Abstract
The Opinion Development System (ODS), an online system for consensus and opinion development was developed as part of a collaboration between Northumbria University and the National NHS-E Community of Action for Trauma Informed Care (TIC). The aim was to determine if an online tool can help a group of people in an organisation to reach consensus (agreement) on some statements relevant to their field. Gathering a range of opinions and building consensus on these topics would inform the direction and strategy for the organisation.
A discrete dynamic opinion development method is implemented in ODS using an Erdos-Renyi network as the underlying communication network to support opinion and consensus development towards achieving a common shared goal among users in a multi-agent system. Erdos-Renyi networks are classical random graphs of nodes with independent edges mimicking connections between nodes or individuals. Each ODS user is assigned to a network comprising a subset of the overall number of users and only sees the views of that subset.
Individuals working in key leaderships positions in health care TIC practice and recovery were recruited to trial the ODS. Five-day trials were conducted. In each case the system successfully established a network of connections between the participants (nodes), who then used it to view the opinion of others in their network on a topical issue and to regularly express their own views. The intention being to evaluate whether the sharing of views over time in such a network supports opinion and consensus development. The system did work as designed with users able to view the opinions of others and enter their own. Results indicated that there were changes in opinions which moved away from consensus (trial one) and towards consensus (trial two). However, there were an insufficient number of interactions to adequately measure consensus development fully (possibly due to staff workload) and to inform direction and strategy. Further research to consider ways of engaging participants and achieving more and sustained interactions is required with larger groups.h
A discrete dynamic opinion development method is implemented in ODS using an Erdos-Renyi network as the underlying communication network to support opinion and consensus development towards achieving a common shared goal among users in a multi-agent system. Erdos-Renyi networks are classical random graphs of nodes with independent edges mimicking connections between nodes or individuals. Each ODS user is assigned to a network comprising a subset of the overall number of users and only sees the views of that subset.
Individuals working in key leaderships positions in health care TIC practice and recovery were recruited to trial the ODS. Five-day trials were conducted. In each case the system successfully established a network of connections between the participants (nodes), who then used it to view the opinion of others in their network on a topical issue and to regularly express their own views. The intention being to evaluate whether the sharing of views over time in such a network supports opinion and consensus development. The system did work as designed with users able to view the opinions of others and enter their own. Results indicated that there were changes in opinions which moved away from consensus (trial one) and towards consensus (trial two). However, there were an insufficient number of interactions to adequately measure consensus development fully (possibly due to staff workload) and to inform direction and strategy. Further research to consider ways of engaging participants and achieving more and sustained interactions is required with larger groups.h
| Original language | English |
|---|---|
| Title of host publication | VIRTUAL 21ST UKSS INTERNATIONAL CONFERENCE |
| Publication status | Published - 18 Jun 2021 |
Keywords
- consensus, human networks, multi-agent systems, trauma informed care