Data Collection
While we often talk about 'data', it's useful to distinguish a few types, as well as where the data originated from.
Sensitivity levels
Some of your data is very sensitive, such as your health details. Other data, such as which city you live in, is less likely to harm your future. In privacy law, there are three levels of sensitivity*.
| Aggregated data | This can be anonymised or 'pseudonimised' data. Data that an organisation has that might say something about you, but isn't directly linked to you. For example, if a company knows that the average age in your street is 28, that says something about you: you live in a neighbourhood with a lot of young people. But it doesn't reveal or guarantee that your age is also 28. |
| Personal data | Your name, phone number or email address is personal data because it's linked to you. Sometimes a combination of data can become personal data. For example, your IP address by itself doesn't reveal a lot, but in reality it can be easily used to find out more about you from specialised parties. For example, advertising companies that remember which IP addresses visited certain websites. |
| Sensitive data | This is a type of personal data that is extra special because if it was leaked, it could seriously harm your future. For example, your religion or sexual preference could be used against you, and in certain countries could lead to imprisionment - or worse. That's why this type of data has extra legal protections, and may not be processed without very good reasons to do so. This category includes: Racial or ethnic origin, Political views, Religion or philosophy, Membership of a trade union, Genetic data, Biometric data, Medical data, sexual behaviour or sexual orientation. |
Origin of the data
Where did the data come from?
| Provided by you | If you enter your email address in an online form, then you are directly and conciously sharing the data. |
| Observed data. | You provide this data too, but the difference is you might not be (continously) aware of it. For example, if a security camera records you. Even if you noticed the camera at first, you may forget about it a moment later. Cookies used to track which websites you also visit fall into this category. |
| Provided by others | This is data that was conciously made available by other sources, but which didn't involve any payment. For example, government institutions need to exchange data in order to function. It can also be 'open data' or publicly available data. |
| Paid data | Any data that is gained by paying money. This can be data that is literally bought, but it can also be 'rented' or gained access to as part of a subscription to a piece of software. |
| Created data | An organisation can combine some bits of data to create a brand new piece of data. For example, a company might label some customers as VIP (very important) based on how often they purchase their products. Algorithms can be used to create predictions about people based on data from their past. |
- Provided by you. If you enter your email address in an online form, then you are directly and conciously sharing the data.
- Observed data. You provide this data too, but the difference is you might not be (continously) aware of it. For example, if a security camera records you. Even if you noticed the camera at first, you may forget about it a moment later. Cookies used to track which websites you also visit fall into this category.
- Provided by others. This is data that was conciously made available by other sources, but which didn't involve any payment. For example, government institutions need to exchange data in order to function. It can also be 'open data' or publicly available data.
- Paid data. Any data that is gained by paying money. This can be data that is literally bought, but it can also be 'rented' or gained access to as part of a subscription to a piece of software.
- Created data. An organisation can combine some bits of data to create a brand new piece of data. For example, a company might label some customers as VIP (very important) based on how often they purchase their products. Algorithms can be used to create predictions about people based on data from their past.