How to Make Statistics Click with Schools Outreach and Interactive Data Visualization

Steve Mallett

Freelance Biostatistics Training | Data Visualisation Expert | Schools Outreach

Steve Mallett is an experienced biostatistician who develops training modules to demystify statistics for professionals who interpret clinical data. His visually engaging materials and interactive exercises help participants consolidate their knowledge and build confidence.

He also provides consultancy on applying effective data visualisation principles to explore clinical data and communicate findings through clear storytelling.

Learn more about Steve’s work at https://www.mallettstats.co.uk/

Transcript

00:00
You are listening to the Effective Statistician Podcast. The weekly podcast with Alexander Schacht and Benjamin Piske designed to help you reach your potential, great science and serve patients while having a great work-life balance.

00:21
In addition to our premium courses on the Effective Statistician Academy, we also have lots of free resources for you across all kind of different topics within that Academy. Head over to theeffectivestatistician.com and find the Academy and much more for you to become an effective statistician.

00:49
I’m producing this podcast in association with PSI, a community dedicated to leading and promoting the use of statistics within the healthcare industry for the benefit of patients. Join PSI today to further develop your statistical capabilities with access to the ever-growing video-on-demand content library, free registration to all PSI webinars, and much, more. Head over to the PSI website at PSIweb.org.

01:18
Learn more about PSI activities and become a PSI member today.

01:31
Welcome to another episode of the Effective Statistician. Today I talk with a good old friend, Steve Mallard. How are doing? Hi Alexei, very well, thank you. Very, very good. Today we will talk about a couple of topics that have a kind of interesting connection to each other. But first let’s talk about a role Steve has just started within PSI. And maybe you can talk a little bit about your career.

02:00
a very long career and how that led you to accepting this role? Yeah, sure. So I recently taken over the chair of the PSI schools team, which sits within the careers function. But yeah, it’s not the first time I’ve been involved. I used to be in the old Calc group, just careers and academic liaison team, probably about 10 years ago. Then I moved on to do other things, mainly data vis, which we can talk about. But yeah, I’ve got a little bit more spare time.

02:30
on my hands at the moment. So I’ve learned to the school’s team. So I’m now chairing the committee. So I’m just holding the fort while Rebecca runs hours on the tenancy leave. So yeah, it’s good to be back in the talk on liaising with schools and getting involved in the school’s outreach. have a lot of experience to share from kind of decades in the industry. So you’re not close to those that are still at school.

02:56
I worked in industry over 30 years as a statistician, about half that time, about 15 years. So hopefully I’ve got a lot of experience under my belt. As I say, I’ve done quite a lot of kind of school’s work. I’ve always been quite interested in training and communication and visualization is part of that as a means of communicating with stakeholders, whether those are students in a classroom or scientists in a pharma company.

03:22
There are some common themes there, but in terms of the schools, our focus really is on attending careers events and also going into schools and running workshops. So what we’re trying to do there is help the students kind of make a connection really between their academic studies. Some of the kids, they like maths, not all of them. Some of them are good at maths, but maybe don’t necessarily see the point of it. I did A level maths and then dropped it because I didn’t really see the purpose.

03:50
So it’s really helped to connect student studies with the kind of the world of work and the kind of work we do and help to show that maths can be applied in the real world. can be useful to society. And hopefully we have a bit of fun along the way as well. And then for the older students who are maybe starting to think about their options in the UK, would be GCSEs, A levels, university courses, have a lot of decisions to make. So we kind of help raise their awareness that…

04:19
You can be a statistician in the farm industry, it’s a thing. They may not even know it’s a thing. And then help them to map out a kind of career pathway into that role. So it’s really just informing them of what options may be ahead for them in the future. Hello, I’m German, I attended German schools. I don’t think I ever attended something like a careers event. How does that look like? They work at different levels. When I was with Calc previously, we used to go to a big national event.

04:49
I think more recently we tend to go to more local or regional events. So there might be a cluster of schools, organize. It’s a like a conference. I suppose you have exhibitors and have stands, have activities and workshops. So it’s really a day away from school for the students and teachers and really a chance for them to take part in hands-on activities. And along the way, just learn a little bit about. We’re not just going to show them some posters about careers that will be quite dull. So usually.

