When taking research literacy lessons, I have the students conduct a basic Google search using their inquiry question. Then, in a new tab, they use only the key words and watch what happens to the number of results. Next, they use a range of Boolean operators with their key words. We compare and discuss the change in the number of results (usually significantly less) and the power of effective search skills. I always point out that despite everyone in the room using the exact same words and operators, we get a different number of results. We speak about Google’s algorithms, it’s filtering of their results, and the power of going beyond page 1. This has always been a valuable discussion point, as some believe the filter bubble can dramatically increase confirmation bias. In a climate of divisive viewpoints, this is important to note. Not only in the personal and social world but also the world of academia. Students must have the opportunity to challenge their thinking to develop deeper understandings and develop their capacity for critical thinking.
In his 2011 TED Talk, Pariser highlighted the need for algorithms to be transparent and customisable to enhance companies’ ethics and “civic responsibility” in terms of how people connect and with what they are exposed to (TED2011, 2011). It seems Google responded. When recently searching in Google, I wanted to see if I could turn off certain algorithms or data collection – could I go back to square one to have a truly pure and uncorrupted search experience. It turns out, in 2018, Google released Your Data in Search which makes deleting your search history and controlling the ads you see much easier. You can also turn off Google’s personalisation. While some studies suggest Google’s attempt at reducing the filter bubble (searching in private mode and when signed out) does not greatly affect the disparity between users’ search results, it is perhaps a step in the right direction. It is worthwhile noting that Google disputes the claim that personalisation greatly effects search results.
The jury may still be out as key players are unsurprisingly at odds, however I have seen the difference in results first hand when working with classes of students. In the realm of their academic research, it may not be as big of an issue as say perpetuating political beliefs or other ideologies, however these algorithms are deciding what it deems most useful or important for these students. This can limit students’ search rather than assist, and popularised click bait can hinder their academic as well as social searching. An alternative search engine, which does not track or store your personal information, is DuckDuckGo. A search engine many librarians and educators have been promoting for some time. The next time I take a research literacy lesson, I will put it to the test and see how it stacks up against Google.
DuckDuckGos Business Model. Cuofano, G. (n.d.). DuckDuckGos Business Model [Image]. Retrieved from https://fourweekmba.com/duckduckgo-business-model/In terms of curation, Valenza (2012) suggests we be mindful of the filter bubble when evaluating the curations of others. Are viewpoints missing? Whose perspective is the curation from? On the other hand, effective human curation can alleviate the filter bubble. Human curators, particularly those participating in collective curation, have the ability to provide multiple perspectives within a curation. This gives users a more comprehensive pool of sources to select from and can expose users to a breadth of viewpoints. Even Apple is using human curation to counter the limitations of algorithmic curation. Apple intends to present a curation of quality-controlled news by leveraging the collective skills and expertise of a curation team. This is also a powerful exercise for students. Collective curating of resources for their research tasks can reduce work load, provide multiple and alternate perspectives and encourage collaborative processes and communication. Shirky (O’Reilly, 2008) highlights an instance in 2008 whereby a Toronto college student created a Facebook study group to mimic an IRL study group. In this group, membership was open and vast. He was quickly charged with cheating by Ryerson College. I personally don’t believe the creation of this group to be in violation of academic integrity. Even though students may be collaboratively curating (something I think should be encouraged), they, themselves as individuals, must still be discerning in their selection of sources and evidence, and must still demonstrate their ability to evaluate, analyse and sythensise. Collective curation provides opportunities for students to debate, widen the available perspectives, and support one another in their academic endeavours.
So, the next topic to explore is appropriate collective curation tools that support students inside and outside the school environment.
O’Reilly. (2008, September 19). Web 2.0 Expo NY: Clay Shirky (shirky.com) It’s Not Information Overload. It’s Filter Failure [Video file]. Retrieved from https://www.youtube.com/watch?v=LabqeJEOQyI
The ability to read and interpret information is a fundamental skill needed to participate fully in the world. These basic skills (although actually quite complex) will continue to be as important, if not more, than the past. The information-rich world is expanding; however, algorithms are filtering the information we see. So, people’s values and beliefs are consistently reinforced, while other perspectives are left out or buried on page 3 of Google search results – an equally ominous fate. In turn, this leads to confirmation bias, which can be detrimental to those who cannot critically evaluate what they are experiencing and reading.
Algorithms present users with a calculated selection of “relevant” information; however, it is clear that users must develop and employ the necessary skills to work within these algorithms. The top search results are not always the most useful. Searchers must not ignore the other titbits of information such as Google’s snippets displayed under each search result. While Google can change the snippet from the meta description of the webpage to their own algorithmically determined snippet (Silver Smith, 2013), it is still a useful port of call that many searchers skip over. This allows savvy searchers to preliminarily assess the relevance and worth of the search results – which are not always at the top of the page (Wineburg & McGrew, 2017). But algorithms are not the only cause of confirmation bias. Ashrafi-Amiri & Al-sader (2016) suggest searchers’ assumptive search queries based on fact retrieval and verification will characteristically retrieve more bias results than if the query were non-assumptive; that is, knowledge acquisition, comparative, analytical, and exploratory in nature. Information literacy instructors must be aware of this and consider this when developing instruction for students.
