This post is a short version of my academic paper on this topic, entitles 3Ws of Data Journalism Education, published by Data Journalism Practice. An open access post-print of the paper could be accesses via UCD academic repository.
Data journalism is an emerging area of practice and study, which draws on knowledge from several disciplines, including journalism, information science, social sciences, data and computer sciences, data analytics, information design, and storytelling.
Having been built on the foundations of Philip Meyer’s precision journalism, data journalism promotes a fact-based and scientific approach to journalism. Meyer calls for journalism to be treated and practised in the same way as scientific investigations, inviting scientific methods, scientific objectivity, scientific transparency, scientific reproducibility — in general the ideals of the scientific method — to the process of journalism and mass communication.
Such scientific methods include quantitative and qualitative data analysis for investigation, producing journalistic content and communicating these content to the public.
This new focus on data calls for a radical reconsideration of journalism programmes across the world, to include training on such methods that equip journalists with skills required for finding the facts in today’s data intensive economy, to understand them, scrutinise them, and communicate them to the public in the most appropriate and understandable manner.
Data Journalism Education
Data journalism, as we know it today, has been growing for the past 10 years. Many media organisations have data journalists in their newsrooms and organisations and/or are increasingly interested in hiring journalists with data skills in the past few years. Similarly, academic and educational institutions are making concerted efforts to include data journalism in their programmes.
Incorporating data journalism in education and training curricula is a rather new development, and only in the past five years or so have we witnessed a growing number of data journalism modules and programmes in university curricula, the establishment of MOOCS, publication of books and, in very few cases, textbooks on data journalism (examples of books include Gray et al. 2012, Miller 2013, Mair et al. 2013, Bradshaw 2015, Felle et al. 2015, Vallance-Jones & David 2017, and Mair et al. 2017).
There are a limited number of research publications studying educational needs and the state of data journalism education. Examples include a survey by the European Journalism Centre in 2011, a study by Berret and Phillips in 2016, and a study by Splendore et al.
The existing research, however, were each focused on specific geographical areas, such as a specific country or a number of countries in a continent, e.g. Europe. There was no research to look into the global state and whereabouts of data journalism education, which in my opinion is an important missing piece of puzzle, both in terms of research, as well as a resource for students who are eager to study data journalism.
If you are interested to know more about previous research and studies on this topic, I have provided a brief review of relevant studies in the paper itself. Here I get straight into my new study.
In this research I looked into (mainly academic) offerings on data journalism around the world. For this, I compiled and studied a dataset of data journalism modules and programmes offered in 219 unique data journalism related modules and programmes around the world.
Note: I use the terms course and module interchangeable. Module is the term used in the UK and Ireland, and course in many other countries, including the US.
Another note: Since the publication of the paper three new courses are added to the dataset, increasing the total to 221. These two courses are both in Spain. This article reports on the finding from the paper, hence the 219 listing. However, the shared dataset, and the map, present the latest data at any given time, and both will potentially keep updating.
This dataset was compiled using data collected from a variety of existing data sources, including Dan Nguyen’s crowd-sourced curated list of data journalism syllabi (144 out of 219 courses are from Dan’s list), the CAR and Data Syllabi by the Investigative Journalism Education Consortium, a list curated by Daniele Palumbo in 2013, a list of academic publications by Splendore et al.,and an extensive search and asking around.
The details of how I compiled this dataset is explained in length in my paper, if you are interested to read more about the method behind it.
The final dataset is composed of the following fields/variables: ‘id’, ‘title’, ‘theme’, ‘organisation’, ‘school/sub-org’, course listing/code, type (full programme or module), level (UG/PG), credit, start year, latest offering, homepage, instructor 1, instructor 1 highest education level, instructor 2, instructor 2 highest education level.
While a great deal of effort and attention was paid toward compiling a comprehensive dataset, by no means do I claim that I have created a complete list of data journalism courses and programmes across the world, nor do I believe such a complete list is easily achieved. In particular, despite best efforts, finding courses, or finding sufficient information on programmes in non-English speaking countries was difficult and not straightforward.
