How to Give an Academic Presentation

Example and Analysis

Bogdan G. Popescu
bgpopescu@tec.mx

Tecnológico de Monterrey

PART I: THE PRESENTATION

Dinas, E., Matakos, K., Xefteris, D., & Hangartner, D. (2019). Waking up the Golden Dawn: Does exposure to the refugee crisis increase support for extreme-right parties? Political Analysis, 27(2), 244–254.

The Paper

Full title: “Waking Up the Golden Dawn: Does Exposure to the Refugee Crisis Increase Support for Extreme-Right Parties?”

Published: Political Analysis (2019)

Authors: Dinas, Matakos, Xefteris & Hangartner

Research Question

Does exposure to refugees increase support for extreme-right parties?

  • Economic threat?
  • Cultural threat?
  • Simple exposure enough?

Research Question

Context: Greece’s Golden Dawn

  • Neo-fascist platform
  • Anti-immigrant agenda
  • Parliamentary party

Golden Dawn rally, Athens (2012). Photo: Steve Jurvetson, CC BY 2.0.

Why This Matters

Substantive:

  • 2015: 850,000+ refugees via Greece
  • Largest displacement crisis since WWII
  • Political violence against refugees

Theoretical:

  • Tests contact theory
  • Challenges realistic group conflict
  • Isolates exposure vs. contact

Core Argument

The Authors’ Hypothesis

  • Claim: Refugee arrivals → Golden Dawn support
  • Not through economic competition (no work permits)
  • Not through sustained contact (48-hour transit)
  • Pure exposure effect

Key insight: Mere visible presence triggers backlash

Proposed Mechanism

graph LR
    A[Refugee Arrival] --> B[Visible Presence]
    B --> C[Perceived<br/>Cultural Threat]
    C --> D[Golden Dawn<br/>Voting]

    A -.->|NOT| E[Economic<br/>Competition]
    E -.->|NOT| D

    A -.->|NOT| F[Sustained<br/>Contact]
    F -.->|NOT| D

    style A fill:#F5F5F5
    style D fill:#F5F5F5
    style E fill:#F5F5F5,stroke-dasharray: 5 5
    style F fill:#F5F5F5,stroke-dasharray: 5 5

Data & Design

The Setting: Greek Islands

The setting:

  • Syrian refugees flee via Turkey
  • Cross Aegean to Greek islands
  • Transit to mainland in <48 hours
  • ~400,000 arrivals Apr–Sep 2015
  • 2.8 refugees per resident on affected islands

The geography and timing of the refugee crisis in the Aegean Sea. Source: Dinas et al. (2019), Figure 1.

Natural Experiment

Research design:

  • Instrument: Distance from Turkish coast
  • Treatment: Refugee arrivals per capita
  • Outcome: Golden Dawn vote share
  • Timing: January 2015 (pre) vs. September 2015 (post)
  • Units: 95 municipalities, 248 townships

Refugees crossing from Turkey to Lesbos, Jan 2016. Photo: Mstyslav Chernov, CC BY-SA 4.0.

Natural Experiment

Two approaches:

  • Difference-in-differences (treated vs. control over time)
  • instrumental variables (distance instruments for arrivals).

Main Findings

Effect Size

Treatment effect:

  • +2 percentage points in GD vote share
  • 44% increase over baseline (4.5%)
  • Statistically significant, robust

Dose-response:

  • 1 refugee per resident → +0.77 pp
  • Effect concentrated in exposed townships
  • Stronger with more refugees

Treatment vs. Control

Proximity to Turkey predicts change in Golden Dawn vote share (left). No such relationship before the crisis - the placebo test (right). Source: Dinas et al. (2019), Figure 4.

Political Consequences

  • GD became third-largest party in September 2015
  • Gained constitutional privileges
  • Votes shifted from center-right Nea Demokratia
  • Also mobilized new voters (turnout increased)

Critical Analysis

Limitation 1: Mechanism Inferred, Not Measured

The gap:

  • No direct measurement of attitudes
  • No survey data from islands
  • Authors infer cultural threat

Why this matters:

  • Can’t distinguish cultural threat from other factors
  • Could be local mobilization
  • Could be media framing

Limitation 2: Modest Effect Given Context

Contextual factors suggest a larger effect might be expected:

  1. Massive shock: ~400,000 refugees in 5 months
  2. High visibility: 2.8 refugees per resident on affected islands
  3. Concurrent crisis: Financial collapse, austerity
  4. Electoral salience: Vote during peak arrivals

Yet only +2 pp - suggests limits to backlash

Limitation 3: Narrow Generalizability

  • One country (Greece)
  • Temporary transit, not settlement
  • Pre-existing far-right party
  • Government-managed dispersal

Open questions:

Does this apply to permanent migration? Countries without far-right parties? Economic migrants?

Conclusion

Key Takeaway

Mere exposure to refugees - without economic competition or sustained contact - can moderately increase support for extreme-right parties.

Discussion question:

If transient exposure triggers backlash, what does this mean for countries serving as transit routes in future crises?

Thank you

Questions?

PART II: WHY THIS IS A STRONG PRESENTATION

The Rubric at a Glance

Criterion What “Strong” Means
Content Knowledge Nuanced, detailed summary
Critical Analysis Sophisticated, linked to broader frameworks
Organization Clear structure, smooth transitions
Clarity of Expression Clear language, appropriate terminology
Engagement Eye contact, enthusiasm, captivates audience
Visual Aids Well-designed, enhance understanding
Time Management Balanced summary and analysis within time

Content Knowledge

Selectivity: What to Include

What the presentation does:

  • Focuses on 2 core contributions: exposure effect and natural experiment
  • Omits regression tables, robustness checks, all control variables
  • Prioritizes the “story” of the paper

Why: Strong presentations show judgment - what’s essential vs. technical minutiae

Translating Technical Language

  • “Natural experiment” → distance creates random-like variation
  • “Difference-in-differences” → compare changes over time
  • “Instrumental variables” → distance predicts treatment

Every technical term paired with plain English

Incorporating Visual Evidence

Part I uses four figures from the paper:

  1. Figure 1 (geography) → establishes the setting
  2. Figure 2 (parallel trends) → shows the key visual pattern
  3. Figure 4 (placebo test) → validates the causal claim
  4. Mechanism diagram → clarifies the causal chain

Figures do the heavy lifting - you don’t need to read regression tables aloud

Critical Analysis

Specific vs. Vague Critiques

Bad critique:

  • “The paper is not well written”
  • “The paper is not clear enough”
  • “The methodology could be better”

Good critique:

  • “No survey data means the mechanism is speculative”
  • “+2 pp despite massive shock suggests limits to backlash”
  • “Transit context limits generalizability to permanent migration”

Each limitation names a specific gap, explains the consequence, and remains fair to the authors

Fair but Honest Assessment

  • Acknowledge what the paper does well before criticizing
  • Ground critique in the paper’s own evidence
  • Suggest what would strengthen the findings
  • Distinguish limitations the authors acknowledge from those they miss

Tone: Respectful colleague, not hostile reviewer

Organization

Evaluation Logic

  1. Question → What are they asking?
  2. Theory → What do they propose?
  3. Design → How do they test it?
  4. Findings → What did they find?
  5. Critique → What are the limits?
  6. Implications → What does it mean?

This is how you evaluate research, not just summarize it

Clear Signposting

Section dividers mark transitions:

  • Core Argument
  • Data & Design
  • Main Findings
  • Critical Analysis
  • Conclusion

Audience always knows: “Where are we in the argument?”

Clarity of Expression

Bullets as Prompts, Not Scripts

Research Design slide uses:

  • “Instrument: Distance from Turkish coast”
  • “Treatment: Refugee arrivals per capita”

Not: “The instrumental variable that the authors use in their analysis is the geographic distance from the Turkish coast…”

Fragments anchor speech, don’t replace it

Active, Direct Language

Passive (avoid):

  • “It has been hypothesized by the authors that…”
  • “Results were found to be significant…”

Active (use):

  • “Hypothesis: Refugee arrivals → Golden Dawn support”
  • “The effect is statistically significant and robust”

No hedging, no passive voice, no filler words

One Idea Per Slide

Test: “If I removed this slide, would the audience lose something essential?”

  • Effect Size → magnitude of the result
  • Political Consequences → what the result means in context
  • Parallel Trends → the key visual evidence

Separating ideas reduces cognitive load

Engagement with Audience

Engaging Techniques

What Part I demonstrates:

  • Provocative question: “What does this mean for future transit countries?”
  • Visual storytelling: Mechanism diagram, figures from the paper, photos
  • Incremental reveal: Fragments build the argument step by step
  • Eye contact: Bullet prompts free you to look at the audience, not read slides

Engagement = preparation + confidence + genuine interest in the topic

Visual Aids

Functional Design

  • Mermaid diagram: Proposed vs. rejected mechanisms - easier to grasp spatially
  • Paper figures: Fig 1 (geography), Fig 2 (trends), Fig 4 (placebo) - let the data speak
  • Photos: Golden Dawn rally, refugee boat - provide context without decoration

Test: “What does this visual do?” If no answer, cut it

Consistent Layout

  • Two-column slides for comparisons (strengths vs. weaknesses, good vs. bad)
  • Single-column slides for arguments that build sequentially
  • Full-width figures when the image is the point
  • Consistent color palette throughout

Variety prevents monotony; consistency prevents confusion

Time Management

Time Budget

  • Introduction + Context: ~2 min
  • Core Argument: ~2.5 min
  • Data & Design: ~3 min
  • Findings: ~2.5 min
  • Critical Analysis: ~3 min
  • Conclusion: ~2 min

~15 minutes total. Rule of thumb: ~1–1.5 minutes per slide. Front-load complexity, keep the end crisp.

Putting It All Together

Core Principles

  1. Content: Focus on unique contributions, not every detail
  2. Critique: Specific gaps + consequences, fair to authors
  3. Organization: Question → evidence → evaluation → significance
  4. Clarity: Bullets as prompts, jargon translated, active voice
  5. Engagement: Provocative question, visual storytelling, eye contact
  6. Visuals: Functional, not decorative - every element earns its place
  7. Time: ~1–1.5 min/slide, front-load complexity

Common Mistakes to Avoid

  • Reading slides word-for-word
  • Including every table from the paper
  • Vague critiques (“not clear enough”)
  • Decorative images that add nothing
  • Running over time by cramming too much
  • No structure - jumping between topics

Remember: A presentation is not just a summary. It is an evaluation.