Team 2 - Annie, Bret, Vicki
Topic Summary
Vicki
The chapters this week discuss several clear cases of manipulation by companies/organizations through their use of big data, as described in O’Neil (2017).
- Online Advertising and “Microtargeting”: Big data can be used to generate different tiers of marketing for different people. This ensures that the ads you see are different (i.e. customized) than the ones your neighbors, friends and colleagues see. Microtargeting is used by politicians to appeal to a wide range of voters by carefully curating themselves to match up to those voters’ expectations. By putting their best foot forward, people end up only seeing the intentionally curated view politicians want us to see.
In a political context, it remains to be seen if these distortions can be used to affect election outcomes.
- For-Profit Colleges: These schools notoriously use targeted ads to reel in uneducated people with low SES. The ads promise upward mobility but charge incredibly high tuition rates and “are [worth] about the same as a high school diploma” in the workplace (p. 80).
- Social Media: The algorithms that run social media sites are notoriously guarded and opaque. Companies like Facebook routinely run experiments that have real-world consequences on elections and other social events.
O’Neil Chapter 4, Propoganda Machine
Annie
Chapter 4 of Weapons of Math Destruction is about the WMDs that power online advertising for for-profit colleges. These institutions use predatory online advertising to target poor people who believe education is the key to upward mobility.
Predatory ads identify and prey on someone’s weaknesses. Advertisers use data generated by our behavior online, our zip code, etc. to identify our shortcomings and insecurities, then deliver ads that offer “false or overpriced promises” (Ch. 4). In the case of for-profit colleges, the promise is a better life through education. In reality, students end up with crippling debt and a degree of little value.
For-profit colleges spend heavily on online advertising because it is so easy to target a specific (vulnerable) segment of the population. Much of the budget goes to lead generation, or building lists of prospects who have clicked an ad and filled out an interest form. Apollo Group, the parent company of University of Phoenix, spends about $2,225 per student on marketing and only $892 per student on instruction. In contrast, Portland Community College spends $5,953 per student on instruction and $185 per student on marketing (p. 79).
Predatory lenders also engage in online advertising. The payday loan industry uses data to identify people in need and targets them with ads for short-term loans with huge interest rates. People who fill out the online form often fall victim to scams and theft.
According to O’Neil, there are some laws on the books that protect health and credit data. She recommends legislation of lead generation to protect personal data. However, she warns that most WMDs operate using “work-arounds” (p. 82-83).
O’Neil Chapter 10, The Targeted Citizen
Bret
We learned in the first course of this EMPA program the difference between power (one’s potential to influence) and influence (the execution of that power to modify behavior). O’Neil (2016) warns in Weapons of Math Destruction that “Facebook’s potency comes not only from its reach but also from its ability to use its own customers to influence their friends” (p. 181). With nearly 2 billion people using Facebook worldwide, the publicly traded corporation has access to an enormous amount of big data which significantly increases its power ranking. Consequently, their ability to influence behavior is significant. While I would like to believe that they will and have used this status primarily for good, it certainly seems reasonable that destructive consequences could precipitate as well.
Consumers of data demand that information is current, relevant and can be attributed to certainty in achieving a desired outcome. This chapter discussed ways that Facebook encouraged increased voter participation during the 2016 US presidential election. Additionally, the author described one independent firm’s ability to create 40,000,000 “psychographic” analyses from Facebook profiles of voters in this country (O’Neil p. 191), the results of which were used to target specific groups with specific messages seeking a specific outcome. My takeaway from this chapter is that the complexities of politics, behavior and outcomes can’t be summarized solely by mathematical equations or algorithms. However, I find it fascinating the enormity of the potential power and influencePOWER & INFLUENCE that companies like Facebook, Google, Apple, Amazon and Verizon have on society. I’m certain that each of these firms knows far more about me than I’ll ever know about them.
https://www.internetworldstats.com/facebook.htm
Discussion Questions
- How does predatory online advertising perpetuate income inequality?
- How can we protect vulnerable people from predatory online advertising and the WMDs that power it?
- If a 15-minute couples’ interview can develop 1000’s of data points, how many data points are you contributing every day?
- Do you consider Facebook to be a tool from which you receive the greater benefit, or is big data the recipient of that advantage?
References
- O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Broadway Books. CH4 propaganda machine
- O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Broadway Books. CH10 targeting citizens