Chapter 14 Quiz: Behavioral Targeting and Real-Time Bidding

Select the best answer for each multiple-choice question. Short-answer questions require a written response of 2–4 sentences.

1. The primary advantage of behavioral targeting over demographic targeting, from the advertiser's perspective, is:

A) Behavioral targeting is less expensive than demographic targeting B) Behavioral targeting reaches individuals who have demonstrated specific interests through their actions, producing significantly higher conversion rates C) Behavioral targeting is more accurate in identifying demographic characteristics like age and income D) Behavioral targeting is better regulated and therefore less legally risky for advertisers

2. In real-time bidding, the winning bidder typically pays:

A) The exact price they bid B) The price of the second-highest bid plus a small increment (second-price auction) C) A fixed floor price set by the publisher D) The average of all bids submitted in the auction

3. "Retargeting" (or "remarketing") in behavioral advertising refers to:

A) Showing ads to users who belong to the same demographic group as prior purchasers B) Adjusting ad bids upward when a user's profile improves over time C) Showing ads to users who have previously visited a website or viewed specific products, following them across other websites D) Re-running advertising campaigns that previously produced high conversion rates

4. The OCEAN model, used in psychographic targeting, is based on:

A) Geographic data about users in different regions (Oceans = regions) B) A five-factor personality model: Openness, Conscientiousness, Extraversion, Agreeableness, Neuroticism C) Online behavioral segmentation: the taxonomy of user interests derived from browsing patterns D) Facebook's proprietary audience segmentation system

5. Cambridge Analytica obtained data on approximately 87 million Facebook users through the consent of approximately 270,000 users. This was possible because:

A) Cambridge Analytica hacked Facebook's systems and obtained the data illegally B) Facebook sold Cambridge Analytica the data directly under a corporate data licensing agreement C) Facebook's then-current policy allowed an app installed by one user to access data about all of that user's friends, who had not consented D) Cambridge Analytica used browser fingerprinting to collect data from Facebook users' devices without their knowledge

6. Academic research on the effectiveness of Cambridge Analytica's psychographic targeting methods found:

A) Strong evidence that psychographic targeting produced significant and measurable electoral effects B) That the methods were highly effective at persuading undecided voters but not committed partisans C) Limited evidence that psychographic targeting produced measurable electoral effects beyond conventional targeting D) That psychographic targeting was uniquely effective in the United States but not in other countries

7. Political microtargeting's "opposition suppression" use refers to:

A) Targeting opposing campaign volunteers to discourage them from canvassing B) Using behavioral targeting to identify and reach likely opposition voters with demobilizing or discouraging messaging to reduce their voting participation C) Using data to identify and target donors to the opposing campaign with counter-messaging D) Suppressing news coverage critical of a campaign through algorithmic manipulation

8. The "RTB data leak" refers to:

A) A specific security breach in which RTB auction data was stolen B) The fact that bid requests in RTB auctions broadcast behavioral profile data to dozens of companies simultaneously, with no user consent C) The leakage of proprietary advertiser bidding strategies to competitors through shared ad exchange data D) Personal data stored by ad exchanges that was accessed without authorization

9. ProPublica's 2016 investigation of Facebook's advertising platform found that advertisers could:

A) Show housing advertisements while excluding users based on "ethnic affinity" categories B) Target advertisements based on users' private message content C) Purchase political advertising without any disclosure of the advertiser's identity D) Access users' precise GPS coordinates for location-based targeting

10. The concept of "redlining 2.0" refers to:

A) The illegal reuse of historical redlining maps for modern insurance pricing B) The reproduction of discriminatory exclusion patterns through algorithmic advertising targeting that encodes or reproduces race-correlated behavioral patterns C) A new regulatory framework that draws "red lines" around acceptable forms of behavioral targeting D) The use of geographic targeting to identify and avoid advertising to low-income neighborhoods

11. Eli Pariser's "filter bubble" concept describes:

A) The tendency of social media platforms to remove content that violates community guidelines B) The tendency of users to actively seek out information that confirms their existing views C) The narrowing of a user's information environment through algorithmic content curation based on behavioral history D) The practice of filtering out competitive advertising from platforms that serve their own first-party advertising