05:18
involves some sort of hands-on activity that’s quite fun, get them involved, and then maybe give them a flyer takeaway just so that both the students, I think, have fun, hopefully. The teachers also get some awareness of careers and probably get a mixture of teachers and careers leaders who, again, may not be even aware of what we do as a career option. So yeah, it works at different levels. I think more recently, it has tended to be more local than national.

05:46
pretty cool. I’m pretty sure in other countries, things like that exist as well. And there is actually one of the things that people reached out to me. There’s all this English material existing within your team and with your colleagues that can be translated in all kinds of different languages. Yeah. Of course, it’s easy to use. Now, it would be great if there would be some volunteers that work with this team.

06:13
So that they check that the translation actually makes sense and just makes sense also from a different age group, things like that. Yeah. This is something that came out of conference recently. So we’ve got very talented team within the schools group that are developed in an escape room, which is aimed at kind of primary school level, so up to year six, really good quality, really professional job. And that would be showcased that at the PSI conference and that gained a lot of interest. And as you say,

06:42
people from other countries of turn an interest. We’re trying to find translators because we’re trying conscious not to make everything we do and the UK centric or English speaking centric. But yeah, I guess if we translate them, as you say, it probably needs to be validated in some way. Just make sure that the workshops still work properly in other languages. So it’s great to have that interest and that we can go global. It’s quite exciting. Connect with Steve on his homepage, which will be linked in the show notes or on LinkedIn.

07:12
or check out the PSI Calc page or no careers page and you’ll find always the correct person if you listen to this much later than we record this kind of middle of 2026. The other point is translation is one thing, but execution is of course another thing. So there’s an opportunity for you to work in with your local schools, local school events to roll these things out.

07:39
And I think it’s a great investment in our community. And so far what I’ve seen, most companies are really happy to support these kinds of activities. Yeah. Something we’ve found is in recent years, that’s another difference I’ve found since when I used to do the careers work is that we have a lot of volunteers who put themselves forward. They’re interested in schools outreach, but maybe don’t yet have the confidence to actually come forward and run a workshop.

08:07
And so what we’re trying to do, in fact, ran a workshop earlier this week. had a group of more experienced presenters running the workshop and then we brought in some volunteers just to help provide support. So that’s a model we’re moving forward with so that people want to help, but they don’t actually want. It’s quite a responsibility to be standing in front of a classroom and leading a workshop. There’s a few of us that will do that, but we always need all hands to pump to facilitate the workshop, answer questions, talk to the students, bring in a larger team. So.

08:36
This is a call for you. People have volunteered interests, but they haven’t quite made that jump yet. Please contact us and we can maybe buddy you up with someone more experienced and give you a bad experience in the classroom. Yeah. Or if you’re experienced, we can buddy up with someone that is less experienced. And not a huge amount of work. Yeah. Definitely doable and a really, really nice activity. And it’s great that PSI provides that opportunity. You also don’t need to be a PSI member.

09:06
to do that. Of course, would be great if you become a PSI member, but that’s another topic. By the way, the PSI membership last time, I think it was something around £100 per year. It’s really not a lot and you get lots of very nice, lots of different activities, reduced price at the conference, access to journals and whatsoever. That’s an easier investment and most companies even reimburse you for that.

09:35
So that’s another topic. You also talked about these classroom events where you actually go to a school and you do something within the school. How does that look like? Yeah. So in fact, we just ran one this week for year 10 and 11. So what we did was the whole school, actually, it was a very small school, about 50 students. So yeah, we started with a brief presentation. Why do we need stats? And there were a few kind of relatable examples about.

10:04
drinking energy drinks and measuring your pulse rate. But then we wrote them into two classrooms and we ran two workshops. We’ve got some new workshops now that use R Shiny. So that’s something I’ve been keen to implement. Cause a lot of the old workshops were quite old fashioned and they were quite cumbersome. I remember working with Claire Britain years ago on a kind of Coke taste test. We had to carry all these bottles of Coke across London. And it’s very scalable now. All the students had a tablet, so they just connect to the Shiny app.

