Why Social Workers Need Data Science
by Gina Griffin, MSW, LCSW
Social work embodies a proud tradition that was born in the settlement houses of London and Chicago in the late nineteenth century (Harkavy & Puckett, 1994). We have progressed and evolved radically since that time. Although we have made progress in gaining recognition as a helping profession, we sometimes struggle to remain current in areas related to technology.
We have integrated the use of data into practice in a variety of ways, such as computerized charts, provision of therapeutic services via telehealth, and the use of wellness apps. However, social work practitioners are often struggling to make the grade in developing general technical skills.
Traditionally, social workers have been fairly resistant to technology. We tend to be late adapters. Sometimes, this has to do with the fact that we tend to skew older, as a whole, and developing the new skills can be intimidating. Maybe we’re not getting the training that we need from our agencies (Goldkind et al., 2016; Bullock & Colvin, 2015). Often, this is because we fear that embracing technology will change the field of social work. Social workers take pride in developing a practice that values face-to-face interaction with clients. So, there is the sentiment that we lose something essential as a profession when we integrate technology into practice (Steyaert & Gould, 2009).
But the world is changing, and social work has to keep up. In this rapidly evolving environment, data science is a multifaceted skill that will help us to achieve multiple goals in multiple settings. Data science has been defined as the ability to gather, manage, translate, and communicate data in an effective way (Berkeley School of Information, 2020). Some of the tools are quite complex, such as machine learning. Others, such as exploratory data analysis and data visualization, can be used in direct practice, academia, policy, and research settings on a daily basis. Although not every social worker needs to aspire to become a data scientist, data science tools can certainly benefit every social worker.
The first manner in which data science can provide benefit is by offering the foundation we need to understand negative and positive ways in which data and technology affect the populations that we serve. The National Association of Social Workers states, “The primary mission of the social work profession is to enhance human well-being and help meet basic and complex needs of all people, with a particular focus on those who are vulnerable, oppressed, and living in poverty” (NASW, 2020).
To that end, we need to develop a basic understanding of how and why tools such as predictive policing and recidivism algorithms often negatively impact the marginalized communities with whom we work. Practices related to big data often negatively impact already marginalized communities at a higher rate. As examples, algorithms that decide mortgages have been known to award lower credit scores to names that sound African American by 71% (Chandler, 2017). Predictive policing tools seem to create confirmation bias; those communities can become over-policed, which can lead to physical and emotional strain for those who live there (Lerman & Weaver, 2014). We are relying more and more heavily on these types of technologies and algorithms. So, it is becoming necessary for social workers to understand how these issues and tools are affecting our clients, and also to understand where and when we might intervene. Ideally, social workers, or other practitioners, could be involved when these tools are developed, or policies are written. Hopefully, this will become standard practice.
The second manner in which data science can be of use to social workers is in daily practice. I like to call this “Everyday Data Science.” As practice in the organizations in which we work is becoming more and more data driven, we need to be able to understand how to leverage that data to provide the best outcomes for the vulnerable and marginalized communities we serve. We collect data from a variety of sources throughout the day, and we need to be able to effectively collect and interpret that data, to provide better outcomes for our clients and communities.
Learning data science skills is part of a process in which we can learn to see with fresh eyes ways we are already collecting data, which we might not have previously realized. It also helps us to begin to think of new and useful ways to put that data to use. As an example, if we are engaged in measurement-based care in a mental health setting, measures such as the PHQ-9 (to report depression symptoms) and the PCL-5 (to measure trauma symptoms) are used on a daily basis. If the combined scores of many of the clients are trending downward, then the case might be made that the services that are being offered are effective. Additionally, you might be able to make the case that this should lead to more funding for services, or more staff members to provide additional services. Even mundane data such as gas and mileage logs for a fleet of agency cars can become a source of rich data. You might simply use data science skills to provide more interesting monthly mileage reports. However, you might also be able to spot trends in the community, such as areas in the community where clients may tend to cluster. Data science skills can ultimately help us to more effectively interpret what this type of data is telling us, and to communicate that to both staff and management.
Right now, many of us use Excel spreadsheets to make these types of calculations, and that might be just fine for you, where you are. However, more advanced data science skills offer versatility. A tool such as R Programming language offers options such as charts and graphs, which managers use on a regular basis to illustrate information. It also offers the ability to perform more advanced statistical functions, such as using ANOVA to manage your budget; to create written reports and slides for presentations using Bookdown; and to explore data for research with basic but powerful Tidyverse skills. (Note: The Tidyverse is a suite of packages that work together within R Studio to manage data, from collection through communication.)
Hopefully, it is becoming clear that data science is a next-level skill set that can begin to add a richness to ways social workers engage with data. Data may sound dry and uninteresting. However, please remember that the data always represent the people we serve, and that it is a means to find better ways in which we can serve them.
References
Berkeley School of Information. (2020). What is data science? https://datascience.berkeley.edu/about/what-is-data-science/
Bullock, A. N., & Colvin, A. D. (2015). Communication technology integration into social work practice. Advances in Social Work, Vol. 16 No. 1 (Spring 2015), 1-14.
Chandler, A. (2017). The racist algorithm? Michigan Law Review, Vol. 115(6).
Goldkind, L., Wolf, L., & Jones, J. (2016). Late adapters? How social workers acquire knowledge and skills about technology tools. Journal of Computers in Human Services. DOI:10.1080/15228835.2016.1250027
Harkavy, I., & Puckett, J. L. (1994). Lessons from Hull House for the contemporary urban university.
Lerman, A. E., & Weaver, V. (2014). Staying out of sight? Concentrated policing and local political action. Annals of the American Academy of Political and Social Science, 651(1), 202-219.
National Association of Social Workers. (2020). Why choose the social work profession? https://www.socialworkers.org/Careers/NASW-Career-Center/Explore-Social-Work/Why-Choose-the-Social-Work-Profession
Steyaert, J., & Goud, N. (2009). Social work and the changing face of the digital divide. British Journal of Social Work (2009) 39, 740–753 doi:10.1093/bjsw/bcp022
Gina Griffin, MSW, LCSW, has been a trauma social worker, working with veterans, for a decade. She will complete her DSW, focusing on social work practice research, in October of 2020.