The Ecology of Algorithmic Literacy: Extending Practices Beyond Representational Media Literacy

Abstract

The significant effect of algorithms on media consumption requires new literacy practices that extend beyond representational media literacy. In this case study, I evaluate traditional media and information literacy instruction in a library setting and argue for the inclusion of algorithmic literacy, which is an essential and foundational skill for source evaluation.

Keywords

Media Literacy, Algorithms, Source Evaluation, Information Literacy

Introduction

Libraries focus on developing information literate individuals who know how to locate, access, evaluate, and use information, skill sets delineated and expounded upon in the ACRL Framework for Information Literacy for Higher Education. Students are taught how to formulate search strategies and evaluate sources using methodologies like CRAAP. Unfortunately, these skills do not adequately prepare students for an information environment driven by algorithms. Algorithms, defined simply as the parameters or steps for solving a problem, present users with a heavily curated environment that is personalized and based on user history, behavior, and location, among other factors (Head et al., 2020; Hobbs, 2020; Bodzag as cited in Cohen, 2018). These same algorithms are often characterized as pervasive, insidious, and opaque (Lloyd, 2019; Cohen, 2018; Head et al., 2020). Furthermore, events like the 2016 presidential elections and the Cambridge Analytica situation—both influenced by or resulting from algorithms–have led to skepticism and loss of faith in media, causing a paradigm shift, to what we now call the “post-truth” era (“post-truth,” 2019 as cited in Valtonen et al., 2019, p. 23). The implications of this abound and have repercussions throughout our democratic society. For this reason, I argue that traditional media and information literacy is inadequate, and that algorithmic literacy is necessary to effectively navigate and participate in this complex information environment.

Defining Algorithmic Literacy

Algorithmic literacy is a relatively new term that is often discussed with or subsumed by other literacies, such as digital or AI (Ridley & Pawlick-Potts, 2021). However, the definition posited by Ridley and Pawlick-Potts offers a clearer understanding of the term and emphases its unique importance:

Algorithmic literacy is the skill, expertise, and awareness to understand and reason about algorithms and their processes; recognize and interpret their use in systems (whether embedded or overt); create and apply algorithmic techniques and tools to problems in a variety of domains; assess the influence and effect of algorithms in social, cultural, economic, and political contexts; and position the individual as a co-constituent in algorithmic decision-making (p. 2).

This definition acknowledges two key points: 1.) Algorithms are foundational and more expansive than technology, and 2.) algorithms reflect the sociocultural environment in which they are created. Additionally, the definition appeals to and recognizes the importance of traditional literacy frameworks. As Ridley and Pawlick-Potts state, “Identifying and acting upon algorithms as a literacy makes them as ‘fundamental as reading, writing, and arithmetic’” (2021, p. 1).

Literacy Practices

Multi-modal media is often analyzed using strategies such as close reading (Hobbs, 2020; Hinrichsen & Coombs, 2013). These traditional literacy practices place the onus or “locus of control” on the consumer/creator of the media, requiring a tangible object—what Nichols and LeBlanc (2021) call “representational media”—to analyze. Indeed, the Center for Media Literacy’s MediaLit Kit five core concepts and key questions ask the consumer to evaluate the media, not the holistic information environment.

So how does one analyze algorithms? Are traditional media literacy practices, such as the the five core concepts, sufficient? Cohen (2018) states

In comparison to media ecology, the algorithm exceeds the screen interaction of the user. Algorithms create feeds based on combinations of thousands of inputs from the larger audience and the specific user actions both on and off the platform. Considering the algorithm as a media environment means to come to an understanding that future platform interactions will disregard legacy consumption methods of collective, meaning mass, media ecologies and move beyond traditional media literacy approaches to deconstruction and analysis. (p. 145).

In other words, algorithms create a media environment that calls for different or at least additional media literacy skills, what Nichols and LeBlanc call “non-local, non-representational, and non-human relations” understanding (2021, p. 395). Literacy practices must address the entire media environment, which includes the algorithm itself, the technology/platform, user behavior, and the sociocultural influences and effects.

