Effective Science Talks, Celia M.

Elliott

Summer 2010

Effective Science Talks

Physics Made Simple

Celia M. Elliott
Board of Trustees of the University of Illinois Urbana, Illinois 2010

Department of Physics University of Illinois

This talk is divided into two parts. In the first part, well talk about the nitty gritty of putting together a good talk: Why scientists give talks Goals for a talk How to organize a talk Presentation software tips Effective figures Presenting numerical data Handling questions The second part consists of some practical advice from the veteransthings to do, and not do, when youre preparing and giving your talk. DISCLAIMER: Do as I say, not as I do. The slides for this talk are standard for an academic lecture. They have way too much text on them for a science talk. In a science talk, you want the audience to look at engaging, visually interesting, instructive images and think about the concepts being presented, not read words.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Never rise to speak till you have something to say, and when you have said it, cease.

Witherspoon, John (1723-1794), was the sixth president of Princeton, a signer of the Declaration of Independence, and from 1776 to 1782 a leading member of the Continental Congress. He came from Scotland in 1768 to assume the presidency of the college and held office until his death a quarter of a century later. Although Witherspoons advice was no doubt influenced by his experience in the Continental Congress, it remains good advice for scientists. Dont attempt to give a talk until you are thoroughly prepared, say what you have to say, and then sit down.

Why science needs talks: Publications lag months to years behind discoverytalks get ideas out into the community and move science forward. Putting your thoughts into words and organizing them will crystallize your thinking and give you insights you can gain no other way. Your future job will require presentations: You will give talks as a job candidate Your will give presentations as an employee You will give presentations as an instructor

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Before you pick up a pencil, answer four strategic questions...

What is your motivation for giving the talk? To disseminate your results to other workers in your field. To test your ideas on other scientists and get their feedback. To establish precedence by announcing your results before your competitors can. To teach the audience something. To achieve fame and fortune; to get noticed or hired; to establish future collaborations; to gain the respect of the community. What is the purpose of the talk? Reporting original, significant research results. Documenting methods or establishing standards. Warning of a hazardous condition. Examining the feasibility of a project. Reinterpreting previously reported results. Providing an overview of the topic for non-experts. Who is the audience for this talk? What are their needs, interests, level of knowledge, motivation for listening? Can you talk about it at all? Do you have permission from your adviser and your collaborators?You cannot patent anything that has been publicly disclosed.

Copyright, of Trustees the of the Copyright Board 2010 The Board of of Trustees Universityof ofIllinois Illinois University

Effective Science Talks, Celia M. Elliott

Summer 2010

Essentials for preparing your talk

Know your audience! Determine the style of your talk; decide on the structure that best fits your audience and your message Find out how much time you have to speak Decide on the key points you want to communicate Determine how best to use graphs and figures to illustrate your key points Consider effective slide aesthetics

This is a horrible exampledo not present slides like this at your talk. It has way too much text and zero visual interest. It tells your audience I might be able to be more boring, but I frankly dont think its worth finding out for the likes of you. In the next few slides, well look at each of these concepts in more detail, and Ill present examples to guide you as you craft your own talk.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Rule #1: Know thy audience!

A successful talk is tailored to the listeners wants and needs

Informal seminar Scientific conference

Report to funders Job interview

Who are they? What is their level of expertise? How motivated are they to listen? How large is the group? The size of the group will affect your presentation stylewill you need to prepare slides that can be projected in a large room, or will you all be sitting around a table? What do you want your audience to do for you? - Give constructive feedback? - Learn about what you are reporting? - Participate by asking relevant questions? - Give you new ideas or insights? - Hire you? Give you a grant? - Buy your product? Use what you know about the audience to build rapport with them. The audience must want to pay attention to you. You want to earn their respect. If you do not build a good relationship with your audience, they will not listen to you, no matter how brilliant or groundbreaking your research is. What two or three key points from your talk should they take home? What background information do they need to understand these points? If you can answer these questions, you can prepare your talk to help you get what you want and need.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

The next biggest constraint: How much time do you have?