TLs must address the critical thinking skills required to work with and within algorithms that reinforce bias. Maynes (2015) identifies the role of information literacy instructors in explicitly teaching students about the forms of bias, ways to identify their own bias, and skills to mitigate the potential effects of their bias. This involves teaching the metacognitive skills students need to not only know the strategies to use but how, when and why to use them (Maynes, 2015). A combination of lateral and vertical reading is useful in all information evaluation situations. While many libraries utilise a CRAP (Currency, Reliability, Authority, Purpose/Point of View) test to step students through the information evaluation process, other steps can also be considered so students tune into their metacognition and identify their bias (Wineburg & McGrew, 2017). Allan (2017) suggests incorporating some form of personal reflection into the information literacy sessions offered to students. Students must not only be taught that confirmation bias exists, they must also be taught the skills to identify it in themselves and to deal with it when it occurs. One such strategy is to identify when a source of information elicits an emotional response from the reader – Does it make you happy? Sad? Reinforce? Challenge? Developing self-regulation triggers the reader to seek additional information and reflection to consider the opposite or alternate (Hirt & Markman, 1995; Lord, Lepper, & Preston, 1984; Mussweiler, Strack, & Pfeiffer, 2000). This requires information searchers to reflect on their reactions at each step and consider whether their evaluation of the usefulness or credibility of the source would be the same if it presented the opposing viewpoint. Deliberately considering the opposing viewpoint requires the searcher to consider their bias and the bias of others. This is a powerful strategy in unveiling subconscious or hidden bias. Allan (2017) posits adding an S (Self-examination or Self-awareness) to the beginning of the CRAP test would highlight the importance of identifying and recognising cognitive and confirmation bias.
SCRAP it: Source evaluation process.
The importance of slowing down the information evaluation process by thinking effortfully and deliberately (Kahneman, 2011) and evaluating laterally (Wineburg & McGrew, 2017) is central to 21st century information and digital literacy. Evaluating laterally requires searchers to seek and consider context and perspective, which means they must seek additional information. Slowing down does not simply mean taking longer to read the article and its parts – it means careful and deliberate consideration and slowing your judgement by first taking your bearings and exploring laterally. This may mean to first leave the site or visit the About Us section to find out more about the author or the organisation, before navigating back to the original source (Wineburg & McGrew, 2017). Thinking laterally can occur in multiple stages of the CRAP test, particularly when assessing the reliability and purpose/point of view present in the source. Searchers will need to explore other sites to learn more about the information. While searchers will not always slow down and employ lateral reading, it is important to know when to slow down. High stakes situations where the searcher may possess a strong bias already or where the information may have significant consequences for the searcher or others, or a highly contested issue or topic may require more deliberate reasoning to ensure the searcher is acquiring balanced, truthful information (Maynes, 2015). Considering the opposite is another practical strategy to employ in these situations.
It is clear that information evaluation and digital literacy skills need to evolve with changing demands and issues within the information landscape. Information literacy instructors must stay abreast of these changes and adapt evaluation strategies as needed. A start might be to model and incorporate lateral reading into existing strategies and follow Allan’s (2017) suggestion and put that S at the beginning of CRAP.
References
Allan, M. (2017). Information literacy and confirmation bias: You can lead a person to information, but can you make him think? Informed Librarian Online, 2017(5). Retrieved from https://asu-ir.tdl.org/handle/2346.1/30699
Ashrafi-Amiri, N. & Al-sader, J. (2016). Effects of confirmation bias on web search engine results and a differentiation [Thesis]. Retrieved from https://core.ac.uk/download/pdf/43564372.pdf
Hirt, E.R., & Markman, K.D. (1995). Multiple explanation: A consider-an-alternative strategy for debiasing judgments. Journal of Personality and Social Psychology, 69(6), 1069– 1086.
Kahneman, D. (2011). Thinking, fast and slow. New York: Farrar, Straus and Giroux
Lord, C. G., Lepper, M. R., & Preston, E. (1984). Considering the opposite: A corrective strategy for social judgment. Journal of Personality & Social Psychology, 47(6), 1231-1243.
Mussweiler, T., Strack, F., & Pfeiffer, T. (2000). Overcoming the inevitable anchoring effect: Considering the opposite compensates for selective accessibility. Personality and Social Psychology Bulletin, 26(9),1142–1150.
