The Web provides abundant possibilities to connect to information and people when trying to grow your learning network, and it is vital that we evaluate the reliability of them effectively.
Many of the more reliable sources will base their information or conclusions on data.
Data can be any words, numbers, images or sounds without a context - in other words in their raw form.
Data is found in the form of reports, results sections of articles, appendices, spreadsheets, tables, computer log files, audio files…etc.
However, even data from the best of sources can be misleading. This can be intentional or accidental, and so when growing our network by connecting to new data we need to remember that:
- Data can be misleading because of the way it is visualised. Visualisations may include charts, graphs, plots, maps, tables…etc. Some visualisations are confusing. Others are a deliberate attempt to trick us
- Samples may be too small or biased
- Data for one time may be presented without taking into account changing trends or simply be out of date (e.g. data from the last UK census, which is commonly used for academic research purposes, is only updated every ten years)
- Correlation (two data sets showing a similar trend over time) may be confused with causation (the reason why a trend occurs)
Although problems like these predate the Web by many decades, as the Guardian suggests, they become even more important as the Web enables more people to access data and data outputs.
One way of getting more value out of data on the Web is to combine different sources. This can be as simple as comparing different but related figures.
Despite the amount of available data and the power of these tools, most experts accept that the public use of data on the Web is very limited. There is a growing body of research, summarised in a Special Issue of the Journal of Community Informatics, which explores why people do not take advantage of data on the Web and how any barriers can be minimised.
Sometimes this is seen as a problem in our education system leading to calls for increased data literacy. But others point out that it is more complicated. It is not easy to identify a specific set of skills needed to find and use data on the Web.
Nevertheless, being able to access, analyse and question the data we connect with can help us to grow our learning network in a productive and reliable way.
Have you ever found data, or data visualisations, to be confusing, misleading or inaccurate?
Share your experiences and any tips for how you dealt with it in the comments below.
© University of Southampton 2017