The Safety of Work

Ep.18 Do Powerpoint slides count as a safety hazard?

Episode Summary

On this episode of Safety of Work, we discuss whether PowerPoint Slides count as a safety hazard. This topic came up when we thought about what often goes wrong with safety communication.

Episode Notes

We use the paper When Redundant On-Screen Text in Multimedia Technical Instruction can Interfere with Learning to frame our discussion.



“I think people genuinely think it’s a good way to convey information.”

“The cognitive load theory is suggesting, in this case, that the worst thing to do is to give them text...and audio at the same time.”

“It definitely doesn’t apply that diagrams plus audio is bad.”


Kalyuga, S., Chandler, P., & Sweller, J. (2004). When redundant on-screen text in multimedia technical instruction can interfere with learning. Human factors, 46(3), 567-581.

Episode Transcription

David: You’re listening to the Safety of Work podcast Episode 18. Today we’re asking the question, do PowerPoint slides count as a safety hazard? Let’s get started.

Hey, everybody. My name’s David Provan. I’m here with Drew Rae, and we’re from the Safety Science Innovation Lab at Griffith University. Welcome to the Safety of Work podcast. If this is your first time listening, then thanks for coming. The podcast is produced every week and the show notes can be found at In each episode, we ask an important question in relation to the safety of work or the work of safety, and we examine the evidence surrounding it.

Drew, what’s today’s question?

Drew: Dave, today’s question first arose for me when I was doing some research into what goes wrong with risk assessments. We floated the paper around the lab, and one of the elderly professors who’d been around the industry for ages suddenly said PowerPoint. I was confused. What’s the relationship between risk assessment and PowerPoint? He said that very often, people do a lot of technical work, but then they present the results, not in the final report in sophisticated form. They just put out a set of PowerPoint slides with the results.

So, that’s the question for our episode. What’s the link between PowerPoint and safety? What specifically do PowerPoint slides count as a hazard?

I guess we just need to be clear as a disclaimer upfront. The PowerPoint is itself a particular branded product from Microsoft, and we’re not having a go at Microsoft in this episode. This problem dates back even before Microsoft existed. I’m going to try to use the name and I hope I don’t forget to use the term “viewgraph slides,” just to be clear. It’s not actually the particular format. It’s just this idea of putting information onto our computer screens or projectors as electronic slides.

David, you’ve got a few examples of where they’ve actually been directly mentioned in accident reports?

David: Yeah, we should have mentioned this in an earlier episode. Haddon Cave mentioned this in his report into the Nimrod accident, if you’ve heard him talk or heard any of his talks since he produced that report. There’s a quote I pulled out where he said, “The ubiquitous use of PowerPoint should be discouraged. It can lead the audience to watch rather than to think.” That’s a little bit different (I think), Drew, to how we sometimes think about PowerPoint slides as dumbing down information. He was talking about PowerPoint slides as not actually conveying understanding to an audience; just providing with a pretty little slideshow.

Drew: Yeah. PowerPoints get mentioned for both of those reasons and it’s going to be the audience reception of PowerPoints that we’re going to particularly focus on in this episode. 

Just to give a little list before we go into that. They get mentioned for that other purpose in the Columbia accident investigation. Talk about doing community presentations of risk information and controls presenting information on PowerPoint. They get mentioned in as you said Dave, in the Haddon Cave report. They get a little bit of a mention in Deepwater Horizon in a couple of the investigation reports as well.

It’s no surprise given that viewgraph slides and PowerPoint are ubiquitous. They’re going to be used by any organization that has an accident. Chances are some of the vital information was presented on slides at some point.

David: Yeah we use them. They’re absolutely everywhere. We work with a lot of organizations, and we put everything on PowerPoint slides, from risk assessment results, to strategy information, to induction material, to almost all safety practices now in some way, shape, or form. Organizations are likely to be either presented on a PowerPoint slide or the outcomes of the practice to be presented on PowerPoint in a meeting somewhere.

