| Title | UKRN: making research more replicable, reproducible, and robust |
| Category | Research |
| Overview | Three large many-lab projects have found that replicability, reproducibility, and robustness across the social sciences are still much lower than we would like (https://www.nature.com/collections/idajfifcfg). Specifically, collecting new data using the same methodology leads to the same conclusion in only ~55% of cases (replicability). Re-analysing existing data using the same statistical approach reproduces the exact original findings in only ~53% of cases (reproducibility). Using different statistical approaches on the same data leads to the same conclusion in 74% of cases (robustness). Collectively, these issues can substantially undermine trust in science. We will discuss these findings and explore ways to make our research more replicable, reproducible, and robust.
Anyone interested is welcome to join the session. The primary target audience is researchers (including PGRs who can book via Proficio) and research-enabling staff. |
| Learning outcomes | - Explore recent findings around replicability, reproducibility, and
robustness in social sciences.
- Discuss the importance of replicability, reproducibility, and robustness.
- Explore ways to make our research more replicable, reproducible, and
robust.
|
| Presenter | Paul Hanel |
| Location | Zoom |
| Campus | Colchester Campus |
| Date and time | 14:00 to 15:00 Tuesday 26 May 2026 |
| Duration | 01:00 |
| Available places | 50 |
| University of Essex employee | Use the link provided above to book your place via HR Organiser. |
| Partner institution | Please discuss with your point of contact at your Partner Institution in the first instance. |
| Student Union contractors; PhD students | Please contact the Organisational Development team. |