 Welcome to town meeting television. I'm here today with Nancy Stetson of the Burlington's new BTV data hub. You want to introduce yourself and tell us who you are? Sure. Hi, I'm Nancy Stetson. I am the senior policy and data analyst for the city of Burlington. I work out of the Department of City Planning. What brought you to this work? To the work of the BTV data hub? Yeah, or to the work of being a policy data analyst. Well, I grew up in Vermont and I moved back here after graduate school and started working actually at the police department as their data analyst. And during the pandemic, people moved around the city and I was moved actually into IT and then into planning. So that's sort of my history at the city. Yeah. And what brought you to the work of doing data, I mean it's a particular love, isn't it? Yeah, I mean I've always been interested in government and how government works and but also had the serve background in math and I feel like looking at data often in government data is underused and it brings clarity to a lot of that work. And so that's sort of what brought me to it. So the term data is thrown around a lot to mean a lot of things and I wonder if there's a succinct definition or how you're using the word data when we're talking about the city data? Yeah, that's an interesting question. I'm sure there is one specific definition that I don't have on the tip of my tongue but I would say data is really just information but it's information aggregated up to a point where you can use it to understand things on a broad level. So it's not a singular piece of information, it's multiple pieces of information you can compare towards an end. Yeah, right. So tell us a little bit about the BTV data set. The data, tell me what the name of the project is. Yeah. Yeah, so the project is the BTV Stat data hub. BTV Stat is its own program. It's a performance evaluation program that is a regular meeting of department heads to go over specific data around city operations. So BTV Stat is that program. The data hub is sort of grown out of that program to capture a lot of the data that we are looking at within the city and put it online in an inaccessible way. What we want to be able to do is have people be able to go on the site and just learn things about the city, learn about their community in an easy way. So we'll get a little bit more into the project and look at the website but I'm just curious what's the overall? I hear you about the goal is to allow people to have access but is there sort of a mission behind this? Do you have an underlying driven mission to provide this data? The city strives to be transparent. I would say that transparency is the central mission there of putting out information so people know what's happening at the city. So and how long has this project been in the works? This version of the open data site has been in the works since last fall but prior to that there were other iterations of open data sites. The last version unfortunately was launched in January 2020 and so it never the pandemic you know threw everything up in the air and it never quite got to the point that we wanted it to be at so we took it back to the beginning and sort of restarted last fall. I mean there's at some point somebody decided to start collecting data and information. We have census data, we have other pieces of information data sets and then there was a decision to make those open and I'm wondering if you have any sense of that history of why when did somebody decide it's important to share this and not just hold it in the hands of a few folks. I think there was a sort of wider movement towards open data you know in the early 2010s and before that across across the country and so Burlington was a part of that. Their very first open data portal was in 2014 and so I think it was a wider movement to promote transparency in government and I think it was born out of sort of knowing that the city has so much data and is collecting all that data and making sure it's used as well as possible because it is useful to other people beyond just city operations. Yeah so maybe let's get a little bit into what the data is that's being collected and what the data sets are so we'll just take a look at this now. I think we're at the website. So if we're here at the BTV data hub, how do I start? Well so one goal of this data hub is to make that while there are you know data sets you can get to we want to make sure that there is information that people can get even if they're not data analysts. So we have these five sections four that are currently in the works and these each have their own sort of visuals that you can look at that summarize that data. So we could we could start maybe oh and then yeah so there's the five main sections and then there's a few standalone visuals. So I would start with one of the main sections maybe the housing section. So if you click into housing you can see a bunch of different ways we're trying to collect information and visualize it for people out in the city about what's happening in Burlington. So this first dashboard shows how many units of housing have been built in the city. On the dashboard you can see both the number of units built each year you can also see where those units are built. So all of these dashboards yes they have different filters you can change around you can see what is you can yeah widen it up you can also look on the map you can hover over and see where the projects are and how many units they are so you can see there that's Cambrian Rise with 70 units and you can add in other years as as you'd like so that's some sort of individual project. An individual it's like a single home being built. Yeah possibly an 80 year yeah yeah and if so where is this data coming from? So it depends on what we're looking at the unit data is coming from our permitting system which includes both an ancient permitting system pulling information out of that and and we're trying to keep it current so you know once a month I look to see if there's been new units built and then the dashboard we're