Trends in digital publishing undoubtedly have an impact on school libraries and their collections in terms of the access and acquisition of resources. Changes in the digital landscape have driven changes to traditional modes of reading and accessing materials and how those materials are marketed to consumers. Shatzkin specifically points out these changes over time; evolving from the power of a vast and plentiful selection in physical book stores to the power of increased access to once out of stock books now available online; almost as if a secret club or entry into a world once inaccessible, which is an enticing proposition to many (2016). Shatzkin was referring to the rise of Amazon and the collapse of Boarders, which had a long lasting impact on the way in which books are distributed (2016). Books, both physical and digital, can now be acquired through a vast array of algorithms and new influences and influencers. Not only do the “Four Horsemen” (Galloway, 2015) drive online sales, so do modern influencers on social media such as Instagram and Twitter. An example of the power of an online influencer, though not related to books, is the reported loss of $1.3 billion dollars suffered by Snapchat in response to a tweet from Kylie Jenner showing her ambivalence toward the app’s update (Shen, 2018). While other factors may also have been at play here, it does highlight the power of online profiles in the consumption of products. This, too, has an impact on the way school libraries identify and select resources, as library users may be desiring a particular resource due to an online review, recommendation and/or hype.
Algorithms and Search Engine Optimization also influence how school libraries identify, select and acquire resources, as they dictate or skew what is found when searching online and online marketing influences decide where and how the resources are purchased (Shatzkin, 2015). With the increase of self-publishing and digital publishing, the “digital advantage” publishers and authors once saw (as Shatzkin alluded to with the birth of Amazon [2016]) is diminishing as the market is crowded (Ruscello, 2017). This can make collection development for school libraries arduous, as locating the most effective resources to support teaching and learning needs requires more sifting than before. On the other hand, the “Four Horsemen” (Galloway, 2015) have intentional marketing strategies to direct consumers to specific resources, which narrow search results. Google’s algorithm updates have consistently responded to consumer satisfaction; thus, the updates have optimised the consumer search experience by increasing the quality of the search results and introducing search entities, which auto-fill search terms and direct users down a variety of digital paths (Carson, 2016). Additionally, the online presence of authors and their profiles does influence the occurrence of their books in search results (Shatzkin, 2016). Ultimately, libraries are at the mercy of budgets; therefore, price point will impact the selection of certain resources. As Shatzkin explains, where a consumer buys a product is not necessarily where they made the decision to buy it; something he refers to as “the fallacy of last click attribution” (2016, para. 21). The ability to easily compare prices online using search engines such as Google, perhaps allows libraries to acquire more resources at a cheaper, more competitive price than before.
A variety of technological trends can affect school libraries and their collections. Libraries need to respond to the reading and research preferences of their clientele. Staff and students want easily accessible, reading level appropriate material to meet their teaching and learning needs. In this sense, libraries are able to strike a balance between material presented digitally and material presented in hard copy form. Thus far, it can be seen that the trends presented by Shatzkin (2015 and 2016) may have a positive influence over resourcing these needs but increased access to resources also requires caution in identifying and selecting effectives resources. The digital landscape requires careful evaluation of resources to sift through the plethora of options to find what is most useful in meeting the needs of library users. Overall, online reviews, heightened use of algorithms and competitive prices change how users source products (both digital and physical) and impact how library collections are developed at the identification, selection and acquisition levels; therefore, evaluation of resources is paramount.
References
Carson, J. (2016, February 4). SEO and psychology: The behavior of the online consumer. The Make Good. Retrieved from http://www.the-makegood.com/2016/02/04/seo-and-psychology-the-behavior-of-the-online-consumer/
Galloway, S. [DLDconference]. (2015, January 20). The four horsemen: Amazon/Apple/Facebook & Google – who wins/loses (Scott Galloway, L2 Inc.)|DLD15 [Video file]. Retrieved from https://www.youtube.com/watch?v=XCvwCcEP74Q
Ruscello, J. (2017, December 22). 15 self-publishing trends to watch in 2018 [Blog post]. Retrieved from http://www.blurb.com/blog/self-publishing-trends-2018/
Shatzkin, M. (2016). Book publishing lives in an environment shaped by larger forces and always has. The Shatzkin Files. Retrieved from http://www.idealog.com/blog/book-publishing-lives-in-an-environment-shaped-by-larger-forces-and-always-has/
Shatzkin, M. (2015). Big focus at DBW 2016 on the tech companies that are shaping the world the book business has to live in. The Shatzkin Files. Retrieved from https://www.idealog.com/blog/big-focus-at-dbw-2016-on-the-tech-companies-that-are-shaping-the-world-the-book-business-has-to-live-in/
Shen, L. (2018, February 22). Why Kylie Jenner may be to blame for Snap’s recent $1 billion loss in value. Fortune. Retrieved from http://fortune.com/2018/02/22/kylie-jenner-snapchat-snap-value-stock/
