Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #99: Cambria Brown
170 Plays7 years ago
Cambria Brown is a Performance Management and Data Specialist with the Office of Planning, Partnerships and Improvement at the Colorado Department of Public Health and Environment. She specializes in performance management, data visualization and data dashboard development for divisions and...
The post Episode #99: Cambria Brown appeared first on PolicyViz.
Recommended
Transcript
Introduction to Episode and Guest
00:00:11
Speaker
Welcome back to the PolicyViz podcast. I'm your host, John Schwabish. On this week's episode, we're actually going to go into data at the state level, how people use data and communicate data from the state agency level. So I'm very happy to have on the show Cambria Brown, who's a data specialist at the Colorado Department of Public Health and Environment. Cambria, welcome to the show. Thank you for having me. How are you doing in lovely Colorado?
00:00:39
Speaker
I'm good. We've had actually a nice mild August, which is unusual. Normally it's really hot in August. Oh, okay. Well, you know, we've had the same thing here in DC. It's actually been lovely. So really no complaints weather wise. Data wise, I'm hoping we don't have many complaints. So we've spoken a lot in the last few months. I'm really excited to hear a little bit more about the projects that you're doing over
Cambria's Role and Background
00:01:02
Speaker
there.
00:01:02
Speaker
and the responsibilities you have when it comes to communicating state-level data. Can we start by having you talk a little bit about yourself so people get a sense of your background and your role and what it means to create visualizations and dashboards at a state agency? Yes, definitely. I have my master's in public health and epidemiology and biostatistics. I went to school to really learn the ins and outs of
00:01:28
Speaker
how to analyze health-related survey data and clinical data. And then I worked for a national research group that did data analysis from national survey data on children, youth, and families. And I spent about three years taking national survey data and analyzing it and then just making it available and approachable to a variety of audiences, so anywhere from family advocates to researchers to policymakers to
00:01:58
Speaker
and doctors. So it was really fun because it was really the start to finish of like collecting the data, developing surveys, taking the raw numbers, waiting them to reflect the population and then making it digestible for all those audiences that I already mentioned.
00:02:15
Speaker
And now for the past actually four and a half years, I don't know where all that time went, I've worked for the Colorado Department of Public Health and Environment.
Statewide Public Health Initiatives
00:02:25
Speaker
And so I work in an office called the Office of Planning and Partnerships and Improvement. And for my first couple years in the office, my role was to really work on statewide public health assessment and public health improvement planning.
00:02:38
Speaker
What we did was take data, look at the state of Colorado in terms of how does our population look in terms of the various factors that we know influence health, so everything from like
00:02:51
Speaker
um, income and education and housing, health behaviors, and then health outcomes. So I worked on creating, it was a few years ago, but the giant report, like a static data report of that was our statewide health assessment, which we then use to develop our statewide public health improvement plan. And those are really things that we work on with our partner agencies to look at what are the priority needs in Colorado? What are the issues that are facing our population and how are we going to work to address these issues? So that's just.
00:03:21
Speaker
what we do in public health, we say, what does the population need and how are we going to make it better? And then you hope five years later, you made some kind of a dent and then there's different needs of the population. Only normally it's more like 10 to 20 years. But, um, so, and now I've really shifted to looking more internally.
Internal Performance Management Shift
00:03:38
Speaker
So I'm as the data specialist, I would do what's called performance management. And that's basically just what we call business intelligence, um, as a government agency.
00:03:47
Speaker
And we're not very good at doing that. We're really good at pushing data out. We have a ton of data, but we have a ton of data requests from stakeholders and from the governor's office and from the Office of State Planning and Budgeting. But we aren't very good at saying, here's some metrics that are important to me, my division, my program. We're going to look at them on a regular basis and use them to make decisions about, you know, here's our workload or here's
00:04:15
Speaker
Some processes we need to improve are not very good at that at all. So what I'm doing is working at all levels of the department, everywhere from our department wide strategic plan down to the division branch program level to say, let me help you think about some metrics that are important for you to look at a regular basis, visualize them so that you have a place to go to see how things look and make adjustments or make decisions accordingly, have those conversations.