Drew: That’s particularly where I’d like to zoom to in this episode, is this deliberate use of PowerPoint to convey safety information. I guess the key example here that’s going to be relevant for most people is when you’re presenting the results of an investigation, or you’re doing some safety training, giving an induction or training about a particular hazard, or a particular topic, or a toolbox talk. Does it add to the communication to be using PowerPoint to do that?

One question I’ve got for you, David, is just everyone talks about PowerPoints as being boring. We’ve got the whole phrase “death by PowerPoint.” Why is it that PowerPoint is used so much? Is it because people think it actually spices up their talk and makes it better? Or do you think the idea is it’s almost like written proof of what we talked about? So people aren’t even trying to do good presentations, it’s more about being covered in what they’ve said.

David: Yeah, Drew. I don’t think it would be for that purpose of assurance or coverage. People genuinely think it’s a good way to convey information. We, even myself (until I read through some of these), I thought we’ve got these AV facilities, computers, and screens, so nowadays in organizations when we have a meeting, or we need to do a presentation, we all believe that gathering our thoughts, putting them on slides, and then talking and showing information on slides at the same time is how you present. I don’t think there’s anything untoward about it other than modern-day organizations think that it’s a good way to convey information and present information.

Drew: So let’s make that then the real question for this episode because that’s what the paper is going to directly speak to. Do we actually think based on the evidence? If you’re going to give a talk anyway, does writing down some of the key points of what you say on a PowerPoint slide behind you help or hinder with people learning from that talk?

David: That’s a great question, and we’ve got a really interesting paper to talk about today. Do you want to introduce the paper?

Drew: Okay, I’ll give a little bit of a stab at it. The title is called, When Redundant On-Screen Text in Multimedia Technical Instruction Can Interfere With Learning. It’s a long way of saying, “Does stuff on PowerPoint behind you help or not?” The authors of the paper are Slava Kalyuga, Paul Chandler, and John Sweller. It’s published in the journal Human Factors. It’s fairly old, 2004.

I’ve done that deliberately because this is pretty much the origin of the central idea when people say that having things written on slides at the same time as speaking them is bad. This is pretty much the paper that they cite. It has been replicated since and worked done between then and now does stay fundamentally the same sorts of things that get said in these papers. I don’t think that you’re dealing with outdated and incorrect information. I just want to give credit to the people who first started seriously exploring this.

John Sweller, in particular—he’s retired now—the first author of this paper is now a distinguished academic with hundreds of papers. Sweller has a massive body of work centered around looking at what creates the best conditions for learning. That body of work is almost all experiments, so creating two conditions and saying under which condition do people learn more.

David: Drew, what else I like about is we talk a bit about safety being a transdisciplinary science. These are professors and academics specifically focused on education and learning, and psychology. They wouldn’t (by any means) consider themselves safety researchers. It’s just a great example of where, in safety, when we go to parent disciplines, or we go to specific experts in specific fields we can get really good insights about things that we should and shouldn’t do in safety. 

Do you want to talk about how much research goes on in the disciplines outside of safety that we should be looking at to inform what we do in safety?

Drew: There’s a lot to be said there. The general principle is that for any activity that we do in safety, you can look at what someone else is doing and try to copy it but guaranteed that somewhere there’s a science for that. You want to put up a poster, there’s a science about good poster design. You want to have a safety conversation, there’s a science about how to have effective conversations. You want to change someone’s mind, belief, or risk perception, there are separate sciences for all three of those. Absolutely, you want to do something which counts as education.

Oddly enough, there’s a lot of people at universities that think an awful lot and experiment an awful lot on how to make adults learn.

David: That’s good insight, Drew, and specifically about this topic, PowerPoint, you’ve got your own opinions before we got to today’s episode. I think you’ve been waiting to do this for a while, and we’ve held off until episode 183. Do you want to just disclose your own thoughts on PowerPoint?

