Abstract
This chapter examines the rhetorical work of analytics and its product, quantitative metrics, as a constructive strategy of intention. This chapter situates analytics as a heuristic process, the intentional curation of data to evoke change within systems. Despite its reputation as a quantitative and computational field, analytics is equally, or perhaps even more so, a societal, behavioral, and rhetorical (humanistic) activity. While scholars across textual studies have shown some interest in big data and analytics, not enough work has emphasized or instrumentalized the contextual and societal dynamics of the practice. Like any rhetorical activity, analytics impose value and intention. While analytics are equally computational and interpretive, much more work has emphasized and developed the computational side of the field even while acknowledging deficits in how meaning is interpreted from such data and in the ethical uses of big data. From simple metrics like heart rates and blood pressures to more complex and interpretive ones like patient outcomes, consumer behavior, and voting patterns, queries invent and aggregate descriptions, assessments, and predictions that are intended to change human behavior.
By discovering the “secret order” of nature, science was seen, and culturally posited, as providing a rational access to the divine plan of creation, as being a way of ascertaining God’s intentions with the world at large…
Gyorgy Markus, “Why Is There No Hermeneutics of Natural Science?” (Markus, Gyorgy. 1987. Why is there no hermeneutics of natural sciences? Some preliminary theses. Science in Context, 1: 5–51.)
The drive to institute metrics often arises from the best of intentions, as a purported solution to real problems.
Jerry Z. Muller, The Tyranny of Metrics (Muller, Jerry. 2018. The Tyranny of Metrics. Princeton: Princeton University)
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
See for example, Zook Matthew, Solon Barocas, Danah Boyd, Kate Crawford, Emily Keller, Seeta Peña Gangadharan, Alyssa Goodman, Rachelle Hollander, Barbara Koenig, Jacob Metcalf, Arvind Narayanan, Alondra Nelson, and Frank Pasquale. 2017. Ten simple rules for responsible big data research. PLoS Computational Biology 13: e1005399. https://doi.org/10.1371/journal.pcbi.1005399. Zook et al. write that “As the size and complexity of available datasets has grown, so too have the ethical questions raised by big data research. These questions become increasingly urgent as data and research agendas move well beyond those typical of the computational and natural sciences, to more directly address sensitive aspects of human behavior, interaction, and health.” Arguing that “the need for direction in responsible big data research is evident” they offer ten rules for big data researchers, including “acknowledge that data are people and can do harm”; “recognize that privacy is more than a binary value”; and “engage with the broader consequences of data and analysis practices.” Similarly, Cristian Calude and Giuseppe Longo argue that computational analysis of big data tends toward the validation of spurious and arbitrary correlations due to the size and not the nature of the data (see, Calude, Cristian & Giuseppe Longo. 2017. The deluge of spurious correlations in big data. Foundations of Science 22: 595–612. https://doi.org/10.1007/s10699-016-9489-4)
- 2.
Kashmir Hill, in an article published on splinternews.com, examines a Facebook feature that used a smart phone’s location data to suggest new friends to users. The app, called “people we may know,” was reported to use a phone’s location data to find intersections among social networks. Hill cites a Facebook spokesperson who stated, “We show you people based on mutual friends, work and education information, networks you are part of, contacts you’ve imported and many other factors.” Among the “many other factors” included location information. The problem here is that the app can de-identify people who attend anonymous events such as a man who claimed to Hill that Facebook tracked his location to a meeting for suicidal teens and identified another anonymous parent as a “person you may know.” Hill rightly explains other scenarios where the app could be problematic or even dangerous for users. In an update to the story, Hill writes that Facebook appears to have discontinued using location as part of the “friends” algorithm (see, Hill, Kashmir 2016). Facebook is using your phone’s location to suggest new friends—which could be a privacy disaster. SplinterNews.com, June 29. https://splinternews.com/facebook-is-using-your-phones-location-to-suggest-new-f-1793857843
- 3.