10:32
And can move classrooms or whatever, doesn’t matter. So it’s a lot easier to run. And so, yeah, we basically ran a couple of workshops, one of them called the Stroop test, which was great fun, clinical trial simulation together after the break and did a kind of speed networking session. But they got to kind of speak to us about what we do for a job and ask questions and so on. So, yeah, I think it was quite successful, kind of mayhem at times, but yeah, so that’s kind of gives a flavor and.

10:59
As you’d expect, not all the students are a hundred percent engaged. Some aren’t. You’re going to get a wide variety of interests and abilities. But yeah, hopefully we’re planting a few seeds. And even if there’s a few students in there that are years to come and they come to make their choices, they remember us or they go to our website. Then I think it’s worthwhile. Yeah, completely agree. I think it’s also to show that what kind of people work actually in the pharma industry. Yeah. I think the pharma industry from the outside is this monolithic

11:29
block, yeah, and also very often doesn’t have the best branding, to be honest. Although from the inside, we always think about the people that take care of the patients and we should do very noble things. But from the outside, people look at lots of the things that the pharma industry has done wrong or is still doing wrong. Just talking about prices and things like that. Yeah. And so connecting with people in the industry, I think it’s

11:58
good for the overall industry, not just for our function. Yeah. I think it’s in two parts really. There’s the stereotype of a statistician and what people think if they had to draw a statistician, it might be somebody with leather or patches on the beard or I don’t know. It’s kind of an academic. So when you have people, young volunteers, there’s a lot of variety and diversity and people that look like them, but just a few years older. I tend to step back because I’ve got grey hair. You still have hair. I don’t have any hair.

12:28
But I’m probably thinking I’m sort of granddad to them, but I think it’s important for them to see someone that looks like them or they’re relatable. And they think, oh, that could be me. Cause a lot of choices come down to self-image or I’m not a, that sort of person. So it’s breaking down the stereotype around statisticians. And then as you say, people probably need a lot of things online about the farm industry, lot of misinformation. So I think it’s both of those things. So you just mentioned R Shiny interactive data visualization. How has this changed?

12:57
improved the way you communicate data to the students. One of the workshops, they’ll draw a histogram, assimilate some data and they can play with the sliders to change the variability. And then they can look at the histogram changing. And I may have done that in the classroom. They might not quite understood the idea of variability, but then by the side, there’s all these emoticons. It’s basically the patients and then funny faces and show how much pain they’re in. Some of them are smiling, some of are crying. And so it’s kind of making that link between the abstraction, which is the

13:27
histogram and real world patients. And then we ask them questions about if you’re a patient, how you do feel about, compare two groups, but there’s a bit of overlap. So we compare the means, but not all the patients do well. So we ask them, if you’re a patient, how would you feel about this medicine? So it’s really connecting what they may have learned in the classroom, which may be quite dry with something that’s quite visually appealing and relatable. And this is the link in with the data visualization.

13:54
SIG work, because quite often we want people to look at, so might have some modeling results or something, but, or distribution, but you also want to see the patient level data as well. And to make that connection, the different levels of abstraction available. yeah, I’ve been borrowing some of the thinking from the wonderful Wednesday webinar that we do. And obviously the audience is very different. Just trying to borrow some ideas. Simplicity is important. Clarity, removing any clutter and making a screen so it just looks.

14:22
And it’s also, we talk this thing about know your audience. So your audience might be clinicians or decision makers or managers. have to get, think about getting into their shoes and look at, think about what their level of understanding is, what their level of motivation is. The same applies to a bunch of children in school as well. Why should they be interested? If it’s fun, they’re going to take part and what do they know already? Maybe they’ve seen a histogram in their last lesson and now they see something they’ve recognized and they can now play with it. So it’s all about.

14:50
relating to your audience. That’s a kind of common thread, I suppose, between all these different things. That’s a very, very important point. And for those listeners who don’t know about the data visualization special interest group, shortly called DataVis SIG, it’s a special interest group that is organized by PSI and FSPY and runs now for a couple of years, six, seven years, a monthly data visualization challenge.

15:19
And that is called the wonderful Wednesday webinar series. So if you go to the PSI homepage, PSI.org, you will find the all the special interest groups and there you can find the also the data visualization. So it’s probably the last one on the page. And by the way, you will find there also the careers page and within that schools outreach and there’s also other groups more about university and other careers stuff.