Drew: I’ve been a public speaker and debater all of my life, so I’ve got pretty (let’s say) strongly ingrained thoughts based on personal experience about what does and doesn’t work when it comes to speaking. I’ve never ever been a fan of particularly putting text on PowerPoint slides. What I tell all of my students and colleagues when they insist on creating shared PowerPoint packs, if the text matches what you’re saying, then it’s redundant. If the text doesn’t match exactly what you’re saying that particular time, then it’s distracting.

As much as that’s my own personal opinion, it’s one that is shared by lots and lots of good public speakers. In fact, they put it in as one of the core TED Talk principles. In a TED Talk, the presenter is the center of the talk, and anything that’s a visual has to be secondary and supporting. You hardly see a TED Talk (certainly never any of the core official ones) that has words behind the speaker. There’ll be pictures to support what the speaker’s saying but not words.

David: Drew, for our listeners, we had a little bit of an error where I didn’t know if Drew had prepared today’s episode, so when I was reading the article, I thought I was helping him out by preparing it. It was that three things that we said exactly the same. One was about the role of paper pointing accidents, one was a reflection about why you never see PowerPoint slides in TED Talks, and then the third was something that we’ll talk about when we get to the practical conclusions.

But that was interesting. I had that reflection when I went through the paper and I went, “Oh. Actually, you don’t see PowerPoints as we see them in organizations when you’re watching a really good speaker in a TED Talk.”

Drew: Now, there’s a risk that at least one of our students may be listening to the episode, so I am going to point out just for safety, you might want to create an audit trail in teaching their purposes other than effectiveness for putting things on slides. One of the big reasons we do it is to make sure that people who aren’t physically in the classroom, or who have different learning needs, can get the same information. Also, to make sure that if we’ve got a course that is split between two classes, that each class learns the same things.

There are reasons other than effectiveness for putting words on, but generally, if you ever see any of my own slides that they’ve got words on, it’s because I’ve been lazy. It’s because I’ve prepared the slide with words, so they knew what I was saying, and then I haven’t had time or maybe effort to go back and replace all of those words with nice visuals instead.

David: The introduction to this paper goes deep into background with [...] psychology and talks about attention and working memory, and long-term memory formation. Talks about Cognitive Load Theory, dual processing models of memory, all the way through to fairly specific and almost biological descriptions of how the brain receives auditory, visual and other sensory information, makes sense of that, comprehends, and learns. Do you want to talk a little bit about some of the theories that are relevant directly to what the experiments that we’re about to talk about where we’re trying to test?

Drew: This particular set of authors and a whole group of people who follow them, have an approach to learning called Cognitive Load Theory. This is being criticized because it’s not an accurate model of how the brain works, but it’s one of those things that’s not accurate but may in fact be true enough for its purposes.

The idea is when you’re learning, you want to be working really hard. You want your brain to be active so that it is processing and interpreting information, it’s not just being drip-fed. We know that rote learning doesn’t work. On the other hand, do you want your brain to be working on stuff that is really relevant to what you’re learning? You don’t want it to have to do other tasks instead.

For example, if someone has a lot of difficulty reading, then they’re going to be working really hard just to spell out the words. They’re not going to be working really hard to process the underlying concept. If someone has visual difficulties, they’re going to be trying hard to see stuff instead of trying hard to learn. There are all sorts of things that teachers can accidentally do to create this unnecessary load that uses up part of your memory and distracts from the important part of learning.

One of the things that they test a lot is for example if you need to look in two different places, get that information, and put it together so that you’re looking back and forth, that’s bad. That’s creating extra unnecessary work. If you’ve ever noticed when you’ve been reading and there’s a footnote, it’s just harder to read something with footnotes because you got to remember what is there, look down, read the footnote, go back to your place previously. Learning is harder and slower, and unnecessarily.

That’s the thing with Cognitive Load Theory is working around and gives us the hypotheses that are going to be tested in this particular paper. The hypothesis says that if we’ve got a person speaking, and we’ve got text, then there are two different things to process. Those two different things mean that the person has to hold more in their brain, they might in fact learn less. The contrasting possibility is that having information presented in two different ways makes it easier.

