I want to revisit the five points made in this video in future posts, but, for now, I’ll just share with you this excellent video by Nacy Duarte and the staff of Duarte Design.
I’ve written about the creative possibilities in PowerPoint before, and I want you to take note that this incredible presentation was made entirely in PowerPoint 2010. Again, the tools do not make the presentation. You do.
I’ve been looking at the flurry of information (and sometimes lack thereof) regarding iPad, and I’ll probably post some thoughts on Simply Robert before the end of the week. One development that interested and surprised me, though, was the introduction of iWork for iPad.
image © Apple, Inc.
The preacher at the congregation I attend occasionally uses a tablet PC for his PowerPoint presentations, and it seems a nice solution. He can have the machine resting on the podium without the bulk of a laptop. Unfortunately, more than once, he has been left befuddled by some technical complication with the device or the tablet responding incorrectly to his inputs. Say what you will about Apple, but they have a reputation for releasing products that Just Work for a reason, and I could see taking this device into presentation settings.
Apple obviously sees iPad as a potential presentation machine. Otherwise, why bring something like Keynote to the platform? Keep in mind that neither iPhone or the iPod touch have a native version of Keynote (though both can open .key files in email attachments), but iPad gets its own dedicated version of the three iWork applications – versions completely rewritten to take advantage of a touch interface. Of course, this could easily be used in one-on-one settings, looking much better than demoing something on a phone’s screen. A few issues may need to be addressed, however, before an iPad could replace taking my favorite laptop to my next speaking engagement.
Here’s my wishlist:
- A bluetooth remote. iPad will work with Bluetooth keyboards. Why not monetize a first party remote?
- Wireless projector support. You can run to a VGA or component connection through the Dock Connector. Could one not also send a wireless video signal?
- A USB port. Since iPad won’t navigate files using the usual hierarchical metaphors, perhaps this would not be for file storage so much as peripheral support.
- Kiosk mode. If you used one or several iPads to pass around a conference table, it would be nice if attendees were unable to accidentally exit the presentation.
Now a kiosk mode may exist in Keynote for iPad. I haven’t been able to try one out, so I cannot know what the application is able or unable to do. Furthermore, two wishes – remote support and wireless projector support – can possibly be remedied through hardware vendors taking advantage of the App Store. (See the innovative Square credit card payment system for iPhone as an example.) Solutions may become available over time.
Before iPad finally hits the market, debates over closed or open ecosystems will continue. Many will predict its success or failure without ever touching one. I have to admit to being intrigued, and I’m curious if such a device can make life easier for presenters in the long run. What are your thoughts? Could you see yourself using an iPad in a presentations setting? What would be on your wishlist to make it a better presentation device? Share your thoughts in the comments.
Please note: if your comment is awaiting approval, be patient. I cannot approve comments until after I get home from work. Thanks for your understanding.
Over the last few months, I’ve written about presenting data honestly on slides as well as fitting your narrative to the data you have rather than vice versa. Integrity is the point here. Every time we erode our own integrity as presenters, we create an impression of dishonesty (or at least incompetence), and we damage our own potential to positively connect with the members of our audiences. We hurt ourselves when our audience discovers that we – intentionally or otherwise – have been presenting factually inaccurate material.
For example, a presenter I saw a few years ago (of whom I’ve written before) was speaking on the advances of technology, and he was using the iPod as an example. Not only did he get the year the iPod was introduced incorrect, but he also had prices and capacities mismatched. This was a professional presenter who was hired as a consultant, but I had a hard time taking his other facts and statistics seriously – some of which I later discovered were also incorrect – after seeing mistakes that could have been remedied with a simple Wikipedia search.
This issue hit a little closer to home when I was watching some colleagues give a talk about institutional racism earlier this year, and they were using print advertisements to prove their point. One ad they included in their talk was this one:
Yikes.
There is one problem: the image is a fake – and not the only one in the presentation. Proctor & Gamble never marketed the now-vanished Dash detergent this way. Yes, the image illustrates racism, but it does not illustrate any form of institutional racism. Rather, it illustrates racism as used in parody – a completely different topic and debate.