Kahneman, Daniel, Amos Tversky, and Paul Slovic, Eds. 1982. Judgment Under Uncertainty: Heuristics & Biases. Cambridge, UK, Cambridge University Press
- 4.
Aristotle. On Rhetoric, Book 1, part 5. http://classics.mit.edu/Aristotle/rhetoric.1.i.html
- 5.
Lewis, Michael. 2004. MoneyBall: The Art of Winning an Unfair Game. New York: W.W. Norton.
- 6.
For information on the MIT Sloan Sports Analytics Conference, see: http://www.sloansportsconference.com/. Accessed 5 August 2018.
- 7.
The claim here is that analytics and big data have become central to political strategy not that their use in 2012 and 2016 was equivalent. For information on the 2012 campaign, see note 130. On Cambridge Analytica and the illegal harvesting of private Facebook data, see: Rosenberg, Matthew, Nicolas Confessore, and Carole Cadwalladr. 2018. How Trump consultants exploited the Facebook of millions. New York Times, 17 March. https://www.nytimes.com/2018/03/17/us/politics/cambridge-analytica-trump-campaign.html. Accessed 5 August 2018. An important distinction here is that in 2012 analysts used polling data and data from their own surveys, phone calls, and campaign operations while in 2016 social media data was taken from over 50 million Facebook users without their permission. Of course, the use of analytics in elections is not limited to American politics. Susan Ormiston a reporter for the CBC, for example, has described how all three major Canadian political parties have used data mining in federal elections. See Ormiston, Susan. 2015. Federal election 2015: How data mining is changing political campaigns. CBC.ca, http://www.cbc.ca/news/politics/federal-election-2015-how-data-mining-is-changing-political-campaigns-1.3211895. Accessed 5 August 2018.
- 8.
Davenport, Thomas and Jeanne Harris. 2007. Competing on Analytics: The New Science of Winning. Cambridge MA: Harvard Business Press, 7.
- 9.
Whelton Paul, Robert Carey, Wilbert Aronow, Donald Casey Jr., Karen Collins, Cheryl Dennison Himmelfarb, Sondra DePalma, Samuel Gidding, Kenneth Jamerson, Daniel Jones, Eric MacLaughlin, Paul Munter, Bruce Ovbiagele, Sidney Smith Jr., Crystal Spencer, Randall Stafford, Sandra Taler, Randal Thomas, Kim Williams Sr, Jeff Williamson, and Jackson Wright Jr. 2017. ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension. 2018: 71:e13–e115. doi: https://doi.org/10.1161/HY,0000000000000065
- 10.
James Paul, Suzanne Oparil, Barry Carter, William Cushman, Cheeryl Dennison-Himmelfarb, Joel Handler, Daniel Lackland, Michael LeFevre, Thomas MacKenzie, Olugbenga Ogedegbe, Signey Smith, Laura Svetkey, Sandra Taler, Raymond Townsend, Jackson Wright Jr., Andrew Narva, and Eduardo Ortiz. 2014. Evidence-Based Guideline for the Management of High Blood Pressure in Adults Report From the Panel Members Appointed to the Eighth Joint National Committee (JNC 8). Journal of the American Medical Association 311: 507–520. doi:https://doi.org/10.1001/jama.2013.284427
- 11.
Bakris, George and Matthew Sorrentino. 2018. Redefining hypertension—Assessing the new blood-pressure guidelines. New England Journal of Medicine 378: 497–499. https://www.nejm.org/doi/10.1056/NEJMp1716193
- 12.
Duhigg, D., see note 126.
- 13.