Presentation math:

t p, 10
where t is the time in minutes you are allotted for your presentation, and p is the number of points you can make without losing your audience

The amount of time youre allotted determines how much material you can cover in your talk. It takes about 10 minutes to adequately introduce, explain, and summarize one major idea or point in an oral presentation

You cannot present 10 major ideas in a 15-minute APS-style talk, no matter how fast you talk.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Key Goal: Communicate your ideas!

Our group is making seminal d iscoveries ,

so you should pay attention to us!

This fundamental goal should govern every aspect of the design and presentation of your talk!
You are not here to tell the audience everything that is in your publications. You are here to make them interested enough so that they want to look them up and read them! You must, however, give your audience one or two important points to take with them now.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

How do you start? Write down the two to three key ideas you wish to convey!

The introductory material flows from these ideas - What motivated the work? - What background information does the audience need to understand these points? The body of the presentation also flows from these ideas - What supporting evidence and data must be presented? - How can you most effectively present those datain text, figures, graphs, equations? N.B. In most cases, text is the worst way to convey scientific data. For a wonderful introduction to how to present quantitative information, see Edward Tuftes Visual Explanations (Cheshire, CT, Graphics Press, 1997).

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Setting the overall structure of the talk also follows from the key points
Motivate the key messages (Introduction) Preview your main messages (Introduction) Provide support for your messages (Body) Summarize your messages (Conclusion)

In other words, dont let em leave without getting your main messages!

Ideally, youll convey your main messages to the audience three times during the presentationfirst in the overview or introduction, next in the body of the talk, and finally on a summary slide. Take it from a mothertelling somebody something three times is not overkill.

Copyright 2010 The Board of Trustees of the University of Illinois

Effective Science Talks, Celia M. Elliott

Summer 2010

Organizing a 20-min talk

Background and Introduction (24 minutes)
Title slide Overview slide 12 additional slides, if needed Body (912 minutes) Develop only two or three main ideas (2 slides ea) 57 slides Summary (1 minute) 1 slide Questions (3 minutes) (Know thy audience!) 34 back-up slides

Follow some simple rules of thumb: If youd write or draw something on the blackboard or a piece of paper while explaining your ideas to a friend, make a graphic of it. Allow about 2 minutes per slide. Allow more time for the audience to process slides that present: Equations. Complicated schematics. Numerical data in tables or graphs. Back-up slides; consider likely questions or objections and make a slide to answer each of them.

Copyright 2010 The Board of Trustees of the University of Illinois

10

Effective Science Talks, Celia M. Elliott

Summer 2010

The title slide and outline prepares the audience to listen and tells it what to look for
Title slide
Your name and affiliation Venue and date Attention-getting graphic

Outline or overview of presentation*

Prepares the audience to listen Provides a logical structure for your talk Summarizes key points (limit to two or three for a 20-min talk)
*Tip: An outline isnt necessary for a short talk

Use a combination of slides and handouts to deliver your message Use slides to: Emphasize main points Illustrate experimental apparatus, schematics, samples, photographs or simulations of results Present and summarize data Use printed handouts to: Facilitate note-taking Reinforce main points Convey complicated information, e.g., numerical detail or equations Provide additional details and contact information

Copyright 2010 The Board of Trustees of the University of Illinois

11

Effective Science Talks, Celia M. Elliott

Summer 2010

The body of your presentation is the intellectual content of your talk

Problem statement, motivation; prior work 12 slides Method
12 slides

Results
3 slides

Future work
1 slide

Have one Problem Statement slide that tells the audience why your work is important and why they should listen to you. How does it extend prior work? What important question have you answered? Methodkeep this section short unless you are employing an exciting new method, which is one of the main points you want to make. If the audience wants to know the exact composition of your samples and where you set the dial, theyll read the paper. Resultsthis is what the audience came to hear. The results section should be the longest part of your talk and should provide the most detail.