Challenges in Internal Data Communication
00:04:42
Speaker
It's really interesting because in some ways it's backwards of how a lot of places work where they don't do the public communication very well. They figure out how to talk to each other but not how to get information or data out. It sounds like it's sort of the opposite there where there's been a big emphasis or
00:05:01
Speaker
concentration on doing a good job of getting information out to the public but not so much of analysts talking to each other or understanding their data better. That's really fascinating. When you are talking to agencies and talking to analysts and managers and saying, okay, here's a better way, do you get a lot of pushback? What's the culture like and what does it take for someone like you to come in and say, hey, we're going to improve the way we do this internally?
00:05:27
Speaker
So it really varies because we're a really big department and we have different programs that are at different levels of their kind of sophistication with using data. So we actually have an entire data division and they're pretty good at it. They like to find innovative ways of communicating their data. They're a little resistant to the external view in terms of like they know data. They don't really want our office coming in and saying,
00:05:52
Speaker
here's what we think you should do. So we don't, because they do it really well. They don't need our help. Then we have some divisions that are the polar opposite. They have so much data, for example, permitting programs that collect a bunch of information about water permits or waste permitting.
Promoting a Data-Driven Culture
00:06:10
Speaker
And they need to collect that, but they kind of just sit on it. I mean, they do use it.
00:06:15
Speaker
They're not in the practice of using it regularly or at all. Or if they do, they really like their system of like, well, he sends me the numbers, I enter it in the Excel spreadsheet and I really like my pie charts and I don't want anything from them.
00:06:31
Speaker
everywhere in between where some people see what we're offering and they just jump and they say, yes, please, we want you to help with this and this. And that's great. And I really like that. And so it's really the whole gamut. Um, but what we've really, the approach we've taken to make this culture shift is really starting with the people, the middle group, the people that do want the help, the people that see the value. Um, and then with the hopes, like you and I've talked about before that that will spread from there. So, so that you, you do those, those kind of low hanging fruit.
00:07:00
Speaker
examples, you you create some products that really almost kind of blow people's minds in terms of not even realizing
00:07:08
Speaker
how approachable and how easy data can be to access, and then you hope that the others kind of come along later. There's a lot of people who work with data at the department, but I'm really the only person doing this internal data visualization, so there's way more work than one person can do. I just start with the people who want the help, and it really has grown from there. I've only been doing it for a little over a year.
00:07:33
Speaker
I can't believe how much progress we've made. So how did you get interested in doing data visualization as sort of your primary responsibility and like being the person in some ways whose job it is to like spread the gospel and get all these people to buy into it?
Passion for Data Visualization
00:07:50
Speaker
Because I like it. No, that really is what happens. So we needed a way to track that plan that I was talking about earlier, that public improvement plan. And then the data division needed a better tool for communicating
00:08:07
Speaker
They have a ton of data, survey data, vital records data, and so they wanted a better tool for their external website. So they've been exploring Tableau and they kind of demonstrated to us how we could also implement Tableau for plan tracking. And I had that experience as a biostatistician working with like
00:08:26
Speaker
statistical analysis programs, but I hadn't gotten to do that for a few years. So then learning Tableau and kind of getting to use that part of my brain a little bit, like not like I'm doing multivariate analysis or anything, but it's still a similar skill set. I just really, I just took it on. I really liked it. People saw the products and they were like, it just grew very organic. It was like,
00:08:49
Speaker
I started, I learned, I looked for opportunities and now we were looking also for performance management system. So there's a lot of products that are just performance management. So as opposed to data visualizations, we'd consider that. But just with the way that we saw Tableau could meet a variety of needs, I just kind of saw the potential and took it and ran and I worked for a small office.
00:09:12
Speaker
state government isn't really thought of as particularly agile. And we're not as a department of 1,300 people, we're not agile. But my little office of 10 is really nice and agile and has that flexibility to be innovative. So luckily, my director saw the potential as well and just kind of created the environment for me to take it and run. So helped create this structure for we're meeting with everyone regularly, we're demonstrating what we can do. And then we've also created what we call performance reporting forums so that
00:09:43
Speaker
We each division reports once a year and we report on how we're doing on the strategic plan as well and all staff are invited so it's been a great way to demonstrate the work and we get a lot of other you know
00:09:55
Speaker
We're able to promote it that way as well.
Transition to New Tools like Tableau
00:09:58
Speaker
I'm guessing that a lot of the people in the agency as a whole are Excel users or maybe their own statistical programs like State or SPSS or whatever it is, but it sounds like mostly Excel. When you come in and say, we have this new tool and you can create this interactive thing or you create this visualization, it's a little bit easier to do in Tableau than it is in Excel.