15:48
So I really love the idea of interactive data visualization, that you can, for example, play with the assumptions so that you can see what is changing, what does doubling the standard deviation means. If you change the baseline distribution, what’s happening. If you change correlations, what is happening. All of these things, you can have a better idea in terms of for your design.

16:16
or for your analysis part. I have completely the same experience when you use tables. Yeah, it’s usually pretty boring and people don’t see changes so much. Because making calculations in your head is maybe easy for us, but not easy for non-statisticians or for people that are not so easy with mathematics. And very often we have such people, yeah. When we talk to physicians and

16:46
other people. They studied these kinds of things because maybe they didn’t like mathematics. And they are explaining things in visual things is so much easier for them. Also to, for example, understand formula. We might see a formula and then understand what’s happening there. But most people don’t. By far, most people don’t. So just bringing complex formula is the last time they have seen an integral decades ago. So

17:15
You also have a lot of experience and training non-statisticians, physicians, colleagues within your company, but also outside of your company. How do you typically approach such a training? Yeah. As I said before, I’ve always been quite interested in explaining and training in part of my day job, I suppose. And now I’m, I’ve now left employment. I’m working with freelance capacity. So I’m developing kind of training.

17:45
and other statisticians, it’s what I quite enjoy doing. Like you say, it’s this common thread, it’s very tempting to create all these slides, they’re looking really cool, data visualizations, animations, but the key is turning things around and saying, what do the stakeholders need? So if I’m presenting to MSLs or medical writers or understanding, what are the things they struggle with? Cause that might put a lot of effort into explaining what a p-value is. They know that, but they don’t know they struggle with something else. Part of the challenge is really understanding.

18:13
What are the pinch points from their perspective and building the training kind of really around that. Again, it comes back to somebody in a classroom or it’s a person’s motivation. Why do they care? Because if you’re not addressing that becomes your agenda rather than theirs and they’re going to switch off. But having said that, I’m quite interested in my favorite new word is pedagogy, which is as I understand it, it’s the kind of science or study of teaching methods. How do you convey, how do you teach something? It might be kind of stepwise. You really want to anchor an important concept first of lots of

18:43
exercises, that’s secure in that concept before we move on to the next one. Using a lot of interactivity. I’m quite interested in animation because a statistician, we’re very familiar with say, probability distributions, but a lot of people, they’ll look at a bell-shaped curve and they’ll draw, oh, that’s mathsy. But if you simulate draws from that population, so these could be your studies, each one of these could be your study and you see they’re moving around, then you can talk about what’s standard error.

19:09
means and how you construct the confidence interval and what it actually means. Hopefully they’ll it in such a way, step by step, that they finally think, finally, I understand it. That’s the aim. And also through interactivity, they’ll build a tool just for exploring power. Just by increasing subject sample size, they can see the curves get tighter, the critical value changes, which is on a scale they understand. Oh yes, the patients don’t need to achieve so much that hopefully pennies will drop.

19:38
when they get their hands on the tools. So yeah, I’ve given a lot of thought to teaching methods, but I think for me, the next stage will be to get a lot of feedback from people to see if I’m on the right track. And I love that you always start with the audience. Yeah. I think it’s so easy to assume the audience needs X, Y and Z. Yeah. Or to assume that it doesn’t matter whether I talk to diabetologists or to oncology people or to neuroscience people or to

20:07
whatsoever. But actually they care. I can’t use a neuroscience example with oncology people. I can’t use an oncology example with diabetes people. They will say, oh, I don’t care about survival, at least not in my area. My patients survive. Diabetes is largely under control. Why do we want to talk about survival data? And so you always need to make it

20:37
specific for the different people. And then training becomes so much more effective. Now, of course, that is more work. And also preparing it in such a way what you just talked about. With interactivity, with good data visualization, with fine-tuned slides that are not overloaded with all kinds of different things, but fun to look at. That is quite a lot of work. And it’s…

21:06
very often easier actually to call someone in to do that. know all those statisticians I know when I talk to them, they say, I’m buried in work. And for them, the next FDA response is much more important than giving a training or getting the publication out of the door is much more important than the training or designing and study is much more important than the training. Yeah. And in

21:35
all these kind of scenarios, it’s really great to have someone like Steve that can help you support better your training. Because there’s a huge training needs. As soon as you start to speak about your data with clients, with treating physicians, with QAPL leaders, the MSLs, you have usually hundreds of people that you need to train on your data.

