A literature review was conducted to map and understand the research base surrounding loneliness. Research was collected using the resources made available through the Campaign to End Loneliness website and wider searching of free to access, open-data resources was also conducted. This led to 129 research articles being collected from peer-reviewed journals, government and non-government policy projects and online journalistic sources. The full range of data collected has been stored on a google docs folder, which can be accessed at this link:
Upon collection the source documents were split into their constituent strings and combined in a central database. This was done using bespoke ‘Simplexity Analysis’ processes, that analysed the combined string set for the source data and sampled the most frequently occurring keywords (note - this sampling involved taking the top 25 frequently occurring keywords (known as 'L1' Keywords), and then taking the top 20 keywords associated with each of the L1 keywords (known as 'L2' keywords). Using this sample enabled a systems map of the most frequently occurring keywords (and relationships between them to be visualised. This visualisation gives an overview of the main themes and concepts contained in the data. The data was visualised using an open source software platform called ‘Gephi’.
Using Gephi scripts and structured metadata for the source string data enabled us to produce a ‘meta-analysis’ of a wide range of literature research (based on a technique known as ‘grounded theory)’. This has enabled us to understand themes and ideas in the current research and highlight new and emerging ideas. As a result we produced two maps.
Rich data map - this is a visualisation (displayed through gephi) - of the top 25 keywords and their associated top 20 keywords, with relationships between strings charted. This is a powerful means of visualising a large range of data, without human bias as the diagram directly reflects the word frequency scores. Additionally, using gephi enables an analyst to move from the map, through to the actual data (i.e. strings and sources). This enables high level deductions to be tracked back to the source data, which can be useful for policy planning. The full rich data map is available at the following link:
Main findings map - this is a visualisation of the main themes, loosely based on the frequency of the topics occurring. Due to the complexity of the data contained in the rich data map, the main findings map is offered as a stylised representation of the data; its main purpose is to be an early visualisation of research themes and questions surrounding loneliness. Such a map can be more subject to bias (it is an analyst’s interpretation of the data in the rich data map), but (we believe) it's useful at the early stage of project as it provides an simple reference for potential themes and topic areas for researchers seeking to better understand how to tackle the complex, social phenomenon that is ‘loneliness’. This map also contains links (where appropriate) to the source data contained in the project archive, where relevant papers, blogs and newspaper articles highlight interesting concepts and/or emerging research ideas are kept. The full map is available at the following link:
A narrative (made using the top level map) has been produced, using the top five themes to explain current ideas and topics in Loneliness research, forms the basis of the ‘Understanding Loneliness blog’ which is available here.
We hope providing our maps, the data and the method we used to produce this analysis gives a helpful way of understanding current loneliness research and adds to the open-resources that help us better understand this complex, yet very simple, problem.