People often ask me why we teach data skills instead of programming or other IT disciplines. To answer their question, I’d like to compare it to a swimming diploma. It’s a nifty example for a country that’s almost entirely under sea level.
Data skills build on what you already know and allow you to apply your knowledge at work. Data exists in every branch. Just like a swimming diploma it’s an addition to your skill set. Knowing how to swim won’t replace your ability to walk. It means you can do both and combine them as you like.
The inaccessibility of statistics
The world of data is often seen as mysterious, complicated and objective. If we see people use data, we think they have a special kind of intelligence, and they know more than we do. Some people are convinced that if something is based on data, it has to be true. It will feel uncomfortable to criticise and ask questions to get the full story behind the data. To make sure your conclusions really make sense. In reality, it’s not as complicated or mysterious as you think.
Data is not much more than bits of information. When you bring those bits together, you get a dataset. If you need a question answered, you look at the dataset and see if you can find patterns you can analyse and visualise. Then you’ll create something that’s more accessible to others. That’s not too complicated, right?
A diverse perspective on data
Why are there so many data difficulties then? Usually because of human errors and decisions. For example, when we do not ask the right questions, or when we don’t have the data we need. Instead, we base our questions on what we have. This is like having a flat tyre. If you own a shop that only specialises in bicycle chains, you’ll look for a set with spare chains and oil. It doesn’t matter how many chains you replace, it won’t fix your bike. If you don’t ask the right questions or use relevant data, you’ll run into problems.
It’s also important to keep in mind the backgroud of people who work with data. If everyone who looks at the data has the same gender identity, ethnicity or age among other factors, you could fall into groupthink and miss important perspectives. If nobody can offer such a perspective, the outcome of your data research could even be dangerous. For example, when programming a predictive tool to see what neighbourhoods have a higher crime risk, based on civilian data. Or government programs that predict what kind of person is more likely to commit fraud. Not based on facts, but on a lack of perspective from the team that works the data.
What happens with your data?
If we use data that is produced by people, we also have the responsibility to treat sensitive data the right way, and with informed consent. People need to know what kind of data they produce with their devices, actions and preferences. They also need to know how their data is used and what the tradeoff is.
It’s good to know that an app for your supermarket could track your shopping habits to give your personalised discounts. With that comes the possibility you’ll be influenced to spend more money there. If you know this, you can make the decision to accept or decline these services. If you don’t understand how your supermarket app works, your data might not be used in your best interest.
Increasing data literacy is an important instrument to combat inequality. If we have a better understanding of how data influences society when it comes to budgets, goods and services, policies and daily habits, the power of data doesn’t solely stay in the hands of a small group of specialists. Instead, it becomes something that’s open and accessible, questionable, usable to build together instead of build for. And we all benefit from that in the end.