×î×¼µÄÁùºÏ²ÊÂÛ̳

Simon Knowles, ×î×¼µÄÁùºÏ²ÊÂÛ̳’s Head of Coordination (SDGs), describes how the ×î×¼µÄÁùºÏ²ÊÂÛ̳ SDGs Initiative is improving the accuracy and extent of measuring SDG-related activity at ×î×¼µÄÁùºÏ²ÊÂÛ̳.

2021–22 is the second academic year that ×î×¼µÄÁùºÏ²ÊÂÛ̳ has reported on the extent of SDGs-related activity across the university.

Since last year, we have improved the accuracy of how we are measuring our SDGs-related teaching and research. We have also started to measure the extent of our students’ extra-curricular activity, as well as aspects of our external partnerships that are addressing the SDGs.

We are aware some of our measures could be improved: we set out some of the caveats to our methodologies below. We continue to refine how we measure SDGs-related activity by learning from others in the sector and beyond. We will also identify further metrics to illustrate how our community is helping to achieve the Goals.

SDGs-related teaching activity at ×î×¼µÄÁùºÏ²ÊÂÛ̳

For the second year, we classified the descriptions of the 6,797 taught modules in ×î×¼µÄÁùºÏ²ÊÂÛ̳’s online module catalogue by Goal using , a multilingual open-access tool jointly developed by the UN Development Programme SDG AI Lab and research and policy analysis centre PPMI.

Combining several existing sets of SDG categories and augmenting them with additional keywords, OSDG compiled a set of SDG-relevant terminology. The list of original sources is available on OSDG’s website. For last year’s report, OSDG searched for keywords in the module descriptions and attributed an SDG to them if the descriptions contained two or more keywords for that SDG.

OSDG has further improved its methodology since then, by integrating machine learning models that predict the preliminary SDG labels, for them to be verified through the ontology/keyword matching. The updated approach is now able to identify modules that may have been missed previously, so the results should be more accurate than last year’s report.

The methodology relies on module leads using SDG keywords, which many may not have done – they were unaware their description would be mapped – so the number identified is likely to be an underestimate.

Similarly, longer module descriptions have higher chances of being identified as relevant to an SDG simply because they tend to use a wider vocabulary. Figures 1 and 2 (page 9–10 of the report) show the numbers of SDG-related modules by ×î×¼µÄÁùºÏ²ÊÂÛ̳ faculty and by SDG. Ìý

SDGs-related research activity at ×î×¼µÄÁùºÏ²ÊÂÛ̳

Research publications

This year, to broaden our search for SDGs-related research, we mapped ×î×¼µÄÁùºÏ²ÊÂÛ̳ research activity related to each SDG by combining the publications found in database (using keyword searches) and in Clarivate’s Web of Science database (as categorised by their ); last year we only used Scopus. Elsevier also improved the accuracy of their keyword searches this year.

To be included, a paper had to be present in the Scopus or Web of Science databases, with any duplicates removed from the total. We then attributed each paper to ×î×¼µÄÁùºÏ²ÊÂÛ̳ faculties by matching to ×î×¼µÄÁùºÏ²ÊÂÛ̳’s internal publications database by DOIs.

A paper was counted once per faculty (even if it had multiple authors within a faculty), but could be counted in more than one faculty if it has co-authors in multiple faculties. It could be counted in more than one SDG if it is given multiple SDG classifications by Scopus and InCites. Figures 3 and 5 (pages 12 and 16 of the report) show the number of SDGs-related publications by ×î×¼µÄÁùºÏ²ÊÂÛ̳ faculty and by SDG.

SDGs-related research impact case studies

For the first time, we also used OSDG to classify the impact case studies we submitted to the UK Research Assessment Framework (REF), see Figure 4 (page 15 of the report).

Top 10% most cited and international research collaborations

The lists of publications for each SDG (combined from the Scopus and Web of Science databases) were imported as custom datasets into InCites for citation and collaboration analyses.

The percentage of ×î×¼µÄÁùºÏ²ÊÂÛ̳ publications in the top 10% most cited for all research of similar papers was calculated by comparing citations with ‘similar papers’, referring to similar Web of Science subject categories, years, and document types (e.g. articles and reviews).

International research collaborations were measured by the percentage of publications with at least one co-author from a country outside the UK.

Policy citations

Policy citations were sourced from Overton, an index of policy documents, guidelines, think tank publications and working papers, which collects data from more than 1,000 sources worldwide.

Student extra-curricular activity addressing the SDGs

For the first time, we mapped how many of our students were engaged in activity outside of their course which was contributing to the SDGs. We surveyed student societies asking them to state which of the SDGs their activity was supporting. For those societies that didn’t respond, the ×î×¼µÄÁùºÏ²ÊÂÛ̳ SDGs Initiative made that judgement on their behalf, based on information provided by the societies.

The Students’ Union ×î×¼µÄÁùºÏ²ÊÂÛ̳ Volunteering Service did the same for the student-led volunteering projects. We used the number of students who were members of those societies and projects that were judged to be supporting an SDG as an indicator of the extent of student involvement in each Goal.

The figures rely on the subjective judgement of staff and students at ×î×¼µÄÁùºÏ²ÊÂÛ̳. The numbers also reflect particular a moment in the year, while membership numbers fluctuate during the academic year, and students may be members of more than one society or project. Figures 6 and 7 (pages 18–20 of the report) shows the numbers of students involved in addressing the SDGs through their student societies and volunteering projects.

External partnerships activity addressing the SDGs

For the first time, we also mapped the extent to which two types of external partnerships were supporting the SDGs. With ×î×¼µÄÁùºÏ²ÊÂÛ̳ Innovation & Enterprise we measured how many of the student startups supported by the ×î×¼µÄÁùºÏ²ÊÂÛ̳ Entrepreneurship Hub were addressing each of the SDGs (see Figure 8, page 29 of the report). Each startup was asked to identify which (if any) of the SDGs they thought they were addressing.

Similarly, working with the ×î×¼µÄÁùºÏ²ÊÂÛ̳ Office of the Vice-President (Advancement), we measured solicited philanthropic income received by ×î×¼µÄÁùºÏ²ÊÂÛ̳ in support of specific SDGs (see Figure 9, page 30 of the report). OVPA assigned each gift to one ‘lead’ SDG to avoid duplication and double counting.