00:10:20
Speaker
I guess my question is, do people push back because they don't want to learn a new tool, or are they pushing back just on the philosophy of, no, no, no, this is how I've always done it, I get the data, I type it in, or like you said earlier, I make my pie chart and I don't want to try something new, or is it more of a tool constraint? I don't want to have to learn a whole new tool, or is it both? My guess is that's going to be a little bit of both.
00:10:44
Speaker
Well, the learning the new tool part, the way we sell it is I say I'll do it for you. So we do have a lot of other Tableau developers who have external needs. There's probably about 20 or 30 people in the department who have Tableau licenses. But for this internal need, I'm the only one really doing it. So unless they have someone in their program who knows Tableau, we typically say, we'll do it for you. So the learning really isn't a barrier. I would say the barriers are either, I like my process. I don't see the value add.
00:11:14
Speaker
of like I have everything already set up, I'm used to the system, I don't really need you to recreate it in a more automated way. Or there's people, you know, there's people who've been at the department for 20 to 30 years who have seen a lot of like
00:11:30
Speaker
Oh, the brand new shiny tool on. So they're like, they just think we're selling the brand new shiny tool. That's going to be gone in a year. And I'm like, no, I'm not going to let this one go away. I mean, regardless of what the actual software is, having a data visualization software like this, um, it's not going to go away. And then the other thing is sometimes they just, it just takes a while to get there because it's so.
00:11:55
Speaker
So, you know, if you have your database, the access database, and you're so used to like having someone on your staff pull the numbers for you, put an Excel spreadsheet, you put in your own, like you're so used to that process, you don't even think about the fact of like, you don't even, you just connect Tableau directly to the access database, do your calculations, like it's just completely automated. And it's just so different from what they're used to that sometimes it's just a good 10 minute conversation to get them to realize it. And then we're like, oh yeah, I am interested in that. They just, they don't say that first.
00:12:25
Speaker
Yeah, yeah. No, it's really interesting. I had a similar discovery a few days ago when I was talking to someone. They just had a table. I said, well, they wanted to do the sparkly thing. I said, what if we just did spark lines so you could see these little trends? Oh, yeah, but no one does that to look and graph the numbers. Yeah. And changing the way people think is interesting is one of the biggest challenges.
00:12:46
Speaker
I also wanted to get your thoughts on data visualization as a field, because you've been working with data for a long time, but I get the sense that you're relatively new to the field of data visualization as a field.
Successes in Data Visualization and Communication
00:12:58
Speaker
So are there particular challenges that you face personally, both, I guess, the community and how you approach the task of visualizing data? Yeah, so you're right. I am relatively new. I used to just call it data communication, and I didn't really know it was a field.
00:13:16
Speaker
I, it's been really nice to learn that it's a field to learn about the resources. The second I learn about a best practice, I just feel like I just grab onto it and start implementing it right away. So it's been really nice to like see people's examples. Um, a challenge, I don't know if this has to do with data visualization, but I really liked the development and the creative phase of it. So I'm not great at, um,
00:13:41
Speaker
Once I finish a project, I'm just like, want to move on to the next thing. I'm not great at promoting it and doing that. Really. Let's get it out there. Let's get feedback. Let's make sure that it's meeting people's needs because there's so much to do and I get so excited for creative phase. Not that I want to do those iterations too. I just not super intentional about getting that feedback and making, you know, which is a really important part of data visualization, you know, all the best practices, but if your audience isn't
00:14:09
Speaker
getting the information they need from your product, you know? It sounds like a lot of the people that you're working with are internal. If their workflow is a little bit easier, if they're doing a better job analyzing the data because they can see it in a better way, I mean, we can always solicit feedback, I guess. But I guess your audience is in some ways different than a lot of people who, you know, they're trying to get stuff public facing and get lots of feedback where you're enabling people to do better work, which is exciting.
00:14:37
Speaker
I think it's true, but I also, you know, I'll build a dashboard because I have to, because their performance reporting form is coming up and they're like, Oh, this is great. And then they forget to go use it. And they're like, Oh, I use these metrics all the time, but I just, I still get them from my spreadsheet. And so that it's still a big cultural shift of like, we do need that internal feedback to make sure that that's the place they're going.
00:14:58
Speaker
We know the information that they want is there. It's still not always the place they're going to get it. So do you think there's a piece that's missing? Because it seems like you're there. I feel like you have filled all of the holes that I can think of. You've taken down the barrier of having to learn a new tool because you're doing it for them. You're going to their office and showing them a better way. And yet some people still aren't doing it. Is it a management thing? Does there still need to be a carrot and a stick sort of thing?
00:15:27
Speaker
Or is it just like, it's one of those things where it just takes time? I think it is somewhat of a management thing. I think when you have management at all levels saying this is really important to do, I think that helps a ton. And we don't have that. So we have management within some programs and some levels that are taking and running. And they buy me gift cards because they're so excited about what I made them. And they're like, we were able to have all these conversations at our meeting that we'd never been able to have before just because we couldn't see it. We couldn't see in front of us.
00:15:56
Speaker
And so it's just the gamut, and I've probably been overemphasizing the kind of the resistors, but there's plenty of people where it really has taken off and they're just, they're chomping at the bit and are so grateful for how much we've been able to like advance their work by just putting the information they need in front of them in a digestible way. For the successes, did they now turn around and say, oh, the stuff they were putting out publicly, we can improve that? Are they now starting to think about like the external part?
00:16:26
Speaker
Yes, definitely. I haven't talked about it much, but I actually get a lot of requests for those external reports too, and that's not my primary job, but luckily I have the flexibility to take on those projects. I've done several of those external reports too, and that has largely grown from these internal conversations.
Comprehensive Health Profile Project
00:16:44
Speaker
They're like, well, what I really need right now is I have this emissions inventory report that I need to get out there. Can you help build that?
00:16:52
Speaker
So they do see the potential and then want to improve the external stuff. And it's not just me. I mean, I'm not doing all the work for the whole department. It's a lot of the other developers to see it and then want to emulate. And we also we brought in some people to do a training on best practices and data visualization last summer. And we had a ton of interest in that. And so I do think especially among some of the newer employees, there's a lot of interest there. Yeah.
00:17:18
Speaker
Yeah, that's great. That's great. I do want to talk a little bit about some of the public pieces you have out because there's some really interesting uses of Tableau that at least the, you know, the data is folks that I watch and learn from I don't, you know, sort of a different take. So I want to talk you have a health profile visualization that you use for the state. Can you talk a little bit about that one to start? And then there's another one that you did on suicides in Colorado that I thought was really cool to cool for the technique, not for the topic. Yeah,
00:17:49
Speaker
But the health profile one is really interesting. Can you talk a little bit about that one and both the background design of it and then the goals of the communication side of it? Definitely. We're again a big department. On the public health side, we have a lot of different public health programs that are working to impact similar issues but are funded through different funding streams. What happens is
00:18:14
Speaker
You have one program that's working on mental health promotion, and another program that's working on youth substance abuse prevention, and another one that's working on youth tobacco prevention. And all their funding comes from different funding streams from the federal government. So then they push that funding out to locals, all these different funding streams. And by locals, I mean local public health, which is one of our primary partners.
00:18:38
Speaker
Well, local public health is actually most of the agencies are relatively small. And so they get all these different funding streams, but they use it to implement to do similar types of work, like they kind of braid the funds together to do a cohesive program. But we don't get that picture at the state because we're very siloed. There's a tobacco program and there's, you know, there's all these different programs.
00:18:58
Speaker
And so even though we're all working with the same communities. So I've heard it ever since I started four and a half years ago that it's like we just need a better picture of which communities have which grants, what projects are they working on. So that us, we provide technical assistance, not me, but people at the state provide technical assistance so that they can get that better picture and provide that support or also so that we can see where the gaps are. So maybe there's certain parts of the state that have
00:19:24
Speaker
three or four different funding streams to work in the same kind of air shoe area and other parts of the state that don't have any funding to work in that issue area. And we just have never had a good way to get that, just take that step back and look outside the program, just within the program walls.
00:19:40
Speaker
So I had seen all over, so people had created Excel spreadsheets mapping different things together. I was like, all right, I'm going to take everyone's Excel spreadsheets, reformat them a little bit, put it in Tableau. And instead of having like six different concept maps in Excel, we're going to put it on Tableau. You select your community and you can see what funding streams do they have? What are they working on?