Copyright 2010 The Board of Trustees of the University of Illinois

12

Effective Science Talks, Celia M. Elliott

Summer 2010

Use figures to illustrate your key points

Figures promote audience interest, provide supporting evidence, help explain complex ideas and relationships quickly, and give the audience something to remember
Myosin walking on actin Courtesy of P. Selvin

Figures: 1. Promote audience interest. 2. Provide supporting evidence. 3. Help explain complex ideas and relationships quickly. 4. Give the audience something to remember.

Copyright 2010 The Board of Trustees of the University of Illinois

13

Effective Science Talks, Celia M. Elliott

Summer 2010

Who can tell me the four reasons from the previous slide for why you should use figures to illustrate your talk?

Who can describe the image on the previous slide?

I rest my case.

People remember images, even after only one fairly brief exposure. They dont remember words. Design your talk to be visually interesting and memorable.

Copyright 2010 The Board of Trustees of the University of Illinois

14

Effective Science Talks, Celia M. Elliott

Summer 2010

Not all figures are created equal

Good figure visually interesting, attractive, and memorable

Courtesy Gerard Wong

Bad figure visually boring, contentless, and forgettable Tip: If a picture isn't worth a thousand words, to hell with it Ad Reinhardt

First figure could be better; the black background does not offer high-enough contrast, and it is not clear what the white arrows in B are supposed to mean. (uncondensed and condensed actin filaments; at high multivalent ion concentrations, the ions collectively form a CDW and bundle actin filaments)

Copyright 2010 The Board of Trustees of the University of Illinois

15

Effective Science Talks, Celia M. Elliott

Summer 2010

Label all elements in a figure

Point out important features Label both axes of graphs and show units Provide a scale Give credit

The Nike laser system uses discharge pre-amplifiers.

(Courtesy US Navy)

Sample normalized signals from the twobeam optical drive.

(Courtesy C. Michael)

Scientists differ on whether captions are needed for figures on slides. (Theyre absolutely mandatory for figures in publications.) I personally recommend that figure captions be eliminated or kept very brief, but point out important features so the audience knows what its supposed to be looking at. If youve used somebody elses figure, you should at a minimum give credit for it, and perhaps provide a URL or bibliographic reference for where the original may be found. Another tip for ALL figuresif you show a photograph or drawing of something, provide some sort of visual clue to its scale. The audience may have no idea if the apparatus shown below is 5-cm long or 5-m long from just looking at this image.

Copyright 2010 The Board of Trustees of the University of Illinois

16

Effective Science Talks, Celia M. Elliott

Summer 2010

Use graphs and tables to present numerical data

Use to show trends or reveal relationships Specify units of measure (in SI units) Provide a title for each graph or table

Avoid showing tables of raw numerical datapeople just cannot process it and listen to you at the same time. If you absolutely have to show a table (and the size of the audience is amenable), make some hard copies and pass them out.

Copyright 2010 The Board of Trustees of the University of Illinois

17

Effective Science Talks, Celia M. Elliott

Summer 2010

Keep graphs and tables simple convey ideas, not raw data
Verbosity Index
100 Comprehension, % 80 60 40 20 0 10 20 30 40 50
Regrettably, does represent actual data
Women in Top-Ranked Physics Ph.D. Programs (1998)
University Harvard Princeton MIT California-Berkeley Cal Tech Cornell Chicago UIUC Stanford California-Santa Barbara NRC Rank/Score 1 / 4.91 2 / 4.89 3 / 4.87 4 / 4.87 5 / 4.81 6 / 4.75 7 / 4.69 8 / 4.66 9 / 4.53 10 / 4.43 PhD Students 149 110 315 283 154 182 154 295 135 117 Women % 13 13 10 9 18 18 14 8 13 13 PhD Recipients % Women 14 3 12 8 8 12 6 7 12 5

Words per Sentence

Illustration only, does not represent actual data

The example on the right shows how you can present tabular data in a form that people listening to your talk can immediately process. By highlighting the relevant line, you convey the main ideathat UIUC was ranked far down the list. The audience probably doesnt care that Illinoiss score was 4.66 and Harvards was 4.91; they care that Illinois is ranked toward the bottom of its peers, and its percent of women was in single digits. (Weve improved since 1998.)