The main theory that actually promotes this we now don’t believe, which is in thing called learning styles. Learning styles is the idea that different people learn in different ways, so you make visual for some people, audio for other people, diagrams for other people. We know that one isn’t true. Open hypothesis is about whether presenting it in two forms to the same person adds to the learning or takes away from the learning.

David: So Drew, to get there, the researchers did three experiments. Essentially, they used the same group of 25 technical apprentices they have access to. It appears from the acknowledgment in the paper that they had access to a group of technical apprentices at BHP from what we can work out in the early 2000s. They ran three separate experiments to try to refine and test these hypotheses about presenting concurrent visual and audio information versus separate information. How about we just talk through these three experiments in a way that helps people understand what the participants were going through, then we can move on and talk about what was found.

In experiment one they were basically comparing these effects of simultaneously presenting the same information. What they’re actually doing was they had a diagram of how to cut metal or different types of materials basically. It was a three-way, three-axis diagram of cutting speed, revolutions per minute, and material thickness. Then there was a spoken explanation over the top instead of bullet-point process steps on the slide beside the diagram. People had to look at the diagram and read the process for how to use the diagram and listen to a person speak about the process of how to use the diagram. They did that all at the same time for one group and then the second group they just narrated the diagram and then showed the text separately, and then for the third group, they showed the text and narrated as well separately.

Basically, this format’s the same for the other experiments, so they only had to go through it once. Then they asked the subjects on a scale of 1–7 how difficult they felt the diagram was to use, and this is a bit of a subjective assessment of mental load, so how difficult people found the task. It was going to be interesting for the researchers if people found the task a different degree of difficulty by the different way that they presented the information. Then they gave the participants 10 multiple choice questions to test their understanding of the diagram.

Drew, do you want to talk a little bit more about the experiments and the independent and dependent variables?

Drew: We’ve talked about experiments and how they work in previous episodes. Basically we’ve got an independent variable, which is the thing that we’re changing and the dependent variable loots the thing we’re measuring. In this case, the independent variable is what format is it in. We get to test out the differences between audio plus text at the same time, or text followed by audio, looking at them at different times. There’s actually three conditions for this one out there.

David: Yeah, they eliminated one of the modes, which means they no longer showed the text at all.

Drew: That one was just audio. We’re seeing two things. One of them is subjectively how hard they thought it was, and objectively how well did they perform on the multiple-choice.

They have a couple of slightly weird ways of comparing these two output variables. They use a thing called instructional efficiency, which is based around this Cognitive Load Theory, which says that you’d actually combine together the subjective feeling of how hard it is and your objective score to work out how efficient the learning is.

David: I’m glad you explained that Drew because I was trying to work out the mess and it was a little bit complicated, but I think what they’re trying to get out in education is to give people the most understanding with the least cognitive effort for their measure of efficiency.

In experiment one they found that it took longer, instruction time took longer. If you present all of the information, diagrams, written information, words at the same time, it obviously doesn’t take as long. On average, it was taking maybe 20% longer to actually separate out the written text and the audio text.

On the writing scale, that changed quite significantly how people found or reported back how easy the module was to understand. On their scale of 1–7, with one being easy and seven being difficult, it was 2.7 versus 1.6. That’s both pretty easy down the end of the scale, but the thing that I found really interesting, Drew, was the separation of people where they’ve separated out the text and the words and not provided this concurrent information by the two medians at the same time. People scored 6.8 out of 10 versus only 5.8 out 10 when they were given the concurrent information. It’s a small sample size, but that (to me) felt like quite a difference in scores.

Drew: The difference between 60% and 70% on a test can be really quite significant. For each individual, it’s only basically getting one extra question right but when you consistently get that across a group, that’s a real sign that there has been a difference in learning. Particularly given such a small task that they’ve been asked to do, that’s a big difference.

We’ve got some hints already from this experiment that the Cognitive Load Theory is suggesting in this case that the worst thing to do is to give them text, information, and audio at the same time, that putting up together actually makes learning harder rather than easier.

