Language selection


Innovate on Demand, Episode 17: A Conversation with Anil Arora (DDN2-P17)


In this episode, we sat down with Anil Arora, Chief Statistician of Canada, to discuss transformation within the public service and its impact on the way we use and interpret data.

Duration: 00:27:04
Published: November 23, 2020
Type: Podcast

Now playing

Innovate on Demand, Episode 17: A Conversation with Anil Arora

Transcript | All episodes
Listen on: Apple | Google | Spotify | Stitcher


Transcript: Innovate on Demand, Episode 17: A Conversation with Anil Arora

Anil Arora  0:00 
Ask me the questions and we'll just no question is, yeah, I'd rather answer your questions and kind of pretend. 

Natalie Crandall  0:09 
I feel like Anil is going to be the most relaxed of all of us. We should just get going

Todd Lyons  0:17 
I'm Todd Lyons.

Natalie Crandall  0:18 
I'm Natalie Crandall.

Anil Arora  0:20 
And I'm Anil Aurora.

Todd Lyons
And this is the Innovate on Demand podcast.

Todd Lyons  0:39 
Anil Aurora was a career employee of Statistics Canada, reaching the post of Assistant Chief Statistician, before moving to management positions in Natural Resources Canada, and later, Health Canada. Then, in September 2016, he was appointed Chief Statistician of Canada. We're fortunate to have him here in the studio today, as our guest.

Natalie Crandall  1:11 
Welcome Anil.

Anil Arora  1:12 
Thank you very much. Thanks, pleasure to be with you.

Natalie Crandall  1:14 
That's very exciting. We have quite a few questions prepared for you. I'm just joking. We don't have any questions prepared for you

Anil Arora
None? Oh wow, is that the end of the program

Todd Lyons

Natalie Crandall
we're here to talk a little bit about innovation, transformation, these things in the public service. And I guess my first question that I would ask of you is, so you are heading up Statistics Canada right now, an organization which I think probably traditionally has done things in a similar fashion for a long time and what it certainly appears from the outside that you guys are moving and shaking at an incredible rate. Right now. So maybe you could talk to us a little bit about what does that transformation look like? And maybe what are some of the major obstacles and opportunities to do this kind of work in the public service?

Anil Arora  2:05 
Well, thank you. I mean, there's a lot there in that question. So first of all, let me dispel any myths when you think about statisticians or Statistics Canada, that kind of set in their ways and they do certain things. Yes, there's no question there's a rigor that any kind of framework within which then you don't have a definition, and you have to stick to it has some rigor built into it. But let me tell you the reason why Statistics Canada, has been chosen by the economist as the best statistical agency in the world a couple of times, why we're invited to be heading up pretty well, every kind of major initiative in the statistical world is because we are innovative, because we actually kind of take some risks and intelligent risks and what people might not know about is we were the first in the world or the second in the world, I think it was to adopt the system of national accounts. My sort of predecessors, predecessor, predecessor was the first person to be chairing the United Nations statistical commission, because we believe Canada believed in making sure that the language of like trade and population growth and demographics and so on, we're really reflective of our values as Canadians. And so, we're the first ones to actually use some of those very, very complicated machines that do those very complicated calculations in the world. I was privileged to be involved with the first census that had an online response. And, people were like, what do you what is this online thing, the internet, it was just starting to take hold. And we built integrated systems where people could respond whichever way, and you'd be able to keep track of it for an entire country and think about that. Last census, we got 70% of Canadians nearly, just under 70, 68.9%, to be precise as that responded to the census online, that doesn't happen, because you kind of sit back and relax. So, yeah, I would say we're always building on a culture of innovation. And so to your question about transformation, so what how do you take it to that next level. And so that's where I think it takes an organization, good leadership's a little bit of resources, but it has this desire of an organization to give more to Canadians, and to be more present in so when you think about society today everybody's a data producer and anybody can produce something and put it on the, on the web and tomorrow, guess what it's taken as fact. Well, Statistics Canada wants to make sure that when Canadians, whether it's businesses, policymakers, you name it, when they use good information, we want to make sure it comes from a credible source. We want people to know the limitations and strengths of the data and the analysis, and that's what we want to do. We want to make sure that people are using good information to make decisions about know their loved ones their businesses, because we think that that's going to make a better Canada.