Copyright 2010 The Board of Trustees of the University of Illinois

18

Effective Science Talks, Celia M. Elliott

Summer 2010

Use equations only if absolutely necessary to convey your message

If you use equations

Slow down; talk through step by step Explain relevance Make them large enough to be easily read Define your terms!
PowerPoint animations can be useful in presenting equations: Highlight relevant terms in different colors Drop out terms Blow up parts of the equation as you walk the audience through it

Copyright 2010 The Board of Trustees of the University of Illinois

19

Effective Science Talks, Celia M. Elliott

Summer 2010

Provide a summary slide

Recap key results Reiterate principal conclusions Repeat your contact information

The summary slide lets you reiterate your key points and cues the audience that you will soon be taking questions. This slide will probably stay on the screen during the question period and will thus get the longest audience exposuremake it count!

Copyright 2010 The Board of Trustees of the University of Illinois

20

Effective Science Talks, Celia M. Elliott

Summer 2010

Handling questions is an essential part of giving a talk

Always repeat the question What if you dont know the answer? If the questioner disagrees, dont argue Never insult the questioner

Always repeat the question (summarize or paraphrase it) before you plunge ahead with your answer. Not everyone may have heard it, and repeating it not only allows the questioner to clarify if youve misunderstood, it also gives you a few precious seconds to think about your answer. If you dont know the answer, dont bluff! Simply say, Thats an excellent question. We havent looked at that. or Im not sure; Ill have to think about that. Its okay not to know the answer; its not okay to make something up on the fly. If the questioner disagrees, or wanders too far off-topic, you can always say, Lets talk about this further after the session...

Copyright 2010 The Board of Trustees of the University of Illinois

21

Effective Science Talks, Celia M. Elliott

Summer 2010

Advice from a veteran...

From Professor Lance Cooper: Maintain eye contact with audience; dont stare at the monitor or read off the screen. Making eye contact with the audience will build rapport with them and will actually make you feel less nervous. Do not read your talk! Its okay to look at notes, but know your material well enough that you can speak about your points naturally. That takes knowing your material thoroughly and practicing. Avoid nervous mannerismspacing, bobbing, waving arms, jingling coins. User laser pointer or stick directed at screen Dont point directly at overhead on projector Dont block the screen Train yourself to speak slowly and distinctlypractice! Avoid verbal fillers: err, like, um, okay. Turn off your cell phoneand anything else that could distract you. Be enthusiastic! If you dont act excited by your results, dont expect the audience to be!

Copyright 2010 The Board of Trustees of the University of Illinois

22

Effective Science Talks, Celia M. Elliott

Summer 2010

Remember, your goal is to convey your ideas, so avoid distracting text and effects!

Use PPT features judiciously and sparingly. Dont annoy the audience with cheesy text animations, distracting backgrounds, and obnoxious sound effects.

Copyright 2010 The Board of Trustees of the University of Illinois

23

Effective Science Talks, Celia M. Elliott

Summer 2010

Replace the content-less PPT title with a meaningful motivating statement* for the slide

*Tip: Write the statement as a sentence and left-justify it.

Copyright 2010 The Board of Trustees of the University of Illinois

24

Effective Science Talks, Celia M. Elliott

Summer 2010

Slide aesthetics are important

Dont use calligraphy

or serif fonts

USE THE SAME FONT

throughout the talk

Avoid distracting backgrounds

Use sans serif fonts; they show up much better when projected on a large screen.