David: What I also like about this paper, Drew, is sometimes you’ve done that experiment and sometimes we see it in safety a little bit. You’ve got a result or a finding and great, let’s go and publish that. But I do like the way that in this study they went on and did experiment two and three. They obviously thought, “Well hang on a minute.” The groups had different time to look at the concepts because you know that I mentioned that they had a lot when we separate out the information. There was a longer period of time for people to get one piece of information than the other, and they thought, well there might have been some reinforcing effect or a little bit more time that people had, so they thought, “Well actually no, we need to repeat this experiment two and control for training time.”

What they went and did is they actually created these computer animations this time instead of using PowerPoints, but they’re basically animated PowerPoints. This is 2004 so it’s probably not that different to what we do now with our normal PowerPoint program. Training time was exactly the same. They expanded out the scale of how difficult people found to nine points instead of seven because they wanted a bit more differentiation at that lower end, and they gave people 10 multiple-choice questions.

Instead of cutting disks and spades, they’ve changed it to soldering. These were mechanical engineering or technical trades students. Interesting the way that they did this to get the time the same, Drew, is they had basically this diagram on the slide, and the audio explaining the diagram. Then they went to the words, which was like going through the words, separating it out from the audio.

In the other one, they had the diagram, the words, and the audio at the same time, but then they left the information on the page for the same amount of time that the other condition was having it sequentially presented. People got the concept in front of them for the same amount of time. The only difference was in one group they’re all getting the information all together and then letting it sit there, and the other one they were getting it sequentially.

The findings were the same. People found that cognitively, they assessed the task as being more difficult, they scored 6.9 out of 10 versus 5.6 out of 10, so that’s even a greater difference when time was controlled for, and this efficiency gap was also showing as quite wide.

So Drew, thoughts on tightening up these experiments and trying to replicate these findings?

Drew: What I really love is that you’re pleased that they replicated it. They still weren’t satisfied having gotten exactly the result they were expecting and wanting twice in a row. They didn’t say, “Hold on. Maybe there’s another explanation. Maybe the effect isn’t real.” The new idea was maybe giving it the same amount of time isn’t what matters.

Maybe what we’ve done is we’ve given people two bites of the cherry. We’ve given them the information in one format and then giving them in a different format gives them basically a chance to revise, and we know that revision helps people learn. Maybe we shouldn’t be blaming having them both at once yet, we need something that definitely rules out, locks in on this fact that it’s that both at once is a bad thing, rather than some other secret advantage they’re getting. So, experiment number three.

David: Now they took out the diagrams and the difficult comprehension, and they said, “What’s the effect going to be when we just try to give people basic information about mechanical engineering?” They isolated down about 300 or 400 words of the information about general mechanical engineering type principles, and all they wanted to test was this audio and visual at the same time or just audio. Basically, everything else was identical.

One group just sat there with nothing in front of them and just the audio and listened, and they split up the section. Basically they split up the text into four sections, and after each of the sections people did 8 multiple-choice questions, so there were a total of 32 questions.

One group just got the audio, and the other group got the audio in their ears, and they got the text on the screen in front of them. The only difference there was not about time. It was one group seeing text while they’re listening, and one group was just listening. This was probably the most surprising because obviously this didn’t involve a lot of thinking, so both groups write it quite easy but there’s still a difference. The group that was having to listen and read the same information at the same time, found the task cognitively harder.

The results, Drew, the group that was just listening got 25.2 out of 32 and the group that had to read and listen at the same time got 21.6 out of 32. This felt like even a greater spread of actual questions. That’s a big gap for people who’re just sitting there listening to basic principles versus someone having to listen and read.

Drew: Yeah, so the more and more they got closer to exactly mimicking the difference between someone just standing and speaking, and someone standing and speaking with text behind them, the more real and certain this effect became. The only thing that’s different here is that these are being done on computer screens in front of the participants. They’re not being done with a live speaker. There’s a very good reason for that.

To get a totally fair comparison, we need to have the speech happen exactly the same every time, and we just can’t do that with a live person. It’s not exactly the same thing as a speaker with slides. Correct me if you disagree, David, but I think this is a very fair way of testing the idea of a speaker versus speaker with slides.