Indeed, the extent to which such work (and similar deployments of artificial intelligence) deprofessionalizes medicine and other fields and produces results without appropriate oversight raises considerable concern but such concerns have not appeared to influence the continuation of such work. On the role of AI in professional relationships, see: Haupt, Claudia E. 2019. Artificial professional advice. Northeastern University School of Law Research Paper No. 350–2019, June 7. SSRN: https://ssrn.com/abstract=3400898. Steven Sutton, Vicky Arnold, and Matthew Holt argue that deploying AI in knowledge work, like medicine, law, and business, may speed up decision making but at the expense of skill development, knowledge acquisition and retention, and human expertise. See, Sutton, Steven, Vicky Arnold, and Matthew Holt. 2018. How much automation is too much? Kee** the human relevant in knowledge work. Journal of Emerging Technologies in Accounting 15: 15–25. https://doi.org/10.2308/jeta-52311
- 14.
Yet, “discovery rhetoric” in metrics remains strong. Accounts have stressed and assumed the integrative power associated with big data’s heuristic tools but researchers have also shown the ready pitfalls, false correlations, and meaningless random associations of variables that can be easily attained by mining through unstructured data and ho** for surprise revelations. University of Toronto Epidemiologist Peter Austin and his colleagues demonstrated that by testing multiple hypothesis across patient data from all 10,674,945 Ontario residents between 18 and 100 years old, they could show that people born under the astrological sign of Leo had a higher probability of gastrointestinal hemorrhage while those born under the sign of Sagittarius had a higher probability of humerus fracture. See, Austin, Peter, Muhammad Mamdani, David Juurlink, and Janet Hux. 2006. Testing multiple statistical hypotheses resulted in spurious associations: a study of astrological signs and health. Journal of Clinical Epidemiology, 59: 964–969. https://doi.org/10.1016/j.jclinepi.2006.01.012. In some of my own early experimental studies, our data showed conclusively that patients with diabetes were more likely to have higher blood sugar levels and patients with lung disease were more likely to have pneumonia.
- 15.
Muller, 17.
- 16.
Muller, 18.
- 17.
Berger, xiv.
- 18.
See Anscombe, p, 87–89. “We can see that a great many of our descriptions events effected by human beings are formally descriptions of executed intentions.” Anscombe writes of “practical” and “speculative” knowledge, “the account that one could give of what one is doing, without adverting to observation; and the account of exactly what is happening at a given moment (say) to the material one is working on” (88–89). Intentional action is informed by a knowledge about what one is attempting to accomplish at that moment, “knowing one’s way about” as Anscombe puts it. Intentional actions are voluntary, reminds Anscombe. Such actions may be “reluctant” but still voluntary.
- 19.
Muller, 18.
- 20.
See for example, Powell, Emilie, Rahul Khare, Arjun Venkatesh, Ben Van Roo, James Adams, and Giles Reinhardt. 2012. The Relationship between inpatient discharge time and emergency department boarding. Journal of Emergency Medicine 42: 186–196. doi: https://doi.org/10.1016/j.jemermed.2010.06.028. Powell et al. conclude that shifting inpatient peak discharges by 4 hours earlier could eliminate ED boarding.
- 21.
Muller, 19.
- 22.
Shine, Dan. 2014. Discharge before noon: An urban legend. The American Journal of Medicine 128: 445–446. doi: https://doi.org/10.1016/j.amjmed.2014.12.011
- 23.
Shine responds directly to a paper by Wertheimer et al. that argued that discharge before noon actually decreased readmissions. Shine argues that Wertheimer et al.’s study was conducted during an effort to reallocate substantial medical social services to weekends. Shine argues that these other changes to services confound Wertheimer et al.’s findings. Further, Shine notes that Wertheimer et al. assumed but did not measure the effect of pre-noon discharge on ED boarding. See: Wertheimer, Benjamin, Ramon Jacobs, Martha Bailey, Sandy Holstein, Steven Chatfield, Brenda Ohta, Amy Horrocks, and Katherine Hochman. 2014. Discharge before noon: an achievable hospital goal. Journal of Hospital Medicine 9: 210–214.
- 24.
Sine, 3.
- 25.
Muller, 23.
