Analysis interactivity: Automation jobs by the ONS
When I think about interactivity in journalism, one of the first things that pops in my head is bots. Maybe it’s the influence and interest I had about Politibot.
A Spanish project through Facebook Messenger and Telegram, where a chatbot would interact with the user about a relevant news that occurred that day or the day prior and gives you some background.
If you didn’t have time, or even haven’t read that much about the news; you’d be well informed on the subject of the day (most of the times it would be Spain-centric) in less than 10 minutes.
With that in mind, I thought about another story that used a chatbot to interact with the reader.
A few months back I read a data analysis from the ONS about the jobs at higher risk of being automated.
And for it they used a chatbot, which felt a bit ironic but at the same time was on point.
Before being introduced to the robot, the reader is given a general explanation of the results of their analysis, such as how many jobs are at a high risk of being automated.
And then comes in the chatbot who allows you to “play” with him and you’re given two options at the beginning: “give me a specific occupation” or “general facts”.
Once that first choice made, you’re given other options: age, gender, or place in England. It’s an interesting way to interact with the reader and catch your curiosity about a subject that may affect their profession.
Different types of interactivity
But the interesting part is that the bot serves more as an addendum to the story, you can either choose to ask it about information, or scroll down and look at the interactive charts/map.
This could be considered as Jensen’s “Consultational” type of interactivity. You can interact with the same data in different ways. Giving you freedom to choose how you want to interact with the information.
If you are reading it via your computer, you can easily skip through the section you are most interested in knowing more.
And each part has a succinct explanation on what they were able to extract from the data they got. Similar to what you may have been told by the bot.
Back to Jensen’s types of interactivity, those charts are “Transmissional” since you can hover with the mouse and get extra information. Or even search for a particular job you have in mind.
The first interactive chart is about the jobs (see picture above), and they are reunited in 9 categories, and ordered through least threatened to highest chance of being automated.
Each dot is a profession; and if you point at it, you’ll see which one it is; and the percentage of risk from automation.
It’s a simple and easy way to understand the information and involve the reader in looking the data.
Presentation
The structure of the article reminds the Martini Glass Structure, where the reader is first given a brief explanation about the data analysis made by the ONS and with the bot as an introduction to it.
Nothing forces you to use it, but it is implied to at least ask the robot a question to later have liberty to go through the different sections and have a different interactivity.
Another great feature is the fact that the ONS allows you to use their content, with a simple click you can either download the data or embed the code of the chatbot or the graph; this way you can publish it on your website, or your news article.
Is it news for everyone or journalists
The ONS is not a news organisation, and thus you can sense it was not necessarily written for that format; despite not following the inverted pyramid, you still are receiving valuable information.
The content is not complicated to understand for any reader, but at the same time, there’s plenty data material for journalists to work with it and write a story about a particular subject (e.g. “Which city in England is the most likely to be exposed to automation?”).