Copyright 2010 The Board of Trustees of the University of Illinois

25

Effective Science Talks, Celia M. Elliott

Summer 2010

Choose an easy-to-read font (36 pt)

Make sure your audience (32 pt)
Can easily read (28 pt)
Every one of your slides (24 pt)
From the back o

f the room
(14 pt)

(20 pt)

See what I mean?

The larger the room, the bigger the font size! If the room is not full, have the members of the audience come up to the front of the room. Youll establish an immediate rapport with them if you invite them to come up and be a part of your small, select circle.

Copyright 2010 The Board of Trustees of the University of Illinois

26

Effective Science Talks, Celia M. Elliott

Summer 2010

Choose colors carefully*

Strive for easy reading Strive for easy reading Strive for easy reading*

*Tip: Avoid using red and green. Between 8 and 12 percent of white males are red-green colorblindwhos your audience?

LCD projectors change color appearance; text and background that looks fine on your computer screen may look entirely different when it is projected to an image 1.5-m high and 2-m wide. In particular, pastel colors disappear when projected; use a neutral background with a high-contrast, dark text. Dont use color randomly; people expect color to mean something. Avoid using red and green. To see what your image will look like to someone with color blindness, theres a very useful, free emulator at http://aspnetresources.com/tools/colorblindness.aspx.

Copyright 2010 The Board of Trustees of the University of Illinois

27

Effective Science Talks, Celia M. Elliott

Summer 2010

Consider the appearance of the text on the slide

Arrange line breaks so that the text is not interrupted in awkward places.

In PowerPoint, type SHIFT + ENTER to insert a manual line break.

Copyright 2010 The Board of Trustees of the University of Illinois

28

Effective Science Talks, Celia M. Elliott

Summer 2010

Embed special fonts in PPT to avoid unsettling surprises

The Strickler-Berg relation opens the door for comparing measured spectral quantities

Strickler-Berg Relation

Different computer Voila! pencils

(1). Open the document in PowerPoint. (2). Click on the "Tools" tab on the top menu. (3). Click on the "Options" link. (4). Click on the "Save" tab. (5). Locate Font options for current document only and Embed TrueType fonts. (6). Click in the check box to turn on the option. (7). You have to do this for every presentation; PowerPoint does not automatically embed fonts unless you tell it toevery time.

Copyright 2010 The Board of Trustees of the University of Illinois

29

Effective Science Talks, Celia M. Elliott

Summer 2010

Critique this slide

Copyright 2010 The Board of Trustees of the University of Illinois

30

Effective Science Talks, Celia M. Elliott

Summer 2010

This slide is an example of an eating the elephant slide

Where do you take the first bite??

Its too busy, and its really ugly.

Copyright 2010 The Board of Trustees of the University of Illinois

31

Effective Science Talks, Celia M. Elliott

Summer 2010

End with a bang, not a whimper

Dont trail off with an ineffectual Well, I guess thats it John Witherspoons advice aside, dont just stop and let audience guess that youre done. Put up a summary slide, reiterate your two or three important points, thank the audience for their attention, and ask for questions.

Copyright 2010 The Board of Trustees of the University of Illinois

32

Effective Science Talks, Celia M. Elliott

Summer 2010

Finish on time!

Practice your timingyou will get cut off unceremoniously at conferences. Set your watch for a two-minute warning Do not compensate for having too much material by trying to talk faster Simplify. Cover fewer points. Eliminate slides. Think about the importance of each slide. What if, for some reason, your talk must be shortened by five or ten minutes? What slides would you take out? Use the hide slide feature in PPT to easily remove slides.

Copyright 2010 The Board of Trustees of the University of Illinois

33

Effective Science Talks, Celia M. Elliott

Summer 2010

The best way to prepare for a talk is know your material*

It takes three weeks to prepare a good ad-lib speech

*Tip: Knowing your material thoroughly is also the best antidote to stage fright.