- 26.
See: Kabir, Ahmedul, Carolina Ruiz, Sergio Alvarez, and Majaz Moonis. 2018. Regression, classification, and ensemble machine learning approaches to forecasting clinical outcomes in ischemic stroke. In Biomedical engineering systems and technologies. BIOSTEC 2017, ed. Nathalia Peixoto, Margarida Silveira, Hesham Ali, Carlos Maciel, and Egon van den Broek. Communications in Computer and Information Science, vol 881. Springer, Cham.
- 27.
Dreyfus, Herbert L. and Paul Rabinow, 1983. Michel Foucault: Beyond structuralism and hermeneutics. Chicago: University of Chicago Press, 123.
- 28.
Dreyfus and Rabinow, 187.
- 29.
Bruno Latour, 2010. On the Modern Cult of the Factish Gods. Durham, NC and London: Duke University Press.
- 30.
Dreyfus and Rabinow, 185.
- 31.
Dreyfus and Rabinow, 123.
- 32.
Dreyfus and Rabinow, 79.
- 33.
Dreyfus and Rabinow, 79.
- 34.
Using quantitative models in the humanities is not new. Springer publishers for example has managed a book series, Quantitative methods in the Humanities and Social Sciences since 2014 (see, https://springer.longhoe.net/search?facet-series=%2211748%22&facet-content-type=%22Book%22). Where we have differentiated our approach has been in the types of questions we are pursuing and the application of our work. Rather than seeing our work contributing to various projects within a particular humanities-based disciplinary perspective we proposed taking concerns normally raised by humanists and positing these in an applied, extra-disciplinary context.
- 35.
For data sources, see: World Bank. Countries and Economies. http://data.worldbank.org/country. Accessed 1 Jan 2020; Organization for Economic Co-operation and Development (OECD). Economic Outlook for Southeast Asia, China and India 2014 Beyond The Middle-Income Trap, OECD iLibrary. for http://www.oecd.org/site/seao/Pocket%20Edition%20SAEO2014.pdf. Accessed 1 Jan 2020. Gapminder data is available online at www.Gapminder.org
- 36.
Four criteria needed to be met for a clinical trial to be included in the study: The trial had to be completed, have results, be an interventional study, and be a study on drugs.
- 37.
For data sources, see: US Department of Health and Human Services, National Cancer Institute. Comprehensive Cancer Information. http://www.cancer.gov/. Accessed 1 Jan 2020. U.S. National Library of Medicine. Search for Studies (database). http://www.clinicaltrials.gov/ct2/home. Accessed 30 April 2014; WebMD. Liver Cancer Causes, Survival Rate, Tumor Types, and More. http://www.webmd.com/cancer/understanding-liver-cancer-basic-information. 1 May 2014; New York State Department of Health. NYS Cancer Registry and Cancer Statistics. NYS Cancer Registry. http://www.health.ny.gov/statistics/cancer/registry/. April 30, 2014.
- 38.
See, Kamath, Suneel, Sheetal Kircher, and Al Benson. 2019. Comparison of cancer burden and nonprofit organization funding reveals disparities in funding across cancer types. Journal of the National Comprehensive Cancer Network. 17: 849–854. doi: https://doi.org/10.6004/jnccn.2018.7280
- 39.
Suneel et al., 849.
- 40.
Dreyfus & Rabinow, 204.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Faber, B. (2022). Practice: Heuristics and Hermeneutics in Data Science. In: The End of Genre. Postdisciplinary Studies in Discourse. Palgrave Macmillan, Cham. https://doi.org/10.1007/978-3-031-08747-9_6
Download citation
DOI: https://doi.org/10.1007/978-3-031-08747-9_6
Published:
Publisher Name: Palgrave Macmillan, Cham
Print ISBN: 978-3-031-08746-2
Online ISBN: 978-3-031-08747-9
eBook Packages: Social SciencesSocial Sciences (R0)