Articles

I always recommend using a color palette when you build slides. You will get your work done faster, enjoy building slides more and the slides will look more professional. One of the best places for creating a color palette is Kuler.

After you’ve created your color palette, you’ll want to program it into PowerPoint. Let me show you how. First, here is the color palette I use.

To build a custom color palette, the first thing you’ll do is go to Design > Colors > Create New Theme Colors. You’ll meet an interface that looks like this. It’s a bit confusing, so let me explain what you’re looking at.

1. Accent 1. This will be the default color when you draw lines and shapes. I recommend using gray for Accent 1, or whatever color you want your shapes to automatically contain.

2.Accent 2-6. The 5 colors from your color palette go into Accent 2-6. If you have more than 5 colors in your color palette, you can add them as Text/Background colors (described later).

 3. Hyperlinks. For hyperlink colors, choose one of your darker accent colors. For followed hyperlink, choose a lighter version of that accent color. Light colors indicate it’s been “depleted”

4. Text/Background – Dark 1 and Light 1 are your text colors. Dark1 can be black. Light1 can be white. These colors show up in the first two columns of your color menu and if you choose black and white you’ll end up with a complete range of grays in increments of 10%. Or you can choose different text colors if you prefer.

5. Text/Background – Dark 2 and Light 2 are your background colors. To be honest, these are a bit of a mystery to me since I use a white background for business slides or one of my accent colors if I need a color background. If you have more than 5 colors in your color palette, you might add them here, or just choose 2 extra colors you think you might need.

Now just name your new color theme, click “Save” and you’re done. Once your new color theme is saved, you can select it any time and your Color Picker pull-down menu will show your custom color palette, ready for use.

About the author: Bruce Gabrielle is author of Speaking PowerPoint: the New Language of Business, showing a 12-step method for creating clearer and more persuasive PowerPoint slides for boardroom presentations. Subscribe to this blog or join my LinkedIn group to get new posts sent to your inbox.

I read a lot of research papers on the most effective way to present information. So I was intrigued, and then disappointed, to read this paper testing one model, called the Assertion-Evidence model.

The study appears to prove picture slides with a full sentence slide title are more effective than bullet point slides with a topic title – at least for complex concepts. However, the research is badly flawed and the conclusions drawn are overly broad and even biased.

In this blog post, I want to review what the research purports to say, why those conclusions are skewed by poor research design, and what the research actually says.

1. Research Results

In 2006 Michael Alley, author of The Craft of Scientific Presentations, first proposed the Assertion-Evidence model for building slides, where the slide uses a full sentence slide title (the “Assertion”) and then a picture with limited text as the slide content (the “Evidence”).

In 2011, Alley continued studying the A-E model, comparing it against the Topic-Subtopic model; the typical PowerPoint model that uses a brief topic slide title and a list of bullet points, often including a relevant picture.

The results show that, for “complex content”, students learned more from the A-E slides than the T-S slides. On the other hand, they remember statistics better when they are on the slide than when they are simply spoken by the presenter. Alley draws the following conclusions

1. The A-E model is superior to the T-S model for complex content

2. The A-E model and T-S model are similar for remembering statistics, as long as they are included as text on the slide

3. The T-S model is superior for remembering secondary facts that are written on the slide, rather than merely spoken. However, this comes at the expense of understanding complex content

These conclusions are misleading and based on faulty research design. Now I look at the flaws in this research study and propose a new set of conclusions.

2. The Flaws

Michael Alley and his colleagues are to be commended for breaking new ground and testing different models to find the best way to use PowerPoint for educational purposes. My critique of his research is not to be negative, but to caution readers against rejecting other models that might serve your purposes even better.

The main problem with this research is the T-S slides were a mix of two types of slides: Type 1) slides with text only and Type 2) slides with text plus a relevant picture. In this study, 6 of the 11 T-S slides were Type 2 slides.

Type 1) Text-only	Type 2) Text-plus-picture

These Type 2 slides are pure poison. Poison! We’ve known for the past 10 years presenting a slide that combines extensive text plus a picture harms learning. This principle – called the Redundancy Principle –  has been proven in studies by both Richard Mayer and by John Sweller.

The researchers should not have combined text-only and the poisonous text-plus-picture slides into the T-S model. This just gives an unfair advantage to A-E slides that isn’t even worth testing.

Instead, they need to test the A-E model against three other models, which have NOT been proven to harm learning.

1. Full sentence title and text-only

2. Topic title and text-only

3. Topic title and picture-only

Another flaw is the apparent researcher bias. Their goal appears to be to prove the A-E model, not objectively test its merits. This is a shame. The A-E model looks promising, but it’s only by being intellectually honest they will discover its limitations and improve it further.

Consider the way they ignore data that doesn’t support the A-E model. The students wrote a test that included five multiple choice questions. Using the A-E model, students did better on only three questions. What did the researchers do? They ignored the other two questions and only analyzed the three questions (questions 2-4) where the A-E model was superior.

Consider also how the researchers dismiss the fact that students remembered statistics better on the T-S slides, based on fill-in-the-blank questions on the test. The A-E slides do not contain those facts as written text, only as spoken text by the presenter. The researchers dismiss it as not important enough to offset the advantage of A-E slides for learning complex content, rather than unemotionally acknowledge the superiority of T-S slides in some learning situations.

3. The Real Conclusions

You cannot draw conclusions when the two slide models have so many things that are different. Are the higher scores because of the poisonous Type 2 slides? Or the superiority of pictures over bullet points? Or the full sentence slide title? Or the animations only included on the A-E slides?  Which of these actually explains the difference in performance? It’s impossible to know for sure.