As algorithms are “code-based programs that…execute commands,” media literate individuals also need to understand computational thinking, machine learning, and increasingly AI (Cohen, 2018, p. 145). In addition, individuals must decode and analyze emotion, bias, and propaganda, skills traditionally covered in rhetoric and persuasive genre studies (Hobbs, 2020). Lastly, ethical questions surrounding algorithms must be addressed. Sayifa Noble (2018), author of Algorithms of Oppression, found that algorithms perpetuate and reflect the bias of their programmers. There is a cyclical relationship between the user and the algorithm, in that they influence and feed each other. As a code-based program, however, “control is held by those who build and deploy algorithms, not those who use them” or are affected by them (Ridley & Pawlick-Potts, 2021). Case in point, as a human construct, algorithms cannot predict all outcomes, thereby resulting in unanticipated consequences that can and do have disastrous results, particularly when used as predictive analytics (Nichols & LeBlanc, 2021).

Algorithmic Literacy in Libraries

Algorithms affect access to information and seeking behaviors in all contexts, from political spheres, to news, to education. Valtonen et al. called it a “great technological shift…from rule-drive programs [prior to algorithms]—where rules for behavior were explicitly hand-coded by hordes of programmers—to data driven programs—where program behaviors are automatically derived from massive amounts of data (2019, p. 21). And algorithms affect everyone, even students in higher education, who whether knowingly or not, benefit from algorithmic enhancements that foster learning-centered environments (Hobbs, 2020). What is alarming is the lack of knowledge college students have about algorithms. Project Information Literacy (PIL) found that students were aware of how platforms customized their feeds and personalized content but had a profound ambivalence towards algorithms and their effects in general (Head et al., 2020). They lacked a full understanding of what data was collected on them as students and in their personal life and how it was used, especially by large companies such as Google and Amazon, but saw this a tradeoff to the service. And many students were skeptical or critical of news media, demonstrating a lack of faith in journalism as a result of the current proliferation of misinformation. An important takeaway is that students felt they had not been adequately prepared to deal with the algorithmic world. Brodsy et al.’s (2020) study found similar results to PIL, and like other researchers, found that even though students had high media literacy skills their algorithmic literacy that was lacking.

The imbrication between the work of libraries, specifically information literacy instruction, and algorithmic literacy is apparent. The ACRL (2016) defines information literacy as the “set of integrated abilities encompassing the reflective discovery of information, the understanding of how information is produced and valued, and the use of information in creating new knowledge and participating ethically in communities of learning.” For the purposes of this argument, two components of this definition align with and require the understanding of algorithms: reflective discovery of information and the understanding of information production. The ACRL Framework, which details the skills of information literate individuals, acknowledges the information ecosystem in which we live and calls upon librarians to design information literacy curricula that go beyond the representational focus of current media and information literacy practices. Libraries, therefore, are perfectly positioned to the address student deficiencies in algorithmic literacy.

Unfortunately, Project Information Literacy (2020) found that traditional information literacy, with it focus on library research and peer-reviewed journals, inadequately prepares students for the information environment they will encounter post-graduation, a world heavily altered and customized by algorithms. Lloyd (2019) similarly critiques current information literacy pedagogy, stating, “when information literacy education is focused on the operationalisation of skills rather than developing a deeply critical and reflexive approach to understanding and critiquing the conditions which scaffold the operationalisation of information,” students are at risk (2019, p. 1482). This mirrors the criticism leveled against representational media literacy. So, what practices and pedagogy can libraries employ to improve information and media literacy to adequately address algorithmic literacy?

Algorithmic Literacy Instruction

To prepare students for an algorithmically influenced world, libraries should craft curricula that focuses on the sociocultural environment in which algorithms are created and disseminated, and the ethical and moral implications, what Lloyd (2019) calls “reflectivity” (p. 1482-1483). To achieve reflectivity, students should be taught “bias, trust, credibility, opacity, diversity, and social justice” (Lloyd, 2019, p. 1483). As Hobbs (2020) argued, instruction needs to incorporate traditional persuasive genres, the study of propaganda and adversiting, and the evaluation of emotion in media.