Practice, practice, practice, practice Focus on communicating, not performing. Humor is good, but it must be relevant to your subject matter, in good taste, and understandable across cultures. Keep it simple. Prepare key phrases (Notes Pages) Okay to write out material first. Write down each slides main point. If the slide doesnt have a point, eliminate slide! Stay on track. Small digressions are fine if motivated by a question (shows you are paying attention to audience), but get back on track.

Copyright 2010 The Board of Trustees of the University of Illinois

34

Effective Science Talks, Celia M. Elliott

Summer 2010

Check everything before your talk

Check the projector
Make sure you know how to turn it on See that it is plugged in Check which way to position your slides Adjust the focus

Check microphones, pointer, other tools If a clip-on mike is used, make sure it is fastened securely, check the volume, and then leave it alone Arrange your slides, notes, and other materials so you can reach everything without fumbling

Do not expect the conference organizers to take care of all of your needs if you do not tell them what they are ahead of time. Did you request an overhead projector? Slide projector? video-tape player? Make it easy for the audience to pay attention. Clean up clutter; if the room next door is noisy, ask them to quiet down, adjust the lighting. Arrive ahead of your appointed time. Dont dash in at the last minute, panting and out of breath, in your coat, umbrella, galoshes, with a bag of exhibit-hall geegaws. If you are nervous, take some deep breaths. Do not daydream during the presentation just before yours! Be patient if the session is running behind schedule, and be flexible if the moderator asks you to adjust your talk to take a lesser amount of time.

Copyright 2010 The Board of Trustees of the University of Illinois

35

Effective Science Talks, Celia M. Elliott

Summer 2010

A word about appropriate dress

*Tip: Wear comfortable clothes that present a professional appearance.

The day of your talk is not the day to try out your new thong underwear or strapless underwire bra. Wear comfortable shoes. Wear a shirt or blouse that you can clip a portable microphone to, so that it is positioned about 56 in. below your mouth. Turtlenecks and tee shirts should be avoided, because theres no good place to clip the microphone where it wont slip. Wear slacks or a skirt with a waistband or pockets for the microphones power supply.

Copyright 2010 The Board of Trustees of the University of Illinois

36

Effective Science Talks, Celia M. Elliott

Summer 2010

If English is not your native language (and even if it is!)

*Tip: Watch for cues from the audience; if they looked confused, they dont understand you.

1, Do not use jargon unless you explain it (SPH). 2. Choose the simplest wordsimagine that you are giving a talk in English to someone who doesnt speak English as a first or even second language! 3. Do not be embarrassed to ask a native speaker to review your presentation. 4. Practice speaking slowly and distinctly.

Copyright 2010 The Board of Trustees of the University of Illinois

37

Effective Science Talks, Celia M. Elliott

Summer 2010

Express your thanksbriefly

At the beginning of your talk: Acknowledge colleagues and collaborators who contributed to the work. Thank the conference organizers for allowing you to speak. At the end of the talk: Thank your audience for their attention Keep your thanks very brief.

Copyright 2010 The Board of Trustees of the University of Illinois

38

Effective Science Talks, Celia M. Elliott

Summer 2010

Remember your purpose in preparing and giving a talk Your purpose is to tell an interesting, memorable story of your work
Not Not Not Not to read a book to demonstrate how smart you are to attack others work to show how fast you can talk

Copyright 2010 The Board of Trustees of the University of Illinois

39

Effective Science Talks, Celia M. Elliott

Summer 2010

To recap
Identify two to three main points that you want to convey to the audience Design your talk to make these points as clearly as possible Preparation is crucial Graphics enhance audience interest and retention Be relaxed and (try to) keep audience attention Finish on time Giving good talks is a learned skill; the more you practice, the better you will get
Celia M. Elliott University of Illinois [email protected]

Copyright 2010 The Board of Trustees of the University of Illinois

40