Natalie Crandall  5:10 
That's very interesting. So the world of that, in that data world has changed so significantly, in that we have so much data now. So how do we do that? How, how do we take that? I guess, how do you guys take that leadership role? And how we do that, that best stewardship of that data? And how do we find and figure out

Anil Arora  5:32 
First of all, let me let me just say 90% of the data that we have today was just created in the last two years. And if you go back two years, 90% of that was only created in the two years precedent proceeding that, and when you think about the number of devices, sensors that are being placed in, there's cars and under bridges, and fitbits, and other sort of devices and sought, data are just about to explode. So if you think there's a lot today, just wait five years and 10 years and 15 years routinely now I see you go to a restaurant, you see kids two, three years old, and they've got an iPad or some sort of a smart device. And guess what, there's nobody teaching them, they're just clicking and trying and doing things. So on the one hand, you're going to see data just completely overwhelm us. And it already is doing that, by the way. And you see a next generation that only knows how to do data and digital, and they have expectations that everything is going to be data and digital. That combination, if you like, have increased expectations. And the data is just yearning for us, as a society to say, how do we get better at it? How do we do it responsibly? How do we do it in a ways that there's quality baked into it, that we're not being duped by biases in our data sources? And so it's yearning for expertise. It's our yearning for infrastructure. And guess what?  The innovation that we are seeing now, I think we're just at the cusp of that. And you're going to see new applications, new ways of doing things, deeper insights. And when you when you think about what does that all have to do with Statistics Canada? So as organizations start to wake up to the reality, okay, and some are a little further along the way than then perhaps others. But when you see organizations that are struggling with some of the basic things about what is metadata? What is anonymized data? What is synthetic data, what is microdata? What is aggregate data, and so on, so on and so on. You see an organization like Statistics Canada that has been around, for 101 years and guess what we do, and do it arguably better than anybody else in the world, we know how to manage data. And we know how to do it responsibly. We know how to link things we know how to have good governance of data, we know what data management looks like, we have entire divisions for things that organizations struggle with saying, Well, how are we going to manage data acquisition, for example. This is where I think the country can benefit if you like, from an organization like ours, and that's what the public service is about, right? Being helpful. And so yes, we know, our core business is always about producing official statistics. But our mandate to unduplicate, to actually bring harmony within the federal family, within other across governments across our society is absolutely aligned to being helpful in being able to help organizations get better insights, unlock the value that's hidden in their, in their in their data stores. And so this is the kind of information that we're putting on our website, we're doing partnerships with the [Canada School of Public Service], Digital Academies, we're creating modules, we're creating datasets, so that we as a society can get better at it become more numerate and literate, and we can draw the value of that data. We're working on developing standards on data access and data sharing. So that we can have responsible use, we can in fact, get greater insights and greater utility, while protecting privacy and confidentiality.

Natalie Crandall  9:18 
That's great. I've noticed that a lot of people have mentioned their partnerships with Statistics Canada in the context of things we've talked about, even here today at the data conference. So it is certainly evident that StatCan is out there supporting a lot of the diet data science work that's happening in government. So what are some of the some of the opportunities as a as a public service writ large that we could kind of take advantage of that? How do people well get there?

Anil Arora  9:46 
First of all, we're honoured to be an active part of this conference. As  , we were one of the founders in previous years of this and this is exactly what my vision was that we were going to kick it off. Create that capacity, and we're going to hand it over to others, and still work in partnership with them to create this enriched ecosystem within the federal government of better utilization better data, responsible responsibility and stewardship, and so on.  I think it's a collective responsibility. And to your point about what was a future whole like, I think it is just incredibly rich with insights that we are going to glean, because we can now responsibly share data across the federal government, across different levels of government, in ways that we have, we haven't even imagined the kinds of things that we're doing with the province of British Columbia to understand labour market trends in education, because we can now link datasets while protecting privacy and confidentiality, These are the kinds of insights that we would have never been able to do without that that partnership model, the kinds of things that we're doing with Saskatchewan on recidivism and indigenous populations, and what some of the models are of interventions that would perhaps address some of these long held issues are really fascinating, the kind of work that we've done on opioids, and looking at data of people that have unfortunately passed away. But when you start to trace their histories back by linking justice and health data, and so on, you get a much, much richer portrait, if you like, of people who are leaving a digital trace, even while they've lost their lives. So they can serve as illustrative examples of save more lives by better policy interventions, and so on. So, I think we're just at the cusp of insights that we never had before, so that policymakers can then understand better, what are the levers they have, and actually be able to measure the impact of their interventions, and I think similar kinds of things are available on the regulatory side of things. I mean the data streams that we're getting now, by having animals being implanted with a chip by when they're born, we have everything about as a society, what if we were to use that data to create a competitive edge for Canadian producers, in the international markets to showing us as being responsible producers of products, and, and so on. So I think we're just at the cusp. Other countries, I would say, are leading in some sense. But I believe that Canada can be a true leader on these domains. And I think it starts with the federal government, working collaboratively to try and harness these kinds of advantages, and social districts, Canada is a small part of enriching an ecosystem. And so certainly my commitment is that we'll be there to help, while we are doing our kind of business of providing those insights through that, as I said, responsible use of data.

