Social and Political Data Science: Introduction

Data Visualization

Karl Ho

School of Economic, Political and Policy Sciences

University of Texas at Dallas

How to give effective presentation of data visualization?

Killer presentation:

#1 Skill Irreplaceable in the Data Science Revolution Age. 


This is the one tool we have to outsmart smart machines.

Prepare yourself



  1. Watch Great Persuaders

  2. See Yourself as a Chief Storytelling Officer

  3. Reframe and Rehearse 

Prepare yourself: Tools



  1. Online Slides


    2. speaker deck

    3. Prezi

    4. Slideshare

  2. Quarto Presentation (reveal.js)

  3. Keynote vs. Powerpoint

  4. LaTeX Beamer

Prepare yourself: Tools



How to make great (not good) slides

  1. Know your tool
  2. Legible fonts (never use default)
  3. Figures > Tables > Text
  4. Cognitive Science
    • Fancy elements (e.g. interactive, reactive)

      Consider adopting novel design solutions only when the estimated payoff is substantially greater than the cost of learning to use them.


Prepare yourself: On stage



  1. Always look at the audience

  2. Tell us your research question at the beginning of your talk.

  3. Spend a minute or two explaining why this question is important and original.  You are charged with responsibility that you are not copying other people’s papers.

  4. Again, make sure you look at the audience (you see me, I see you!)

Prepare yourself: On stage (cont'd)

  1. Spend as little time as possible summarizing the literature.  But this is staple too. Focus on your argument.

  2. Presenting a table with quantitative results is fine, but a figure illustrating your findings is better.

  3. When presenting quantitative evidence, focus on the hypothesis you are testing.

  4. Repeat the big take-away points from your paper at the end.

Prepare yourself: On stage (cont'd)

  1. Reserve one minute for Q&A                     



Helpful Tips

  1. Get your own remote control

  2. Arrive early

  3. Control your own techs.

  4. TED talk

  5. Read books (not online materials)

For Data Visualization

  1. Mind your detail

    1. Theme

    2. Color

      1. font

      2. frame and grid

    3. Font (depart from default)

  2. Know your environment (stage, projector)

    1. Be ready to change theme/background color

  3. Presentation proportional to:

    1. Effect = \(f(Main  takeaway, time, skills, audience )\)