12. Jordan discovers they are being charged $50 more than Marcus for the same laptop. The most likely explanation offered in the chapter is:

A) A technical glitch in the retailer's pricing system B) Jordan's browsing history signals greater purchase commitment or lower price sensitivity, triggering dynamic pricing C) Jordan and Marcus have different credit scores, which the retailer is using for dynamic pricing D) The retailer is charging more to users accessing the site from a university IP address

13. Which of the following accurately describes why political microtargeting is more difficult to counter than broadcast political advertising?

A) Political microtargeting is not subject to campaign finance disclosure requirements, unlike broadcast advertising B) Microtargeted messages are shown only to their intended recipients, making them invisible to journalists, fact-checkers, and the opposing campaign C) Political microtargeting is technically more sophisticated and therefore harder for regulators to monitor D) Microtargeting allows campaigns to reach voters who would normally be protected by ad-blocking software

14. The chapter argues that "personalization" is a euphemism. Which of the following most accurately describes what "personalization" means from the platform's perspective?

A) Tailoring content and advertising to match user preferences as defined by the user themselves B) Using behavioral surveillance to increase commercial conversion rates and advertising revenue, with user benefit as an incidental byproduct C) Providing users with a customized interface that reflects their individual aesthetic preferences D) Reducing advertising clutter by showing users only the most relevant advertisements

15. Under the second-price auction model used in RTB, what is the strategic implication for DSP bidding?

A) DSPs should bid exactly what they believe the impression is worth to them, because overbidding wastes money B) DSPs should bid as low as possible to minimize advertising spend C) DSPs should bid exactly what the impression is worth to them (truthful bidding), because they will only pay the second-highest price if they win D) DSPs should overbid strategically to prevent competitors from winning high-value impressions

16. The Fair Housing Act's prohibition on discriminatory advertising applies to behavioral targeting because:

A) The law explicitly covers all forms of digital advertising B) Showing housing advertisements only to certain racial or ethnic groups — whether through explicit exclusion or through behavioral proxies that reproduce racial sorting — denies protected groups equal access to housing opportunity C) Any use of behavioral data in housing advertising is presumptively discriminatory under the Fair Housing Act D) The Fair Housing Act covers only print and broadcast advertising, not digital targeting

17. "Lookalike audience" targeting creates a redlining 2.0 risk because:

A) It explicitly uses racial categories to identify similar users B) If the original audience is racially homogeneous due to historical discrimination, the lookalike model extends that homogeneity, perpetuating the discrimination without explicit racial intent C) Lookalike audiences have been found to systematically identify lower-income users for more aggressive advertising D) The machine learning models used for lookalike targeting are known to encode racial stereotypes in their training data

18. Short Answer: The chapter describes real-time bidding as broadcasting your behavioral profile to dozens of companies with every page load. Explain why this aspect of RTB is particularly problematic from a consent and privacy perspective. Then briefly assess whether GDPR's consent requirements (as described in Chapter 12) are capable of addressing this specific problem.

[Answer space — 150–250 words]

Answer Key

1. B
2. B
3. C
4. B
5. C
6. C
7. B
8. B
9. A
10. B
11. C
12. B
13. B
14. B
15. C
16. B
17. B
18. Rubric: Full credit requires (1) a clear explanation of the broadcast nature of RTB bid requests — that a single page load simultaneously discloses behavioral profile data to dozens of companies, all of which receive the data whether or not they win the auction; (2) the observation that this differs from traditional consent models, which envision bilateral relationships between user and specific data collector; (3) a substantive assessment of GDPR's limitations for RTB — the Belgian DPA's finding that IAB's TCF was non-compliant reflects the difficulty of applying individual consent models to a system involving hundreds of simultaneous data disclosures. Strong answers will note the structural gap between consent as a bilateral mechanism and surveillance as a network phenomenon.

Chapter 14 | Part 3: Commercial Surveillance