Image
Abstract: Data Storytelling: Tools for Ethics and Ease
McDowell (2023) presents structures for data storytelling, emphasizing classic narrative structures of “continuity, discover, and transformation” (para. 12). Libraries have many stories to tell and the data to tell those stories. However, ethical issues around data use abound. This paper presents a brief look at the data ethics issues of telling the library story, discusses some free tools for libraries to create data stories, and details ways to group data to protect patron privacy while still telling compelling library stories designed to spur community action.
Data use is a fraught issue. There are no clear guidelines on ‘thresholds’ of acceptable data sharing. Situations range from the relatively free, if unscrupulous, reign of private sector data analysis (e.g. Cambridge Analytica’s purchase of Facebook data for political campaigning), to specific and restrictive use of human subjects data (e.g. NIH All of Us). This is further complicated by the reality that according to United States law, collected data cannot be copyrighted as they are considered “facts” (Supreme Court, 1991). Once data are transformed into some form other than collected facts – such as a database or a story – they are subject to more protection because an element of creativity was required.
Data manipulation can be complicated. While various data storytelling toolkits exist (UIUC, IMLS), many use R or other computationally intense languages to process data. Furthermore, some know-how is required to match the right visualization to data in a way that is truthful, shows the story, and inspires the user of a visualization. Thus, the intellectual burden of creating of balancing protection of patron privacy, truthfully telling stories, having enough expertise to collect and protect data, and to visualize something meaningful. Fortunately, free and easy-to-use tools are becoming increasingly popular, and this session will introduce some of those tools.
Storytelling is an act of power, and it comes with potential harms to the people whose story is being told. McDowell (2023) advocated for the use of a community hero, or amalgam of heroes to share how the library helped the hero(es) overcome some obstacle. In smaller communities, protecting the identities of patrons, especially patrons with protected statuses (i.e. LGBTQ youth in states with anti-LGBTQ book bans), becomes a delicate balance. While the U.S. Census Bureau uses more complex techniques like data swapping to protect resident information, there are simpler strategies afforded by tools like Tableau for librarians to reduce identification potential in their own data.
Data use is a fraught issue. There are no clear guidelines on ‘thresholds’ of acceptable data sharing. Situations range from the relatively free, if unscrupulous, reign of private sector data analysis (e.g. Cambridge Analytica’s purchase of Facebook data for political campaigning), to specific and restrictive use of human subjects data (e.g. NIH All of Us). This is further complicated by the reality that according to United States law, collected data cannot be copyrighted as they are considered “facts” (Supreme Court, 1991). Once data are transformed into some form other than collected facts – such as a database or a story – they are subject to more protection because an element of creativity was required.
Data manipulation can be complicated. While various data storytelling toolkits exist (UIUC, IMLS), many use R or other computationally intense languages to process data. Furthermore, some know-how is required to match the right visualization to data in a way that is truthful, shows the story, and inspires the user of a visualization. Thus, the intellectual burden of creating of balancing protection of patron privacy, truthfully telling stories, having enough expertise to collect and protect data, and to visualize something meaningful. Fortunately, free and easy-to-use tools are becoming increasingly popular, and this session will introduce some of those tools.
Storytelling is an act of power, and it comes with potential harms to the people whose story is being told. McDowell (2023) advocated for the use of a community hero, or amalgam of heroes to share how the library helped the hero(es) overcome some obstacle. In smaller communities, protecting the identities of patrons, especially patrons with protected statuses (i.e. LGBTQ youth in states with anti-LGBTQ book bans), becomes a delicate balance. While the U.S. Census Bureau uses more complex techniques like data swapping to protect resident information, there are simpler strategies afforded by tools like Tableau for librarians to reduce identification potential in their own data.