looking at now is is pulling in census data from the American Community Survey looking at overall in Burlington some statistics comparing to other cities in the country and if you know again if I'm not a data analyst how do I compare these sets of data I mean are these comparable sets of data new housing units and renters and owners in terms of timescale I mean I guess how do you manage to make sure that the data sets can interact with one another? I would say each of these dashboards are showing sort of their own their their own complete look into a certain into a certain subject so housing units all come from one data set and they're comparable within each other the census data is not directly connected to that data and this is looking at more of you know it's a survey of households so what these dashboards are trying to do in total I think there's six altogether is to paint a picture of what the housing market what housing is like in Burlington and these are the dashboards that you're referring to this public safety or I mean the six there's actually I think six dashboards within the housing section I see I see I see I see the new housing units renters and owners housing costs rental housing quality homelessness and additional data so how do you have any sense yet I know that the site is relatively new do you have any sense yet or are you going to be able to have a sense of who's using this data and for what purposes we haven't taken a deep dive on that yet we do have some like Google analytics like tracking of how many people are using it which I haven't looked at that closely I don't know we don't know who is downloading the data like we don't get so I'm not sure I know with our previous version of the data site there were a lot of like college projects so people would download we download the police data and do sort of like data projects you know like classes from Champlain would use that so I I sort of heard about that through the grapevine but we don't have a great sense of where where the data is used yeah and I didn't mean to shrug my shoulders it's not you know it's not necessarily assumed that when you come to this site that you're not being tracked and and that there's not privacy but what you're saying is you don't track people individually who use this site no no though we would we would love to hear about projects if you did use the data you know it'd be it'd be great to hear about what they're used for I mean I'm kind of curious about so you know part of the the transparency and open data has this you know very valuable side of allowing citizens to make decisions understand how government works journalists students etc and on the other side you have sort of the big data companies who want to mine and collect as much data as they can and now we have a pile of open data sets and I'm curious if there's tension or awareness of that in you know are there any issues with being a being providing all of this data to not just the public but anyone else who wants to use it um I mean we I would say the the main tension is around like privacy data so we would want to be sure that we're not publishing names but most of this data in a lot of ways is actually already out on the internet already like for example the permitting data you can already look up people's who we know who has been pulling permits you know who has been building new houses like you can see that information already this is just aggregating it in one place so it's easier to use we I haven't heard specifically about people aggregating it and sort of in a wider in a bigger way I'm not I'm not sure sort of what they would use that for why why somebody would want to be able to yeah I have seen the crime data sometimes I think sometimes it gets pulled up into sort of wider sites that are using it to track crime um yeah um I mean I guess that brings me to the question of how is who's paying attention to that question of how the data is being collected how the data is being entered is it accurate is it the right data to be shared publicly yeah I would say that the on the collecting side that the data and that we pull comes from a bunch of different places and it's sort of on the department level accuracy is something that I would check and that's something that I would say there's always going to be small issues but that's something we sort of review as the data comes in to get a sense of um does that answer your question sure I mean I think back to the I mean the thing with you know when when folks say well this data is all publicly available already right your phone numbers in the phone book um and so you know why would you care and then you think well except my phone number is in a phone book that's in a particular home in a particular community and it's very different than having that phone number available across geographic lines you know say you're somebody who's um running from somebody in domestic you know you're in a domestic violent situation and you didn't mind you know in the 1980s having your phone number published in your local phone book but it becomes very different when your phone number is published yeah across the country you're you're building up information is in a local planning commission but it becomes different when that information is publicly accessible and searchable again across so you know I think there are there are issues there that I'm curious about if there's a governance structure if there is a way that Burlington is addressing and thinking about what data are we providing how are we providing it and and is it um is it the right data yeah we do have an open data policy that that covers sort of how you should think about what data is published and we and we are sensitive to publishing individually identifiable data um I would say the housing unit data is the one that sort of is on the narrowest um like geographic location but we we're not publishing definitely not publishing phone numbers um and most of it is on sort of a uh bigger lens yeah so if the information so I'm assuming there's something in here about um well I don't know