How to Give a Good PowerPoint Lecture, 2012

Posted by Chad Orzel on March 5, 2012
(6)
More »

My timekeeping course this term is a “Scholars Research Seminar,” which means it’s supposed to emphasize research and writing skills. Lots of these will include some sort of poster session at the end of the term, but I decided I preferred the idea of doing in-class oral presentations. Having assigned that, of course, I felt I ought to give them a class with advice on how to give an oral presentation. I went looking for advice on this, and found that I wrote a guide to giving good PowerPoint lectures back in 2006 (God, I’m a blogging dinosaur…), which holds up pretty well. So, I dusted that off, and made up some slides on the topic, deliberately demonstrating some of the things people do wrong:

How to Give a Good PowerPoint Presentation
View more presentations from Chad Orzel.

Just to liven things up, I also decided to show them an example of somebody doing a really good job of giving a talk with PowerPoint, so I found a short TED talk with a vague connection to timekeeping, and played that at the end of the class:

It’s a very entertaining talk, but it also demonstrates most of the good things you should do with PowerPoint: his slides are mostly fairly iconic images, when he has text, it’s huge, and limited to only the most relevant items, and he very clearly knows exactly what comes next at every point. It’s not something I expect first-year students, even very bright ones, to achieve, but it’s a goal to aspire to. Reach should exceed grasp, etc.

Anyway, since I have this, I thought I should throw this out for the general edification of the Internet. And because I’m still pretty happy with that old post that I wrote.

(One of my “infintie free time” projects is to go back through the nearly-ten-year archive of this blog, and pull together a “best of” collection. In my infinite free time…)

(6)
More »

Comments

  1. #1 Chad Orzel
    March 5, 2012

    A couple of notes about things that may not immediately be clear from the slides themselves:

    The list of stuff on the “Limit Your Material” slide is a list of statistics relating to my own presentations: a typical lecture averages around 20 slides, my standard research talk is 26, my wonders-of-science-blogging talk is 32, and my “What Every Dog Should Know About Quantum Physics” talk is 39. Those are all conceived of as 50-minute (-ish) lectures.

    The oddly spaced text on the “Use Animation Sparingly” slide is a bunch of single words, each with a different animation effect.

    The Twilight book covers are a reference to the fact that nearly every writer I’ve heard talk about them says that those books are awful, breaking all sorts of rules about how to write novels, and yet they’ve sold a bazillion copies, so clearly they work for somebody.

  2. #2 Eric Lund
    March 5, 2012

    Another useful resource is the satirical “Guidelines for Giving a Truly Terrible Talk”, several copies of which are scattered about the internet. Many of the pitfalls of PowerPoint were around in the era of 35-mm slides (the original version of it was apparently titled “35-mm Slides: A Manual for Technical Presentations”), which would be before my time. PowerPoint did two things: (1) made it easier to create a bad presentation, and (2) added animation to the mix.

    A definite ditto on the one slide per minute ceiling–anything more than that is not merely asking for trouble but sending an engraved invitation.

    Another personal pet peeve: Your summary should fit into a single slide. Especially for conference talks (20 minutes is generous for one of those), if you can’t fit the main points onto a single slide, you’re trying to cover too much ground.

  3. #3 CCPhysicist
    March 5, 2012

    I like your use of bad examples in your instructions, but telling them that 20 slides is OK was a bad move. I’ve seen a pro go through them at several a minute, but many of those had only one word on them. Like “1932″. Only way to improve that TED talk would have been to have punch lines like that fill the screen!

    How many minutes did it take to run through your 11 slides?

    That should be their reference point.

    And my tip: Don’t talk about the last slide. Just put up the conclusions and ask for questions.

    PS – The number one rule: Tell them what you are going to tell them, tell them, then tell them what you told them.

  4. #4 Tom Singer
    March 5, 2012

    If at all possible, run your presentation through a handy projection system, in the same lighting conditions that you’ll give it under. Things will show up differently on a projector screen than they will on your monitor.

    Remember that you’re presenting to your audience, not to your screen.

    And, Chad? The red text on dark grey background, particularly slides 5 and 7. That was intentional, right? :-)

  5. #5 jim
    March 6, 2012

    I did something similar a few years ago. I took a bunch of short clips from TED talks showing people using slides well: blank screen when they wanted the audience to concentrate on their words; pictures, photographs, little text; diagrams, animated diagrams. I gave some advice: one slide per idea, one idea per slide; the slides aren’t for you, they’re for your audience. I put up one slide that said in as large a font as would fit: DON’T READ THEM. Showed a picture I got off the net of a guy facing the screen reading the big block of text on his slide: “Look at him. Even his co-presenter looks bored. Don’t be like him.” And I ended with about six minutes of a Larry Lessig presentation, stopping it at intervals to ask them: “What three things about that anecdote does he want you to remember?” “Why was that a punch line?”

    But it didn’t do much good.

  6. #6 CCPhysicist
    March 11, 2012

    You should share some of the things in this IHE article

    http://www.insidehighered.com/views/2012/03/09/essay-how-apprentice-prompted-change-professors-requirements

    with your students. Also, I’ll add that a talk I saw this past week reminded me that no one ever complains if you finish a minute or two early as long as you deliver on the main promise of talk. Best of all, the speaker could have gone on for hours about one esoteric detail that I infer is her primary contribution to the collaboration, but only gave it the 10 seconds it deserved for the audience that heard the talk.