David: Yeah. What we spoke about a few times, Drew, that is if there’s actually a real effect there or if there is an underlying phenomenon that the researchers are finding. As we do more of the studies, as we get more controlled with the studies, and the design of those studies, then we should see that effect continue, and we’ve even spoken about we should see that effect materialize more and more.

That’s definitely [...]. It’s nice to see a paper that’s actually gone and run three separate experiments, refined the method for that experiment each time based on the previous findings, and not wanting to just claim that something that had been found before they really honed in and controlled for other possible explanations for what was going on.

Drew: It’s great that we have come across something where we’ve got information which is consistent in these experiments with what a lot of people have found in practice. Sometimes, we find ourselves criticizing practices that have been around for a long time on the basis that when we study them, the efficacy of those practices disappear.

In fact, what we’re seeing here is that things like the principles that they put into TED Talks or into teacher training, in fact, are based on really quite solid science. This idea of just having a single presenter presenting a clear message instead of cluttering up people’s brains with different sources of information is in fact good advice. It’s advice that is backed up not just by this study but by other similar studies that are followed on afterward.

David: Yeah, and if we think about how often in safety we do things that are a form of education would benefit from this type of knowledge. I mentioned earlier things like induction training. How many induction training programs do we go into now, where we sit in a room, someone puts a set of PowerPoints up, someone basically reads off those PowerPoints at the same time and flicks through and some people sleep and some people roll their eyes, versus the situation where a site manager or someone just stands up with nothing behind them, gets all of the attention in the room focused on them, and delivers the information that they want delivered for people who are coming on to that site.

So, there's some common practices in safety which we basically have PowerPoints at the center, where maybe we need to think about putting people back at the center of.

Drew: I guess I should briefly throw in here that this is a slice of quite a wide body of research. One of the other things that they've studied is the use of diagrams and where it definitely doesn't apply that diagrams plus audio is bad. 

The idea is if you are going to have any text, the text should actually be part of a diagram rather than separate from the diagram, even if you are not separate in the terms of explained by a voice. Information should already be as integrated as possible. If that means a single person speaking or if that means a single diagram with a small amount of text in the diagram, that's great. This isn't like, "Don't use PowerPoint." This particularly applies to bullet points on PowerPoint. 

David: Yeah, there is something in the conclusion of the paper like you said, Drew, only putting information in the same place if people actually have to integrate that information. If you need to do that, that's actually integrating information in the way that it is presented, is really actually important. Otherwise, just don't present information in the same place at the same time. Even though when we present graphs, then explanations of graphs in our safety performance reporting, and they talk about it all at the same time, we are basically doing exactly what the experiment says don't do.

Drew: Is this a fair time to move on to some practical takeaways, David?

David: Yeah, we might already be bordering across there, so let's dive full into practical takeaways. Drew, how about you start?

Drew: As David mentioned earlier, we ended up preparing twice and separately for this, so we have some different practical takeaways, some of it have ended up being exactly the same. The first big one for me is when we give presentations to workers or to other staff, it's worth stopping and thinking what is the central thing we want them to learn, and why is the rest of the stuff there?

That last one is a rhetorical question because all education theory says that giving people any extra work actively hurts learning. Teachers have to constantly remind themselves of this. I have to do it all the time to myself. There's always something else you want to say. One extra bullet point, one extra fact, one extra example, one extra reminder. You just need to have a big poster on your wall saying, "Adding that extra bullet point can actually take away information." It deprives people of learning to say more. If you can just indoctrinate yourself with that, it really changes your whole approach teaching in presentations.

If you say less, people will learn more. One that I constantly run into is that I teach high school debating and high school debating is very closely timed. There's a bell to tell you when you've got one minute left and there are two bells to tell you when the time is out. Young debaters have to be constantly told that, "When that second bell goes, shut up and sit down," because you are now only taking away from any score that you get. Nothing you say beyond this point is helping. You know you feel that you got to say it, but there's a time to just make yourself shut up.