22:05
And not just on your data, but also on the competitor data. That’s the other part. And of course, these kinds of things change all the time. There’s always a new, depending on which area you work, but in also a little bit more crowded areas, there’s always a new publication coming out. And then people will think, oh, why do these data look different than our? Why do they appear to be different than our?

22:33
Oh, they have used a different way to adjust for multiplicity. How does that relate to our? Oh, they have stopped early. Does that mean that they are better? Hundreds of questions. And that’s really good to have someone that can talk to your different people. And that’s a great way also to get more relationships of the overall stats department.

23:01
all these marketing people, sales people, MSL people, so the medical scientific areas that are usually in the medical affairs area. Yeah, I mean, I have found there’s enormous appetite. You might think people that don’t get statistics or struggle with statistics, then they might turn off. But in fact, they understand certain roles like MSLs who are maybe in the commercial organization and they have to do the A’s with Salesforce. They want to get it right. They have to say things about clinical trial results.

23:29
very anxious not to say the wrong thing. And they don’t have to keep referring to a statistician. They want to be able to stand on their own two feet and speak confidently about what is multiplicity. So I found that there’s an appetite there for people to build their confidence as part of their professional development. yeah, like you say, I found this myself when I worked in big pharma that quite often it’s under-resourced. if you’re a project statistician or you don’t have time to talk to the MSLs when you’re in the middle of paying for an ad board or something like that. So yeah, so it’s interesting.

23:59
I think the other point is it’s usually not a question of money. It’s a question of ad count. The marketing and sales organizations, they have more than enough money to spend for a couple of hours for training and things like this. That is lost in the roundings of the overall marketing budget, I can tell you. So it’s really fun activity. When I was freelancing the last couple of years, I’ve done that as well. Always a really, really fun thing to do.

24:27
And as a statistician, if you work in a global team, there’s no way you can respond to every request that comes from Austria and Sweden and Argentina and Japan and South Korea and somewhere in the US. Yeah. That’s impossible. Yeah. So for that purpose, it’s really, good to have someone like Steve. Yeah. And to connect it back into the schools again, there’s a theme that whoever presented to them, helping them build.

24:57
demystify statistics and help build confidence. that whether you’re a year 11 or whether you’re working in a pharma company, you might think statistics is a mystery to me. It’s mathematical, there’s formulas and Greek letters, and it’s not actually that difficult if you look at it in a certain way, or you can build confidence with it, confidence in people, how people view it and how they interact with it. So yeah, I do see a common thread, even though it’s at a very different level that…

25:22
how people perceive statistics, even the word quite often puts people off. It’s something I’m certainly passionate about is breaking down those barriers. So to sum it up, there’s a couple of key learnings I would say. So first is you need to always start with your audiences. Yeah. For any communication presentations, especially your trainings. Yeah. What is their pain point? Start from there. So second is in order to explain these data.

25:50
In order to explain more difficult concepts, data visualizations are really, really good tools. And if you want to go through different scenarios and things like that, interactive data visualizations could be our shiny, could be something else super helpful. And in the end, it’s all about good communication skills. You need to be able to stand there confidently, talk about your things and

26:19
then you can create these aha moments that make it much easier for people to then also confidently talk about what they have learned. Absolutely. And a big shout out to the school’s team, because even though I’m chairing the committee, my life’s made so much easier by having a really talented team. So I’m working with Alex Spears, who’s an ex-teacher. And he taught me the word pedagogy and he just thinks that way like a teacher. That’s really helped. We’ve got some placement students who are really talented.

26:49
very energetic. So that makes my life not easier. I’m just kind of at the helm, but they’re doing a lot of the real work. Yeah. I just wanted to give a shout out to the school’s team doing a great job. Thanks so much. And I have the feeling we’ll probably have you again for another podcast. Maybe we’ll talk a little bit more about data visualization, because we just talked a little bit about what the data visualization SICK is doing, but that will be a lot of fun. Yeah. I’d love to come back.

27:21
This show was created in association with PSI. Thanks to Reien for the help with the show in the background and thank you for listening. Reach your potential, leak great science and serve the patients. Just be an effective statistician.