Natalie Crandall  13:01 
It's very interesting how a lot of the conversations today keep coming back to this is all about the various perspectives and the collaboration with a previous guest who said, it's not a government problem, it's a governance problem.

Anil Arora  13:18 
Well, look, it's an opportunity it's a, it's a full contact sport, as I say, but it's also a team sport data, our data are powerful information is powerful people define their worth by what insights or what data they put out. So I get it, I get it that these are these conflicting kinds of things on the one hand, this is an innate desire to distinguish and separate using information and so on. On the other hand, there's this innate desire to collaborate and share, and,  , bring greater insights, well, guess what welcome to the messy world, that's these things have to coexist. My aim, as I said, as an institute, leader of an institution, that's just amazing, is how do we do it in a way that we have quality embedded in it, that we don't do,  , damage to by thinking that data is saying something else, when it means something else. So, that expertise and things that most people may not necessarily sort of think about, as  perhaps the most sexy part of this whole business is really the most important part of this business that that how do you classify data? How do you make sure that about the biases, what are the metadata, what are the restrictions, is no sense in mixing up two data sources when one is only for adults, and the other one is the entire population thinking that it's representing the same thing, right. So those are the kinds of things that we, we need to avoid. And that's only going to happen because we're going to get better at asking the right questions, and having the infrastructure and having the expertise to do well with it. So as you as I said earlier the data is going to explode, but also the use of that data through machine learning and artificial intelligence, because they are voracious eaters, if you like, of data, the more you push in them, the better right? Well, if you don't know the characteristics or the limitations of that data, we could actually come out of there with erroneous kinds of results, and we certainly don't want to do that.

Natalie Crandall  15:08 
That's right. I'm still stuck on thinking of your original statistic where you're saying that 90% of the data we have available today was generated in last two years. So what we have today is only 10% of what we're going to have in two years.

Anil Arora  15:24 
We'll see, it's the kind of projections that have been put out there, in terms of sensors and devices and so on, spewing data, no it'll be in all forms, sizes and shapes, and some of it will be very technical in nature and scientific in nature.  Some people think this is something that's new in the sense, but it is, what's happening is that the velocity and the volume and the veracity of this is changing. But the use of data in clinical trials, for example, to come at statistical models of the risk and the benefit the harm risk benefit kind of equation, that's nothing new in the scientific world, using data to come at is this a correlation that fits a statistical model, not, that's nothing new. So the use of responsible data and applications is, is not new in itself. But what we are seeing is an explosion, if you like, where that use of data is disrupting, business models, disrupting organizations, and it's creating new wealth is creating new ideas and opportunities. I do believe that we're at the cusp, how much of it will be usable, and how much will we be able to consume an move our society and our economies forward? Well, we'll see. I think those are the challenges that we have to overcome.

Natalie Crandall  16:51 
I have one small question that Valaria Sosa who's normally my co host, but is on leave right now. sent along, and she specifically was curious about your role as a deputy minister, Deputy Minister level as our chief statistician. How, what are some of the I guess, best opportunities that you have to have that impact and that influence to open this space up for the people who work in your organization?

Anil Arora  17:21 
Well it's a dual thing, right, being at the deputy tables certainly gives me the ability to open some doors, that perhaps others may struggle a bit, my ability to phone up one of my colleague deputies, and have a conversation about a collaborative partnership and so on, to see how we can share from data to you, whatever have you is something that is a privilege. But it goes with a huge responsibility so how do we do it in a way that respects the utility, the privacy we want to be very, very mindful that as, as an entire system, we clearly have regulatory roles where there is no question that sharing of data, doesn't really it has to be done with extreme extreme caution. We're not building surveillance states to some people's kind of fears, right, I mean, we are building systems that on the one hand protect that fundamental if you like, expectation of citizens, that we're not  going to we're not going to use the data in in ways that are unethical. So, I think we have an opportunity as leaders to demonstrate that we are in it only for the public good we're not out there to make money, we're not out there to market and we're not out there to kind of do things that we'll leave to others, not that those are terrible goals or anything like that, but we have a singular focus, how do we make our society and our economy run better? And how do we have insights that policymakers and others have information to use responsibly, so, as a deputy, now, that ability to build that kind of responsible access and responsible use is something that I take very, very seriously. And then, absolutely, it comes with the ability to move resources to set governance to,  set a particular direction and a vision. And, again, it's very exciting to see when your vision and your idea starts to take shape, and it starts to have some resonance and people feel, yeah, it's valued. But, it comes with a with a deep sense of responsibility, because change and change management can be quite stressful to people. So, how do you as a leader ensure that an organization sees that. We will work together, we're there to make sure that they have the resources that are that are going to be needed to make that change, that they have a supportive culture, that when ideas germinate where there is an organization working together as colleagues, to see if we can get the best out of that and remove the obstacles out of the way, rather than to kill the idea. Being a deputy has certain sort of advantages, and that role comes with it, but it also has equally a huge set of responsibilities. It's a dual responsibility that I certainly take seriously. And I know my colleagues take very seriously as well. So we're working so when we think about things, like the data strategy for the federal government and we build it together, it is something that as a community, we take seriously, as leaders, and we are committed to making sure that it's carried out in a responsible way,

