it is not in line with some update tools that journalist can use. As far I know, only Universitas Multimedia Nusantara (UMN) in Jakarta that now has a module in data journalism. In contrast, we are at a stage where many media organisations are increasingly interested in hiring journalists with data skills (Harevi, 2018). At this point, intensive training in data journalism for the Indonesian journalists is needed. Those media organisations should equip their journalists with training in data journalism for the sake of the better journalism in the Indonesian media industry. Also, the universities cannot close their eyes to this trend in journalism. They should update their syllabus and start to teach data journalism. 3 4 https://jaring.id/stories/sepuluh-kepala-daerah-dengan-nilai-korupsi-terbesar/ https://tirto.id/biaya-revolusi-mental-dari-iklan-kkn-mahasiswa-hingga-taman-cPPT 10 7 BIBLIOGRAPHY Appelgrcoen, E. and Nygren, G. (2014) Data Journalism in Sweden. Digital Journalism, 2:3/11 pages. Available at: https://doi.org/10.1080/21670811.2014.884344 [Acessed 20 February 2018]. Fink, K. and Anderson, C. W. (2015) Data Journalism in the United States. Journalism Studies, 16:4, 467-481. Available at: https://doi.org/10.1080/1461670X.2014.939852 [Accessed 20 February 2018]. Harevi, B. (2018) 3WS of data journalism education: what, where and who? Journalism Practice. 18 pages. Available at: https://doi.org/10.1080/17512786.2018.1463167 [Accessed 30 June 2018]. Howard, A. (2014) The art and science of data-driven journalism. [Online]. Available at: http://towcenter.org/wp-content/uploads/2014/05/Tow-Center-Data-Driven-Journalism.pdf [Accessed 30 May 2018]. Karlsen, J., and E. Stavelin. (2014) Computational journalism in Norwegian newsrooms. Journalism Practice, 8 (1)/14 pages. Available at: https://doi.org/10.1080/17512786.2013.813190 [Accessed 20 February 2018]. Knight, M. (2015) Data journalism in the UK: a preliminary analysis of form and content. Journal of Media Practice, 16:1/18 pages. Available at: https://doi.org/10.1080/14682753.2015.1015801 [Accessed 20 February 2018]. Lorenz, M. (2011) Training data-driven journalism: Mind the gaps. [Online]. Available at: http://datadrivenjournalism.net/news_and_analysis/training_data_driven_journalism_mind_the_ga ps [Accessed 20 August 2018]. Meyer, P. (1991). The new precision journalism. Bloomington: Indiana University Press. Uskali, T & Kuutti, H. (2015) Models and streams of data journalism. [Online]. Available at: https://www.researchgate.net/publication/276343897_Models_and_Streams_of_Data_Journalism [Accessed 20 August 2018]. Additional readings Davies, M. and Hughes, N. (2014). Doing a successful research project using qualitative or quantitative methods, 2nd ed. London: Palgrave Macmillan. Raune, J.M. (2005). Essentials of Research Methods: A Guide to Social Science Research. Oxford, Blackwell. Chapter 2: Ethics - It's the right thing to do (pp16-31). Sapsford, R. and Jupp, Victor. (2006). Data collection and analysis, 2nd ed. London: SAGE Publication. Timpany, G. (2016). Tips for measuring frequency of usage in surveys. [Online]. Available at: https://blog.cvent.com/events/feedback-surveys/tips-measuring-frequency-usage-surveys/ [Accessed 31 August 2018]. 11