But some things you can conclude are:

1. For important statistics, include them as text on a slide, not spoken text. Students remembered statistics better when they were included as slide text, rather than simply spoken by the presenter. This suggests there will be times slides with a lot of text will outperform slides with limited text and pictures.

2. Pictures don’t help students remember statistics. When the slides contained the statistics as text, students did just as well with the A-E slides and the T-S slides. The focus on pictures did not give the A-E slides an advantage.

Both of these conclusions argue AGAINST the A-E model when the goal is to remember statistics. Instead, they recommend a focus on slide text over images.

I applaud Professor Michael Alley for continuing to test and refine the A-E model as a promising direction for educators and other presenters. I encourage him to continue studying the A-E model to discover the elements that improve learning and the conditions where the A-E model is superior, as well as inferior, to other models.

Unfortunately, the only conclusion we can make from this study is that students remember statistics better when they are included as slide text than when they are spoken by the presenter or presented as a picture. The rest of the conclusions must be discarded because of the research design flaws.

About the author: Bruce Gabrielle is author of Speaking PowerPoint: the New Language of Business, showing a 12-step method for creating clearer and more persuasive PowerPoint slides for boardroom presentations. Subscribe to this blog or join my LinkedIn group to get new posts sent to your inbox.

Want to wow your boss with your next presentation? In Oct 2012 I developed the SlideShare deck 5 Tips for Presenting to Executives where I compared executives to Simon Cowell – difficult judges who were always on the hunt for talented presenters.

I was therefore flattered when Dirk Hannemann, a communications trainer in Berlin, Germany, took that metaphor and developed this inspirational video. I love the use of contrast, showing the skeptical judges at the start and then their shocked and amazed faces at the end.

That’s your goal. Not merely to inform – but to wow. Now, go nail your next exec presentation!

About the author: Bruce Gabrielle is author of Speaking PowerPoint: the New Language of Business, showing a 12-step method for creating clearer and more persuasive PowerPoint slides for boardroom presentations. Subscribe to this blog or join my LinkedIn group to get new posts sent to your inbox.

Who will win the 2012 presidential election? I was a bit confused by all the different polls being reported – some saying Romney was gaining momentum, others arguing Obama had the edge. So I decided to create a simple set of graphs that clearly tells the story of the 2012 US presidential election.

1. Romney is favored by 51% of Americans according to Gallup. They not only ask people who they would vote for, but also how likely they are to vote on election day. Those who are not likely to vote are not counted.

2. But Obama has more electoral votes. US presidents are not elected by “national votes” but by “electoral votes“. That is, Americans elect local state representatives. The state grants all its electoral votes to whichever party has more state representatives. Many states have a long history of voting Democrat or Republican, based on their own culture and values, and recent polls show that trend is likely to continue in 2012.

270 electoral votes gives a candidate the majority. And based on state polls (usually “likely voters”) and historical trends, Obama is closer to that goal than Romney.

3. The swing states will decide the election. There are 8 states where the polls are roughly tied, or the candidate has a 1% to 3% lead which changes every few days. We call these “swing states” because they could swing either way.

Florida is the largest swing state, with 29 electoral votes. Romney leads in Florida by 2% and has lead for several weeks, so there’s a good chance Florida will swing to Romney. Ohio is the second largest swing state, with 18 electoral votes, and that’s why Obama and Romney are campaigning so hard in this state.

Romney still needs 79 electoral votes. Even with the 57 votes he’ll get from the 3 states leaning his way, he still needs another 22. Romney needs Ohio. Without Ohio, it will be tough to get the other 22 votes he needs. There’s a good chance Ohio’s vote will determine your next president.

Ohio is currently leaning toward Obama by about 2%. We’ll see if that lead holds until November 6.

If you enjoyed these graphs, you’ll love my new book “Storytelling with Graphs”. Subscribe to this blog to learn when it’s available.

About the author: Bruce Gabrielle is author of Speaking PowerPoint: the New Language of Business, showing a 12-step method for creating clearer and more persuasive PowerPoint slides for boardroom presentations. Subscribe to this blog or join my LinkedIn group to get new posts sent to your inbox.

I spend a lot of time these days studying infographics. Part of the research for my new book “Storytelling with Graphs”.

But I keep running across atrocities like this – nice design, terrible understanding of data. Do you see the problem? Study the picture, then scroll down for my analysis.

My comments

1. First, this donut charts adds up to 138%! What does this even mean?

2. The blue Twitter.com donut slice barely makes it halfway around the donut – it’s more like 60%, not 78%. Unfortunately, some reader is going to grab that statistic and share it with others. This mistruth will be picked up as “fact” by people who don’t double-check the source, and repeated on their blogs, Twitter accounts, PowerPoint decks.

3. There are several donut slices that could be color-coded differently to form mini-groups. Especially, the mobile apps could be different shades of green, grouping them into a mobile access sub-group. Then the message gets clearer: Twitter.com is the most common way people send tweets, mobile phones is the second most common.

Designers are ahead of the rest of us in exploring the possibilities in data visualization for a mass audience. But I hope they will continue to educate themselves on data visualization techniques that bring the truth out. Design can be a powerful tool to reveal the truth, or to conceal the truth.

About the author: Bruce Gabrielle is author of Speaking PowerPoint: the New Language of Business, showing a 12-step method for creating clearer and more persuasive PowerPoint slides for boardroom presentations. Subscribe to this blog or join my LinkedIn group to get new posts sent to your inbox.