David: I think that there was a takeaway in the first experiment about the pacing of learning, and I don't think we explained that very well when we were going through the method for the first experiment. In the first experiment, the people looked at the diagram and when they clicked on the text, it basically showed the text and spoke the text at the same time. Then basically, when they are ready for the next step, they would click on the next step, and they recorded all the times people talk.

They basically concluded, even though they went on to refine the experiment because they didn't want this pacing of learning to be a confounder for what they are really trying to understand about this concurrency of presenting information, they said that the pacing of learning makes a really big difference and a big impact, and it's quite different for different people. I suppose that's the reason that we talk quite a bit about self-paced learning.

In safety, with a lot of our learning process, there's not a lot of things that we do that are self-paced, but actually giving people the information and letting the learners stay with it until they are ready to move on is actually a really important thing to do.

Drew: There are some quite well-established practices that embody this. One of them is something that has fallen by the wayside as people have got more and more PowerPoint, which is giving people handouts. The idea of a handout isn't something to have in front of them to distract them while you are speaking. It's a simple summary of that stuff that you might have put up on PowerPoint slides that you give people at the end as something to take away, look back, and reflect on.

It supplements any notes they've taken themselves. It means they can pay attention to you speaking rather than trying to take notes. It means that they get that benefit of the two different formats. The spoken word and the written word, but not in a way it is the same time making it harder to learn.

David: Yeah, there's a general conclusion here, Drew, that if you are putting information on a PowerPoint, which is just repeat of the things you are saying at the exact same time, then not only is it a waste of time or effort, but it's actually going to be watering down your message. This study actually says don't do that. Don't just put words that are just points, summaries, or things for the sake of having something behind you when you are speaking because you are actually watering down the retention or what people are going to take away from what you are saying to them.

If you want to read more about this, then you could go into a psychology journal and read up about split attentions and redundancy effects. You actually want people to have redundant cognitive load, so they can be thinking and internalizing, or comprehending what you are saying rather than splitting that attention across different mediums.

Drew: And if you just want to trust two people who derive parts of their income from giving public talks, then just trust us. Don't put words on your PowerPoint slides, it doesn't help. It will be a better presentation without it. If you have to put words, do what I do and put bullet points and then go back and delete them all. 

If you need to put them on notes to remind yourself, put them on notes to remind yourself and just replace every bullet point with a Google search with an image that matches the keyword on the slide.

David: Yeah, maybe both of us talk a bit, Drew, but any of us derive some income from it. I don't think I would have been paid to talk, but there's always a teacher. What would we like to know from our listeners, Drew?

Drew: Love for you to tell us your own pet hates about presentations. What do you find difficult when you have to sit through inductions or training? What do you wish people would stop doing so that you could focus on your own learning?

On the other side, when you are giving presentations, when you are giving inductions to [...] talks, what do you use as your measures as to how well they are working?  Do you use feedback? Do you have some formal assessment? Do you get people to rate the talks? Do you give them a test? I'm interested in what methods people use.

Mostly I'm interested because I've seen some pretty bad ones. I've seen lots of organizations that put in a test at the end of the inductions, and I've seen that actually gets applied by the person giving the induction sitting there telling them the answers. I'm interested if there's some good practices out there. 

David: So Drew, this week we asked the question, "Do PowerPoint slides count as a safety hazard?" and I guess we have to say, well yes they can. If managing safety risk in your workplace involves conveying important information to people, and you put that information on a PowerPoint slide and speak it to people at the same time, then they are probably getting less of that information than they otherwise could, if you just said it to them or if you just gave it to them to read. Is that a fair answer to that question?

Drew: Yes. I think it's fair to say that if your control involves people holding information, then the text on that PowerPoint slide is directly weakening your control. 

David: There we go. That's it for this week. We hope you found this episode thought-provoking and ultimately useful in shaping the safety of work in your own organization. Join in our discussion on this episode on LinkedIn at The Safety of Work Group, or send any comments, questions, or ideas for future episodes to us at