Close colleagues may be pretty understanding of comparable slip-ups, but imagine you are pitching an idea, presenting to a larger audience, or representing your organization at a trade show or national conference. Such mistakes are not only embarrassing but they undermine the very message or product you are trying to pitch. Fact-checking has become easier thanks to the Internet. We shouldn’t let our presentation quality suffer because we get fooled by a fake image or we don’t know how much an iPod costs.
In my last post, I critiqued an informational graphic shown by a Fox affiliate in Chicago that displayed a pie graph adding up to 193%. During the post, I opined that the chart was poorly implemented for a few reasons, including it being entirely the wrong kind of graph for the data they were presenting. Also, toward the end of the post, I proposed a different method of presenting the data that was more true to the data gathered. That graph, however, may not have been a good solution.
In watching the video again (which you can watch a clip of here if you wish), it’s clear that the story is framed to discuss potential future Republican presidential candidates during the primaries, which would:
- exclude Oprah and, apparently, Gingrich.
- exclude Democrats who would not be voting in the primaries.
- exclude Independents who would potentially not vote in Republican primaries.
Still, my original criticism that the language citing these percentages as “backing” a particular candidate still stands. A more appropriate visual might have looked like this.
The problem returns to one of trying to make the data fit the story rather than the other way around. In a post I made a few months ago (and that I’ll be revisiting again soon) about keeping our slides honest, I wrote of the dangers of being too selective with the data we present. In our information culture, we face audiences that distrust statistics because they see the numbers skewed so often. If we want to overcome that cynicism, we have to be honest with the data we present. Our story should fit the data we have. We should not be shoehorning the data into a predetermined narrative as was done in this example.
By now, you’ve probably seen this graphic from a Chicago-based Fox affiliate:
Hmm, something’s not adding up here.
There are a couple of problem’s right away, some funnier than others:
- Pie graphs are, by nature, supposed to add up to 100%. This one tallies up to 193%.
- The graph does not match the intent of the actual poll.
- Sticking back in front of every name makes a couple look like medical conditions. Just saying.
Picking the Right Graph
I have to admit that I sometimes pick the wrong kind of graph for the data I’m presenting. I usually find a fifth grade math text handy in reminding me what to do. (I admit that, no, I am not smarter than a fifth grader.)
- Bar Graphs are suitable for straight data comparisons.
- Line Graphs demonstrate data changes over time.
- Circle Graphs help show data that are parts of a whole.
In the case of this particular data set, a bar graph would have probably been most appropriate, but that only cures the most superficial flaw. The true problem with this graph comes with the way the data is presented in contrast with how it was collected.
Present Data As Collected
I think my recent manager would call this “Presenting with Fidelity.” Simply put, present the data in a way that most accurately reflects what was being collected. In the case of this graphic, the data and the graph are saying two different things. The graph makes the improbable claim that these percentages represent Republicans who back a given candidate. That, however, is not how the data was collected. The telephone poll in question surveyed 900 registered voters, asked them to rate certain names as favorable, unfavorable, no opinion, or never heard of. Then the data was sorted by political affiliation.
There’s a big difference between claiming to have a favorable opinion of someone and backing them for a presidential nomination. That also explains the seemingly strange math involved here. How could a majority of respondents claim they would back each of these hypothetical candidates? Such a poll should reflect data that adds up to something close to 100% (assuming a few respondents chose to back none). It’s entirely possible, though, for a majority of respondents to favor each individual, especially when the statistics are confined to one political party.
Before we finally fix this graph, though, a couple more problems present themselves.
- Only three of the five names are included. Oprah Winfrey and Newt Gingrich were also part of the survey.
- The data is presented in isolation. Where are the comparisons with the Democratic respondents, Independent respondents, and the big picture?
It’s especially important to include the other demographics because Republicans only accounted for 36% of all respondents. Therefore, this graph represents roughly 324 of the original 900 surveyed.
This is what the survey really represents. This is, admittedly, very full for one graph. A better solution might be to break the data out into four separate graphs, sorted by political affiliation or by individual. However you would choose to represent the data, do so with fidelity. Make your graph or chart say the same thing the data does, and try to make sure you’re using a visual representation that is well suited to the data at hand.
