The Data Hazard labels make a great resource for teaching, and this has been a popular use for them so far. This page contains information that might be useful if you are planning to include the Data Hazards in your teaching, and is based on student seminars that we delivered at the University of Bristol. The format is generally aimed at undergraduate STEM students in a seminar, but could easily be adapted to suit the context. At the sidebar you will also find an example of the prompt page we gave to students to guide them through the session.

The Data Hazards resources have also been used for undergraduate and masters students studying data storytelling, digital health and data science at other institutions.


This teaching example is adapted from the general Data Hazards Workshop format, which is used by those who want feedback on specific research ideas. Here we keep the reflective discussion format of the workshop, but suggest the lecturer comes up with a case study to suit their students.

We suggest starting with a 10-15 minute presentation to introduce the Data Hazards to the students, along with the reasoning of why they exist. The lecturer should then present a case study that the students can discuss in small groups. Prior to the discussion, students should have the opportunity to ask clarifying questions about the project.

We provided a summary page for students too, to use during the seminar.


17:00 - 17:05 - Settling in
17:05 - 17:15 - Background presentation (see link to slides below) 17:15 - 17:20 - Time to get into small groups
17:20 - 17:25 - Case study presentation by lecturer
17:25 - 17:30 - Small group discussions, with lecturer walking around the groups to assist 17:30 - 17:45 - Larger group feedback, identifying trends and voting on Hazards
17:45 - 17:50 - Wrap up

Preparation Required#

Case Study#

The teaching team should have a case study prepared that suits the background and level of their students.
The case study should be about a specific application of data science to a problem. To be clear, this could be anything from simple statistical analysis or creation of a dataset, all the way to a highly complex neural network design - it all depends on your students.

We recommend you include as many of the following details in any case study description as they’ll help students to apply the Hazards:

  • The purpose of the research, e.g. what problem the research is solving if any?

  • What data is used and how that was collected

  • What kinds of methodology are used (e.g. stats, algorithms, machine learning).

  • Who is involved in the research, e.g researchers, communities, sponsors?

  • Potential applications/routes to impact

Interactive questions#

We used slido to ask questions to the entire group about their application of the Data Hazards. This requires a few minutes of setup. Alternatively, the questions could equally be done by a show of hands, in which case you might want to focus your attention on 1-3 relevant Hazard labels.

We usually ask:

  • Following the discussion, which Hazards do you think apply to the case study? (Multiple choices: choose as many as you like)

  • Did you change your mind about which Hazards applied during the discussion? (Yes/No)

  • What have you taken away from the group discussion?


You can find the slides we have used to present the background information about the Data Hazards in a seminar at this link. There is also a preview of the slides below.

Tell us you if used these resources#

If you used these resources in your teaching, you are under no obligation to do so, but we would be so pleased if you could send us a brief email to let us know. We’d love to hear what case study you used and how the students reacted, and if there’s anything we could do to make these resources easier to use. You can read examples of how other people used them on our Examples page.