Chapter 24 Exercises

Exercise 24.1: Self-Placement on the Intensity Spectrum (Individual, 30 minutes)

For a fandom you participate in (or have participated in), place yourself on the intensity spectrum using the behavioral markers from section 24.1. Answer the following:

  1. Which level — casual, fan, enthusiast, stan — best describes your engagement? What specific behaviors led you to this self-classification?
  2. Has your position on the spectrum changed over time? If so, describe what drove the movement (upward or downward). Was it primarily platform design, community engagement, individual psychology, or some combination?
  3. Identify the single behavioral practice that most clearly distinguishes your current level from the level above it (or, if you classify yourself as a stan, from the level below you).
  4. If you have fandom involvement in multiple properties, do you occupy different positions on the spectrum for different fandoms simultaneously? What explains the variation?

Exercise 24.2: Running the Python Analysis (Technical, 90 minutes)

Prerequisites: Python 3.8+, pip install vaderSentiment pandas matplotlib numpy sklearn

  1. Run stan_sentiment_analysis.py in the code/ directory. Examine the output and visualizations. Answer: - Which post category shows the highest mean compound score? Is this consistent with what parasocial theory would predict? - The script shows an in-group/out-group sentiment comparison. Explain in your own words what this difference reveals about the emotional structure of stan communities. - Look at the weekly sentiment time series. During "controversy weeks," how does average sentiment change? What community dynamics might explain this pattern?

  2. Run fan_intensity_classifier.py. Examine the cluster profiles and character placements. Answer: - Which features (behavioral variables) most strongly distinguish Casual Interest from Stan clusters? How can you tell? - The script places Mireille and TheresaK in the same cluster. Given the qualitative descriptions of each character in Chapter 24, does this cluster assignment make sense? What important difference between them does the cluster model not capture? - How does IronHeartForever's feature profile differ from TheresaK's, even though both are classified at high intensity levels? What does this difference reveal about the multiple dimensions of fan intensity?

  3. Extension (optional): Modify stan_sentiment_analysis.py to add a fifth post category of your own design (e.g., "anniversary celebration" or "member hiatus response"). Create at least 8 template posts for this category. Run the modified script and compare your new category's sentiment profile to the existing ones.

Exercise 24.3: Parasocial Architecture Reverse Engineering (Analysis, 60 minutes)

Choose one of the following BTS content formats (or a comparable format from another artist if BTS content is not accessible to you):

  • A Bangtan Bomb video
  • A BTS Weverse post from a member
  • A Run BTS! episode segment
  • A BTS concert VCR (video between sets)

Analyze your chosen content item for its parasocial architecture. Identify:

  1. Direct address instances: How many times does the content address the viewer/audience directly? What register (formal, informal, intimate)?
  2. Apparent disclosure moments: What personal information does the content appear to disclose? Is this likely to be genuinely private or managed "apparent private"?
  3. Consistency signals: What elements connect this content to previous or ongoing content in ways that build PSR?
  4. Bias-system engagement: Does the content preferentially develop parasocial bonds with specific members? How?
  5. HYBE's design hand: Are there elements of the content that clearly reflect management strategy rather than spontaneous personal expression?

Write a 400-word analysis of how this content item functions as parasocial architecture, drawing on section 24.2's framework.

Exercise 24.4: The Bias System in Practice (Group Discussion, 45 minutes)

In small groups, discuss:

  1. The K-pop "bias system" creates a specific form of parasocial attachment within multi-member group fandoms. Does the Western celebrity world have an equivalent? Consider: ship preferences in actor fandoms, "favorite character" dynamics in ensemble show fandoms, position players vs. pitchers in baseball. How are these similar to and different from the K-pop bias system?

  2. Mireille manages a Discord where members have seven different biases. She describes navigating the "dynamics of a community where seven parasocial attachments compete for primacy." What governance challenges does this create? How might community rules need to address bias-related conflict?

  3. The solo stan — a fan exclusively devoted to one member, sometimes to the point of negativity toward others — represents an extreme specialization of parasocial attachment. Discuss: Is solo stan behavior a predictable outcome of deeply intensified PSR, or does it represent something qualitatively different? What would need to be true for a stan community to successfully incorporate both OT7 fans and solo stans without conflict?

Exercise 24.5: Sentiment Analysis Design (Research Methods, 60 minutes)

You have been asked to design a sentiment analysis study of a fan community of your choice (K-pop, MCU, Supernatural, or another fandom you are familiar with).

Design a study that addresses the following research question: To what extent do in-group and out-group sentiment patterns in this fan community differ, and what community events trigger sentiment shifts?

Your design should specify:

  1. Data collection: What platform would you collect data from? How many posts/tweets/comments? Over what time period? What sampling strategy?
  2. Post categories: What categories would you code posts into? (Do not simply copy Chapter 24's four categories — design for your specific community.) Justify each category.
  3. Tool choice: Would you use VADER, another sentiment tool, or a combination? Why? What are the limitations of your chosen tool for this specific community's language?
  4. Ethical considerations: What ethical issues does this research raise? (Consent, privacy, potential harm to the community being studied.) How would you address them?
  5. Interpretation plan: What patterns would you predict? What would confirm or disconfirm those predictions?

Exercise 24.6: The Harmful Fan Behavior Case Study (Ethics, 45 minutes)

Read the following scenario and respond to the discussion questions:

Scenario: A music journalist publishes a critical review of BTS's most recent album, arguing that the production has become "formulaic" and that the group's English-language output prioritizes Western chart performance over artistic integrity. Within 24 hours, the journalist receives 5,000 Twitter mentions, most of which are critical. Approximately 200 of these are personal attacks, including comments about the journalist's appearance and threats to report her account. Her employer receives emails asking for her dismissal. Several ARMY Discord servers, including one with 100,000 members, have pinned a thread about the review.

  1. Map this scenario onto the "overprotection dynamic" from section 24.6. Identify the cognitive and emotional steps from parasocial investment to the specific harmful behaviors.
  2. How should Mireille respond if this thread appears in her Discord? What governance tools does she have available? What should she say to her community?
  3. Is the music journalist's review protected speech that the fan community has an obligation to respect? Or does the community have legitimate interests in responding to criticism? Where is the line between vigorous disagreement and harassment?
  4. What would it mean for HYBE to exercise its "duty of care" (as discussed in Chapter 25) in this scenario? What, if anything, can or should the company do?