I have shared my dataset publicly (some fields are removed for privacy reasons), and have made it open for all to make recommendations of new courses in this.
Where are all these Data Journalism courses?
So the first question, specially if you are looking for a place to study data journalism, is where is data journalism being taught? Which countries? Which universities?
I have compiled a map of all courses and programmes in my dataset to make it easy to see the trends, also for you to find a university close to you if you are looking for a place to study data journalism. You can browse the map for specific modules and programmes, and their information, including links to syllabi.
Looking into the geography of data journalism courses taught in universities and academic institutions, immediately it is clear that the United States has the largest offerings in data journalism related modules and programmes. Furthermore from the map we can see that East coast the US has the highest intensity of data related courses in journalism education, followed by theWest Coast US.
To put it in numbers, 146 of the courses are in the United States, 8 are in Canada, which, excluding online courses, leaves only 63 courses and programmes outside of North America offering data journalism related topics altogether.
These new figures are plausibly in line with a decade old study by Yarnall et al., which looked at how CAR was taught in Journalism schools. New figures show that while the United States still has a higher (and in fact in 2018 a considerably higher), coverage of data-related skills in their journalism programmes in comparison to non-US schools, unlike in 2008, such data related skills are now an integral part of the structure of many journalism programmes in the United States, and to a smaller extent across the world in 2018.
Outside of North America, Europe has the second highest concentration of data journalism related courses. The United Kingdom, the Netherlands, Ireland and Australia are the countries with the highest number of data journalism related modules and programmes outside of North America. The number of courses per country, where two or more modules/programmes are available in a country, is presented in the following table.
To what extent and at which levels are they teaching it?
Amongst these 219 modules and programmes listed in the dataset, only 25 are programmes fully and specifically on data journalism. In other words only 25 universities around the world provide degrees or programmes dedicated to data journalism. Despite this, many universities consider data journalism an important topic, and there are153 instances of stand-alone modules on data journalism in varying non-data journalism focused programmes. The rest are online, vocational or undefined.
Overall out of 24 countries present in this dataset only the United States, United Kingdom, Ireland, Germany, Canada, Spain and Hong Kong present a strong focus on data journalism as a programme of its own, with more than one module dedicated to data journalism, or having postgraduate programmes in data journalism.
Out of all European countries in the dataset, only the United Kingdom (three universities), Ireland (one university) and Spain (two universities in the original dataset, 4 universities after updates) have a strong focus on data journalism as a self-contained programme. The rest of the countries in Europe only provide one or two stand-alone modules in this area.
In terms of level of education — undergraduate level module/programme or postgraduate level — 48% of these modules or programmes are postgraduate courses, 35% are undergraduate courses, and the rest are undefined, used in both undergraduate and postgraduate, vocational or online courses.
What data journalism skills are taught in university programmes?
To understand the types of data related courses and programmes taught across the world, I categorised the 219 modules to 17 themes. These 17 themes are: CAR, coding, computational journalism, data analysis, data journalism, data science, data visualisation, digital humanities, digital Journalism, digital media, Investigative journalism, Journalism, online journalism, precision journalism, Research and Web programming, other and unknown.
The results of the 219 module analysis show that most of the courses listed are, not surprisingly, focused on ‘data journalism’ as an overarching topic. These courses have a primary focus on the complete workflow of data journalism, from finding, collecting and cleaning data, to analysis, visualisation and communication.
This category forms the over half (55%) of data-related courses taught in journalism programmes across the world. These courses, while often on the less advanced end of the spectrum, prepare their students to complete a data journalism project and produce data journalism output individually or in small groups, even if they are not using advanced tools and methods.
This category of courses may also be prevalent because they provide a foundation for more advanced courses in some universities. In many cases, the universities who have more advanced courses, such as coding, computational journalism or data analysis and data science related courses, list a foundational data journalism course as a prerequisite of these more advanced courses.