Natalie Crandall  21:00 
it's interesting to think that a lot of people at the working level, who want to push and work in that innovation space see themselves as requiring the cover from their senior leaders to have space and time that they need to do that work. And so it was, it's very interesting to see it from the other side, and I can see that it would be very, very anchored in that,  , that, that responsibility, accountability of making it work, and ensuring that the people working for you are able to do that

Anil Arora  21:30 
I think most deputies would agree with the comment that we own the kind of actions of our of our departments, yes, we know that there's a ministry of accountability. And in our kind of parliamentary system, there's accountability that we have to respect. But we deputies and heads of organizations take those responsibilities very, very seriously. So, we don't want to take stupid risks. On the other hand, we want to take intelligent risks, we know that failure is always something that we have to be mindful of, we need to make sure that those failures don't end up hurting anybody. And that we are transparent, and that we build the trust with citizens, with our colleagues across. That role of the deputy can sometimes be a bit lonely and challenging. And then sometimes you have to kind of shoulder when things don't go exactly as planned, but I think most Canadians would like us to take intelligent risks to kind of move the yardsticks for because they know that Canada isn't where it is because we just kind of sat and did nothing. So I think most Canadians would be fine, as long as we're transparent about things that we tried when they worked out, and when they didn't, and that we're acting responsibly, and that we're learning from those failures. Being the head of an organization comes with it some of those special if you like, responsibilities, and we're always trying to say, how do we kind of inculcate that culture and through the public service through our individual areas, and I think we're doing this exactly what we're trying to do, we're trying to say, build partnerships, understand some of the limitations build some of the mechanisms to better manage risk, and use response, you use resources and data that citizens have given us with the kind of care that they deserve. And I think by and large, an organization like ours, for example, this is what we do, we trade and privacy, right, we take a little bit of data turn it into insights, and then we lock it down now. So, so this is something that we're quite comfortable doing, there's a lot that prevents us from abusing it. Those responsibilities of access come with real legal obligations of protection. So those responsibilities have to be inherent, and that culture has to be propagated by the most senior levels.

Natalie Crandall  24:01 
thank you. Maybe I think we just have a couple minutes left. But I was wondering if there was maybe one file or something that you guys are working on, that you just love, and you wanted to tell us a little bit about, or give us maybe a sneak peek for next year's census on something that'll be new or interesting. Something you want to or a piece of advice you want to leave us with?

Anil Arora  24:22 
Yeah, I mean, there's so many innovative projects that are currently underway.  Most people have heard about our cannabis crowdsourcing and the use of wastewater to get at legal versus illegal portions if you like, of cannabis consumption in different parts of the country. If you haven't you might want to go to our website and have a look, we regularly not produce statistics on this using these kinds of new methods. Another area that we've been working on is how do we collect the vast amount of information that we get from agricultural operators without, ultimately, our goal is to produce the same if not better quality and detail of information without asking a single question of farmers. So we're using satellite imagery, which now is so good. The resolution is such that we can actually differentiate the crop that is being planted by farmers and what the yields are, we can calculate them using satellite imagery and working with the crop insurance companies, and we'll try to build this win-win kind of partnerships, we're doing work with our colleagues at CFIA to and the Pork Council to say, Okay, what can we do to kind of track if you like, the Piglet all the way to the, the market. We have a sense of sort of what that might offer from a health and safety perspective, and a regulatory and a competitiveness kind of perspective. These are through these kind of collaborative things, we're getting better statistics out to real users and clients, which then can have a positive impact on the industry. But it's also bringing us closer together as public servants, and colleagues, and saying how can we get how can we kind of build those, if you like, that expertise, and be more responsibly using data that we've got, which is essentially siloed and locked away. And so we're saying, Okay, how do we work horizontally, so that we can create greater value new ways of doing things, innovative ways of bringing greater value to Canadians?

Todd Lyons  26:36 
You've been listening to innovate on demand brought to you by the Canada School of Public Service.

Our music is by grapes.

I'm Todd Lyons, producer of this series. Thank you for listening


Todd Lyons
Canada School of Public Service

Natalie Crandall
Project Lead, Human Resources Business Intelligence
Canada School of Public Service

Anil Arora
Chief Statistician of Canada
Statistics Canada

Tell us what you think

Share your comments on this episode by using our feedback form.

Related links

Date modified: