Data Sources and the Digital Footprint

Every piece of scientific knowledge starts with data, and that data has to come from somewhere. A data source is simply where your data comes from. Scientists divide sources into two big groups. A primary source is data you collect yourself, by taking measurements, running an experiment, sending out a survey or reading a sensor. A secondary source is data that someone else collected, such as a published study, a government database or a textbook table.

Neither type is better than the other. Primary data lets you control exactly how it was collected, but it is slow and limited to what one person or team can measure. Secondary data gives you huge amounts of information instantly, often spanning decades or whole countries, but you have to trust that whoever collected it did so carefully. Good scientists often combine both: they collect their own primary data and then compare it with secondary data to check whether their results fit the bigger picture.

The skill is choosing the right source for the question. To find out how a plant in your classroom grows, primary data is best. To find out how rainfall across New South Wales has changed since 1900, no student could collect that alone, so a trustworthy secondary source is the only practical option.

Australia is rich in open data, secondary sources that anyone can download for free. The Bureau of Meteorology publishes weather and climate records going back more than a century. The Australian Bureau of Statistics, or ABS, runs the national census and releases population, health and employment data. The site data.gov.au gathers thousands of government datasets in one place, while CSIRO and Geoscience Australia publish data on oceans, climate, minerals and earthquakes.

A lot of this data is generated by sensors and instruments rather than people. Weather stations, satellites, ocean buoys and even fitness trackers record measurements automatically, around the clock. The result is big data: datasets so large that you cannot read them by hand and need computers to find the patterns. A single satellite can produce more readings in a day than a person could record in a lifetime.

Whenever you use a secondary source, you must check its reliability. Ask who collected it, when, how and why. A dataset from the ABS has clear methods and is checked by experts. A figure shared on social media might have no source at all. Good science means tracing data back to where it really came from.

Every time you go online you leave a trail of data called your digital footprint. It includes the obvious things you choose to share, like posts and photos, but also a huge amount of data collected automatically: the searches you type, the links you click, your location, your purchases, which apps you open and how long you watch a video before scrolling on.

Companies collect this data to power recommendation algorithms and targeted advertising. That is why a video app seems to know what you want to watch next, it is reading patterns in your footprint. But the same kind of data can be used for genuine science. During the COVID-19 pandemic, anonymised mobility data from phones helped researchers see whether people were staying home, and patterns in Google search trends have been used to estimate where flu outbreaks are starting before hospitals report them.

This is why your footprint is both useful and sensitive. The same data that helps scientists track a disease could also reveal where a single person lives, works and travels. Understanding that you are constantly generating data, and that this data has real value, is the first step to using technology wisely.

Because data about people is so powerful, scientists must think carefully about ethics and privacy. The key questions are: who owns this data, did the people give consent for it to be used, and could anyone be harmed by sharing it? Data about a single named person is far more sensitive than a count of how many people did something.

One of the most important tools is anonymisation: removing names and other identifying details so the data describes a crowd rather than an individual. Researchers also aggregate data, combining it into totals and averages, so no single person can be picked out. There is a clear line between using anonymised data for science, such as tracking a disease, and surveillance, which means tracking identifiable individuals without their knowledge or permission.

Ownership matters too. When you use an app, you often agree to let the company store and use your data. That data may be valuable, but you usually do not control where it goes next. Being a good scientist, and a careful citizen, means understanding these trade-offs and choosing data sources that are both reliable and ethical.