Computer Assisted Reporting (CAR) is the second most prevalent data-related category of topics taught in journalism programmes, with 10% of courses being tagged as ‘CAR’. The syllabi of the courses in this category are often not too far away from those tagged as ‘data journalism’, but CAR courses have a stronger focus on databases than on data analysis and visualisation.
CAR elated courses often include lectures on databases, as well as SQL and other methods and tools for creating and interrogating databases. While SQL and similar database query languages are sometimes present in ‘data journalism’ modules, they often are not the main focus in that category. CAR courses tend to have less focus on data and statistical analysis than their ‘data journalism’ tagged counterparts, but that is not always the defining factor.
Except for one case (which was in Thailand, and is no longer an active course), the courses tagged as CAR are all taught in North America, with 18 being in the United States, and 2 in Canada. Those in the United States are mostly courses that came to existence before the more recent popularity of data journalism as a whole. In some cases the content of the modules under the CAR tag had changed since the time they were initiated, and are closer to the content of the ‘data journalism’ category at the present time. However, if the content was closer to ‘data journalism’ but the name was a variation of CAR, I kept the tag as CAR.
Overall, ‘data journalism’ and ‘CAR’ categories are both foundational and in many cases similar topics, and while I have put them in two separate categories, we can conclude that an overarching 65% of the 219 courses form a variation of foundational data journalism and CAR, despite the content and historical differences.
Data journalism and CAR are followed by courses which have a stronger focus on ‘data visualisation’ (7%), and then courses focused on ‘coding’ (5%) and ‘computational journalism’ (4%) respectively.
The computational journalism tag is used for courses that pay strong attention to the process of data journalism, while the main focus is on computational methods, as opposed to data analysis. This is distinguished from those tagged as ‘coding’, which are courses focused solely on teaching coding and programming, such as Python, and less so on the process of data or computational journalism.
Courses focused primarily or only on data analysis, as opposed to the basic data analysis covered in foundational data journalism courses, form only 2% of the courses in the dataset.
What is lacking?
Putting the above themes taught in data journalism courses and programmes next to the results gathered from my earlier study on the state of data journalism (the Global Data Journalism survey), it becomes apparent that while most journalists are interested in learning about data analysis, apart from the basic data analytics covered in foundational ‘data journalism’ courses, very few courses on more advanced data analytics are offered in journalism and related programmes. I have included a chart from this study below, so you won’t have to go back and forth.
Perhaps this is the reason why most journalists feel they need to learn more about data analysis, and a reason behind journalists’ ‘number phobia’ highlighted by Nguyen & Lugo-Ocando.
This results, put next to former calls on the importance of statistics and data analysis in journalism, (e.g. Nguyen & Lugo-Ocando’s), calls for an immediate attention in covering more data analytics and statistics courses in journalism and data journalism higher education programmes.
Apart from the apparent gap in courses covering ‘data analysis’, the rest of the categories and areas that interest journalists are not too far away from programmes offered on the topics across the globe.
While many journalism programmes still do not cover data skills, many have introductory offerings on the topic, and a few have more advanced offerings. As a general observation, journalists and journalism graduates lack sufficient data skills. However, they express interest and the need to further their skills in these areas.
Put next to previous study on the state of data journalism globally, the result shows that there is a gap between what the journalists want to learn and what the universities are offering. This gap is widest when it comes to Data Analysis skills.
This gap calls for a reform in journalism programmes across the world.
Data journalism is an interdisciplinary field, and while I expect we will see an increase of data offerings in journalism programmes in the upcoming years, we require a broadened approach to data journalism training. This approach must be mature enough to facilitate entrants from the various disciplines that converge to create data journalism professionals.
Furthermore, the new and interdisciplinary nature of data journalism requires educators with a diverse set of skills, alongside a deep understanding of both academic and practical aspects of disciplines involved.
This post is a short version of my academic paper entitles 3Ws of Data Journalism Education, published by Data Journalism Practice. An open access post-print of the paper could be accesses via UCD academic repository.
The dataset behind this study could be found here.
If you know of a module, course or programme that is missing from my dataset, please consider adding it here.