 Hello everyone my name is Kira Keenan and today I'm going to present on a package created by Dr. Neil Hadaway and myself. The package is called EvieMapper and it produces interactive bubble plots of evidence. So EvieMapper is particularly useful for a type of evidence synthesis called an evidence and gap map. This type of evidence synthesis relies upon systematic and scoping review methods. The search strategy in an evidence and gap map is typically designed to be comprehensive but not necessarily exhaustive. One of the main features of an evidence and gap map is providing an overview of the research that exists and very importantly then where the gaps are on a specific topic. Unfortunately today I don't have a lot of time to go into all the benefits of an evidence and gap map and but I will focus today on the benefits of visualizing research this way and through the EvieMapper package. So through mapping and visual presentation of all the existing and emerging evidence on a particular topic across relevant outcomes it's possible to graphically highlight the gaps in evidence and the areas in which there is sufficient research for evidence synthesis. There's numerous benefits of this but one main benefit is that key stakeholders can use it for their various needs. So a funder could quickly access the map and assess the areas where there's either a saturation of evidence or to see where there's a gap in knowledge to direct the much needed resources towards the areas in which we don't know much. Also policy makers can access the map to see where enough robust information exists to inform policy and practice decisions and then members of the public so depending on which members of the public are most interested in your research area so there could be patients or parents or courage but they could all quickly access the same information which is of particular relevance to them and then for me as a researcher evidence and gap maps are really useful because it minimizes the the research waste which occurs due to duplication of effort. Interactivity in data visualization is particularly useful for simple science communication because it helps the reader access the information at their own pace and depending on their own needs. So every mapper plots three dimensions of categorical data so the x-axis the y-axis and then the bubble color and one numerical variable which is corresponding to the bubble size which really condenses a lot of information into a visual representation and into just one single plot but readers often want to see the studies which also correspond to each point and so instead of cross-referencing tables of studies every mapper will allow a user to hover over and click on the bubble which lets them deep dive into the data behind the plot and the interactivity is provided through hyperlinks to external web pages and my silver tool tips containing more detail. So this example bubble plot shows three dimensions so the outcomes are along the y-axis and then the context so in this study it's type of educational institution and their plot along the x-axis and then the location and so where this study was conducted corresponds to the color of the bubble and then you can see some of these bubbles are of varying sizes and that's related to a quantitative variable which is the number of studies and if you hover over any of these bubbles and you can click on which updates the HTML table that you see along the bottom and so those are filtering the studies which are then embedded in an iframe and which you know filters and depending on the bubble that you click on. So this slide is just covering the main components of the package so the data prep for the summary data needed across three variables uses the count function and the base plot is a j-on point plot. Tool tips are provided by converting the plot to a HTML output in plotly and hyperlinks are then appended as JavaScript to the HTML output afterwards and the table that we presented in the previous slide is produced using the Revis package that uses a hidden column in the table to filter results for each bubble but actually any other link can be embedded in the bubble too. So as I'm sure you appreciate I covered quite a lot of points there in a really short space of time but Neil and I are both open for questions or emails afterwards and you can explore the package in much more detail by visiting Neil's GitHub page which is provided on screen now. So thanks a million everyone and hope you enjoy the rest of the conference.