Unit 5 — Media and Public Opinion

Template: Activate → Input → Practise → Produce → Reflect.
Niveau: E. Abitur (school-leaving examination) Klausur prep across the year.
Course tagging: basic course (Basisfach, E-BF) and advanced course (Leistungsfach, E-LF).

Learning objectives#

  • I can read short texts on platform algorithms and public opinion (Tufekci, Wu) and identify the writer’s central causal claim.
  • I can use vocabulary of platform-media analysis (algorithmic amplification, attention rent, recommendation system, network effect, platform governance).
  • I can write a 450-word media-analysis essay sustaining a complex causal argument.

curriculum framework (“Bildungsplan”) alignment#

  • 3.4.1 / 3.5.1 Soziokulturelles Orientierungswissen / Themen
  • 3.4.3.2 / 3.5.3.2 Leseverstehen
  • 3.4.3.5 / 3.5.3.5 Schreiben
  • 3.4.4 / 3.5.4 Text- und Medienkompetenz

(Sources: https://www.bildungsplaene-bw.de/,Lde/LS/BP2016BW/ALLG/GYM/E1/IK/11-12-LF / https://www.bildungsplaene-bw.de/,Lde/LS/BP2016BW/ALLG/GYM/E1/IK/11-12-BF )

Lead-in story#

The class read short extracts from Zeynep Tufekci’s Twitter and Tear Gas (2017) and Tim Wu’s The Attention Merchants (2016). Mr. Yilmaz framed the question: what is the difference between public opinion and aggregated platform behaviour? The class noticed quickly that the answer is not nothing and not everything.

1. Activate#

Causal-claim scan. With your partner, list 3 causal claims you have heard about social media + politics. Mark each as plausible / contested / tabloid.

2. Input#

Reading — two extracts (paraphrased)#

Tufekci (2017): Algorithmic amplification produces attention asymmetries that are not the same as public opinion. The visible majority on a platform is a function of the platform’s ranking signals, not of the underlying population.

Wu (2016): The history of mass media is the history of attention merchants — actors whose business model is the harvest, packaging, and resale of attention. Platforms inherit and intensify this model.

Vocabulary — platform-media analysis#

algorithmic amplification, attention rent, recommendation system, network effect, platform governance, content moderation, attention economy, asymmetric visibility, the visible majority, micro-targeting.

3. Practise#

Niveau E — controlled#

  1. Match: algorithmic amplification → ranking-driven visibility; attention rent → the price of your attention; the visible majority → not the same as the actual majority.
  2. T or F: platforms inherit the attention-merchant model; algorithmic amplification reflects underlying public opinion 1:1.

Niveau E — productive#

  1. Build 4 sentences applying platform-media vocabulary to a recent online controversy.
Answer key

Controlled. 1. all true. 2. T, F (it is shaped by ranking signals).

Productive. 3. Open.

4. Produce#

Media-analysis essay, 450 words. Argue a complex causal claim about platforms and public opinion. Use 6 platform-media terms + 7 academic discourse markers + 2 cleft structures + 2 integrated quotes + 1 hedge.

Sample#

The single most useful conceptual distinction in this year’s reading on platform media is Zeynep Tufekci’s between public opinion and the visible majority. The available evidence suggests that algorithmic amplification produces attention asymmetries that systematically diverge from the underlying distribution of views in a given population. Accordingly, what looks like a consensus on a platform — a trending hashtag, a saturated reply space, a viral take — is not, in the strict sense, public opinion. It is a function of the platform’s ranking signals interacting with user behaviour. By contrast with the older mass-media model, where editorial gatekeepers were visible and contestable, platform amplification is structurally opaque. It is precisely this opacity that makes Tim Wu’s attention merchant genealogy useful: the business model is continuous with earlier media, but the visibility of the gatekeepers has decreased while the precision of micro-targeting has increased. More specifically, the common claim that the algorithm is showing what people want is, on inspection, a circular one. The algorithm is showing what its ranking signals treat as engaging; engagement-as-measured is shaped by what the algorithm previously elevated. The system is, in this regard, a feedback loop with no neutral baseline. Critics will counter, fairly, that some signal of underlying interest is preserved — otherwise the platform would lose users. I accept that, but the signal is filtered through ranking choices that are themselves political-economic. The honest analytical move is to decompose the visible majority into three components: underlying preferences (real but partial), ranking-induced amplification (large and shaping), and feedback-loop reinforcement (non-trivial). It is precisely this decomposition that the public discourse has been bad at. Caution is warranted; the next decade of platform governance will, in my reading, hinge on whether regulators can compel transparency on the second and third components. Without that, public opinion and aggregated platform behaviour will continue to be confused — to the political advantage of whoever sets the ranking signals.

5. Reflect#

One thing in your notebook: Write one sentence using something you learned in this Unit.

Exam example#

Open the separate assessment .

Teacher notes#

**Slide deck timing.** 90 minutes total (Doppelstunde — typical in the Oberstufe). Lead-in 6 min · Activate 8 min · Input 25 min · Practise 15 min · Produce 30 min · Reflect 6 min.

Differentiation — support (below Niveau E).

Build your essay from these sentence starters (finish each in your own words, then link them):

  • The key distinction I want to explore is between public opinion and the visible majority, because ______.
  • What actually shapes the visible majority is not real preference but ______.
  • This matters politically because ______.
  • One objection is that ______; however, ______.

Useful phrase bank: it is important to note that, this suggests that, by contrast, as a result, ranking signals decide what we see, engagement is not the same as agreement.

Extension (above Niveau E / Oberstufe).

Write a comparative critical piece (roughly 550 words) that tests Tufekci’s public-opinion / visible-majority distinction against a second explanatory frame of your choice — for example Wu’s attention-merchant genealogy, or a public-sphere model such as Habermas’s — and argue which frame better predicts how a regulator should measure platform influence. Sustain a two-sided argument, concede the strongest counter-position honestly, and close on a falsifiable claim about what evidence would settle the disagreement.

Common pitfalls#

  • The algorithm is showing what people want is circular — flag it.
  • Don’t conflate engagement with preference.
  • Causal claims need decomposition, not assertion.

Further reading / listening#

  • Zeynep Tufekci, Twitter and Tear Gas (2017).
  • Tim Wu, The Attention Merchants (2016).
  • Joseph Bernstein — accessible journalism on platforms.

Downloads

Listen

Audio and transcripts for this unit.

Text 1
Show text
The class read short extracts from Zeynep Tufekci's Twitter and Tear Gas (2017) and Tim Wu's The Attention Merchants (2016). Mr. Yilmaz framed the question: what is the difference between public opinion and aggregated platform behaviour? The class noticed quickly that the answer is not nothing and not everything.
Vocabulary 1
Show text
algorithmic amplification · attention rent · recommendation system · network effect · platform governance · content moderation · attention economy · asymmetric visibility · the visible majority · micro-targeting