These practices are paramount to achieving algorithmic literacy. How does one address these practices in library instruction? I offer the one-shot I teach for a sociology class on race as an example. In this class, students are introduced to algorithms and big data; the terms are defined and demonstrated using Noble’s (2018) research as a framework. Students are shown Google results for women and girls from 2009 and then asked to do a search themselves. A comparison of the two searches produces a dramatic difference: the results from 2009 show a dearth of diversity in any way for either group. The current results are much improved, but not fully representative. Further analysis demonstrates that Google’s search results are often a reflection of society and its views/mores, whether good or bad. For example, doing a search for Black men results in valid and credible resources that address the social/cultural issues facing Black men, but that also may inadvertently perpetuate stereotypes. This is often a startling and eye-opening moment for students as for the first time, students are looking beyond search strategies and source evaluation to cultural implications of algorithmic culture, thus allowing students to expand past traditional information and media literacy to algorithmic literacy.

Conclusion

Algorithms present a unique challenge as they are pervasive and opaque, operating to reflect and create society. These qualities affect a great influence on the information environment and require a broader, more expansive set of practices than what traditional representational media literacy can provide. Similarly, the focus of information literacy inadequately prepares students to deal with our current information environment. What is needed is curricula and practices that address the sociocultural and ethical aspects of algorithms. Libraries are uniquely positioned to offer this instruction as they sit at the intersection of both media and information literacy. Also, given that library instruction is somewhat discipline agnostic, students can develop algorithmic literacy that spans the curriculum and educational experience. With intentionally designed library instruction that moves beyond traditional literacy practices, students will gain the reflectivity necessary to be active, socially aware, and social justice-minded individuals in society.

References

ACRL. (2016, January 11). Framework for information literacy. https://www.ala.org/acrl/standards/ilframework

Brodsky, J. E., Zomberg, D., Powers, K. L., & Brooks, P. J. (2020). Assessing and fostering college students’ algorithm awareness across online contexts. Journal of Media Literacy Education, 12(3), 43-57. https://doi.org/10.23860/JMLE-2020-12-3-5

Center for Media Literacy. (2005). Five key questions that can change the world. www.medialit.org  

Cohen, J. N. (2018). Exploring echo-systems: How algorithms shape immersive media environments. Journal of Media Literacy Education, 10(2), 139-151. https://doi.org/10.23860/JMLE-2018-10-2-8

Head, A. J., Fister, B., & MacMillan, M. (2020, January 15). Information literacy in the age of algorithms: Student experiences with news and information, and the need for change. Project Information Literacy. https://projectinfolit.org/pubs/algorithm-study/pil_algorithm-study_2020-01-15.pdf

Hinrichsen, J. & Coombs, A. (2013). The five resources of critical digital literacy: A framework for curriculum integration. Research in Learning Technology, 21, 1-16. http://dx.doi.org/10.3402/rlt.v21.21334

Hobbs, R. (2020). Propaganda in an age of algorithmic personalization: Expanding literacy research and practice. Reading Research Quarterly, 0(0), 1-13. doi:10.1002/rrq.301

Lloyd, A. (2019). Chasing Frankenstein’s monster: Information literacy in the black box society. Journal of Documentation, 75(6), 1475-1485. DOI 10.1108/JD-02-2019-0035

Nichols, T. P., & LeBlanc, R. J. (2021). Media education and the limits of “literacy”: Ecological orientations to performative platforms. Curriculum Inquiry, 51(4), 389-412. https://doi.org/10.1080/03626784.2020.1865104

Noble, S. U. (2018). Algorithms of oppression: How search engines reinforce racism. New York University Press.

Ridley, M., & Pawlick-Potts, D. (2021, June). Algorithmic literacy and the role for libraries. Information Technology and Libraries, 40(2), 1-15. https://doi.org/10.6017/ital.v40i2.12963

Valtonen, T., Tedre, M., Mäkitalo, K., & Vartiainen, H. (2019). Media literacy education in the age of machine learning. Journal of Media Literacy Education, 11(2), 20-36. doi:10.23860/JMLE-2019-11-2-2