Development of Suicides Dashboard
00:20:05
Speaker
What are they using that money for?
00:20:07
Speaker
across the different funding streams. So someone from each one can go and get that global view for the communities that they're working with. And that actually, I don't know, I just was like, it's not really data visualization. I'm just using Tableau as a tool to pull in more action.
00:20:26
Speaker
information from I think I have a good seven or eight different Google sheets that populate it and right now it only covers about four different programs but I have had we buy this is one actually have been promoting and so every time I show it to someone they're like I want all my programs on here so I'm
00:20:45
Speaker
It's great. The vision was for it to be more comprehensive. I just started with the information I had readily available. I'm actively adding a good eight
00:20:57
Speaker
eight or 12 more programs to that profile, so it's even more comprehensive. That's been a really good success. Yeah, that's a great one. I'll link that one on the show page. The other one I wanted to ask you about was the suicides dashboard visualization piece. And this is actually how we got in touch. A friend of mine in Colorado works on this at the Denver Hospital on suicides and gun prevention. So I was curious about that piece you just want to talk about a little bit.
00:21:26
Speaker
Yes. So that one was one of those products that came from when we had that internal performance management meeting with that program. And afterwards, one of the people just chatted with me of like, would this be a good use for Tableau? And I was like, yes, this would be a great use for Tableau because we have a high suicide rate in Colorado, one of the highest in the nation. We have just a lot of initiatives to try to make a dent in that. And despite all the work that's being done, the rate just continues to go up.
00:21:53
Speaker
Um, so we have like a Colorado suicide collaborative that really hadn't wanted to get the data to tell a better story. So they had been working with a series of like 12 or 14 maps and each one was a static separate PDF map of your suicide by age, your suicide by County. And like, it was just hard to get that global picture of like, really where is the burden? Um, so.
00:22:18
Speaker
I just worked with the data expert, Ethan Jamison, who is really new to the ins and outs of the actual data sources. And I did the visualization side and we created like a three-tab profile that really just helps people working in suicide prevention identify
00:22:35
Speaker
where the burden is, what populations are being most effective, just really readily, whether they're working at the state or county level. And then also you're able to drill in to see what are the industry and occupation of those, where the suicides are so that you can really target your efforts. So actually in Colorado, we see a huge chunk of the suicides are actually among middle aged men working in the construction industry. Well, that's huge to know because
00:23:02
Speaker
Now you can have more groups. You can target those groups, yeah. Yeah, exactly. But that's not necessarily true for every community. So it's true as a state, but each community can look and see really what populations they target. Right. So as you're going around the agency, does this happen all the time where you're like, oh, that's really cool data. I wish I had an extra three days to make a really cool viz, but we've got to move on to the next thing.
00:23:27
Speaker
What sort of like for these external reports? Yeah. I mean, it seems like, you know, there's there's 1300 people or so and you're working with a good chunk of them. They must have like great data. And every once in a while there must be something where you're like, oh, this would make a really cool visualization. But I have to make you this dashboard and go on to the next project.
00:23:45
Speaker
Oh, gosh, that is the story of my life.
Balancing Internal and External Reports
00:23:49
Speaker
I like the internal stuff, but I mean, I think that's why my director, not Letz, because she's Letz isn't quite the right word, but supports me doing these external projects because it's much more interesting because mostly otherwise just making run charts. Right, right, right.
00:24:07
Speaker
It's fun, but it also gets a little old. Well, hopefully the success will keep going upwards and they'll let you hire someone. You can build your own team and you can start to farm out the boring stuff and you can work on the fun stuff. We have a user group, so also just a lot of it's supporting other people to do the fun stuff, which is good.
00:24:28
Speaker
even if I have a little bit of data vis and be sometimes. Yeah, sometimes. I think we all have that. Yeah.
Conclusion and Listener Engagement
00:24:35
Speaker
Great. Cambra, this has been really interesting. It's really exciting to hear how you're changing the data culture there in Colorado. So thanks so much for coming on the show. Yeah, thank you for having me.
00:24:47
Speaker
And thanks everyone for tuning into this week's episode. If you have comments or questions, please do let me know. And please do rate or review the show on your favorite podcast provider. So until next time, this has been the Policy Vis Podcast. Thanks so much for listening.