Beth Astridge (Project Archivist, UKPA) and Clair Waller (Digital Archivist), University of Kent Special Collections and Archives.
Contemporary collecting and mapping other collections
This is our final blog post in a series from the University of Kent Special Collections & Archives describing some of our ongoing work to implement robust workflow and processes for the acquisition and management of born digital records, driven by our work to establish the UK Philanthropy Archive.
In this blog we discuss how we embarked on our first attempt at collecting social media content and how we are supporting researchers by mapping other collections of potential interest using Google Maps.
Using tweets to record philanthropic responses to COVID 19
Once the pandemic had hit the UK in March 2020 and our team had settled in to working at home, our thoughts turned to how we could proactively collect the philanthropic community response to COVID 19, which was often communicated using social media, especially using Twitter. We were interested in how we could ensure that the tweets produced by organisations in the UK Philanthropic sector, in response to COVID-19, were captured and preserved, and how we might be able to make these tweets accessible to people researching this area.
We decided to trial collecting tweet data using the Twitter API and Python. This was a totally new way of collecting material for us and is an ongoing learning process. We had virtually zero experience in using Python (or any coding language for that matter), but felt it was a good time to give it a go and develop our skills. A quick internet search turned up a number of blog posts and github sites detailing how Python could be used by social scientists for just this sort of thing, so we got to reading and experimenting (we found this script by yanofsky on GIthub particularly helpful).
Alongside experimenting with our code, we also started exploring exactly what data we wanted to collect, and any issues we could foresee in terms of storage and access, and of course the ethical implications. We had a look at resources that were already available, such as those shared by Documenting the Now who have a number of tools and information to help those embarking on contemporary collecting using social media. We also discussed whether we needed to draw up a formal policy for social media collecting but felt that we were at an exploratory stage and therefore would not be able to define our policy at this point.
We drew up a list of Twitter accounts that would be of interest to the UKPA and set parameters in the script to collect tweets by these users. We considered collecting individual accounts and corporate accounts, and decided that individual accounts presented more of a risk that the content of tweets could include information and data that would require additional protection under the data protection act and GDPR legislation. Therefore, for this dataset, we decided to focus on corporate accounts of philanthropic organisations and initiatives, such as the Institute for Voluntary Action Research (IVAR) and the London Funders network.
After a few weeks of editing and adapting our script, we finally had something up and running. We did hit a few bugs along the way, but luckily we have very supportive colleagues who work in IT that assisted us when this happened! We now have a working script that can collect tweets from a specified user within a specified time period. This script can be run regularly to collect data, which can then be saved into a database.
Any data we have collected so far is not yet available for public access. We’re still in the process of investigating exactly how we will make this data available in a useful and, more importantly, ethical way that fits within the confines of Twitter's developer agreement and policy. Our current thinking is that as public sharing is restricted to sharing the Twitter IDs only – rather than the full content of the tweet - and we cannot replicate the look and feel of Twitter ourselves, we will point users to a “rehydration” tool so they can access the data. We have not selected which tool we will use yet – but it would have a graphical user interface that will make viewing the data easier. We intend to develop guidance for researchers in how best to use our chosen rehydration tool to access the data, and search for keywords and terms.
We’d love to hear from any other small institutions that have embarked on this type of contemporary collecting and share our practice! Email details are below.
Using Google maps to provide info on philanthropy collections
For the UK Philanthropy Archive, we are actively collecting the archives and papers of philanthropists, trusts and foundations and related philanthropy sector organisations – to develop a research resource for the study of philanthropy.
Alongside our collections, we know there are many collections relating to philanthropy and charitable giving that have been deposited elsewhere. For example, at a local records office - especially if a geographical area was a focus of giving for an organisation. Alternatively, at another university, who might perhaps have an alumni relationship with a philanthropist, or at a specialist archive service for collections that have a particular subject specialism.
We want to help researchers interested in philanthropy to understand where these wider collections are held, and therefore have created this map of collections using Google maps. This map aims to identify locations of relevant archives, provide links where possible, and indicate whether they are open for research. Please drop us an email at specialcollections@kent.ac.uk if you know of other philanthropy archive collections in the UK that we can add to our map!