is there how is BED is the data that BED collects around energy usage part of this data set no okay no why why are why not why would they not be included in open data sets um we haven't we haven't gotten there yet I it's it's not automatic to publish some open data you know because it does need to be checked for privacy and accuracy um I personally like I think it would be interesting to look at energy use across the city but it definitely is not something we we would publish online at an address level um anytime I do not see that happening um so it's it's not something we've gotten to I think it could be interesting to see but each data set sort of it takes time to publish and to get the to get it set up so it can be updated regularly great um when you talk about the transparency and the um folks who are able to use this data can you give some other could you give an example maybe of how I mean you mentioned the students but could you give another example of how somebody might be using some of this data for sure I mean the other um dashboard we could look at is the police dashboard I mean that has a ton of information the police department is publishing data every single month so it should be updated all the way through June um and this is this is an area where um I I think it serves two purposes I think it serves a transparency focus it also um it it serves like sort of a a grounding about what is happening at the city um so for example if you were concerned about your car getting stolen you could see where all the cars have gone stolen in Burlington over the this is over the past month but you can change the filter so you can look at the past year um there's other pages of this dashboard where you can look at sort of trends over time and see whether um cars are getting stolen in your area so so so this provides both transparency um but also it can be a useful tool for someone who is concerned about what's happening in their neighborhood um well this is a good example of um you know how data is collected and the the conversations around oh you know you can make data um say whatever you want tell whatever story you want um and I'm wondering if you know we think of data as just these information sets that have no value behind them and yet what you choose to pay attention to right we've we've learned that um and I may or may not be completely accurate here but is there a point that the Burlington Burlington Police Department did start um including racial data with something like traffic stops and then you know to what degree do you rely on that racial data are you asking somebody in a car how would you like to identify when I fell out this traffic ticket uh so um traffic tickets are governed by actually a state law so the way the police department collects racial data is specifically governed by this sort of sort of law the police department did actually start collecting racial data before the state required it um but the the way the state asks police officers to record race at traffic stops is based on the officer's perception of the person's race and that is because the point of that data collection is to understand whether police officers are treating people differently based on how they perceive them to be um so that that's a sort of like narrow case of how that data collection works um but seems like very important right because if somebody says well I perceive that person to be white but the person doesn't think of themselves as white and has a different perception right you have that it is interesting to think of that data set as can be value driven yeah um and yes and that that has come up in the past um again because that data field is the perceived race is specifically trying to measure how officers are viewing um here and the other way you can see that it's up at the top you can look through tabs so there is a bit on traffic stops oh did that go to the wrong one here try there we go um so this is the dashboard that looks at traffic stops so it seems like the map maybe is not working um but you can look at both overall trends over time and then on the top left you can filter for um race or years or why someone was pulled over um so let's just do an example of that so let's say 2016 search select on the search and this is built on GIS this is ArcGIS uh this is not so the website itself is built on ArcGIS this the dashboards are called it's a program called Power BI so all so this is just all the traffic stops in 2016 and um if I were to look at this I go 4,000 white 466 black right 200 2,000 male 1,700 female um and then what what would I do with that information what do you think what how are the different ways that you might somebody might interpret that information I think it depends on what they're looking for I mean one I think one way to look at this information is to look at the trend over time so right now you're just looking at one year but 2016 had some of the most traffic stops that have been done in the city what it says 5,000 6,000 um whereas in the past year there were maybe 500 traffic stops um so you can see sort of what areas the police department are um focusing on and that they don't you know policy has changed so that traffic stops is not a major part of their work at this point well it says here there's oh I see I've included I'm aggregating data by continuing to click um so it's interesting that you come to the conclusion that that the policy has changed maybe policy is the wrong word but their their focus has changed your focus has changed so you may know that because of your understanding relationship to someone who's been working inside the police department but it that would be a huge leap to make to just read those two data sets and say you'd just say oh there was 4,800 and now there's that could be oh my god the people are really behaving well sure there's no wow people are that does not occur really but um have curbed their there's no more traffic violations anymore I mean which is important I mean uh the mayor of New York right um and yeah decided they were gonna have a zero um bike fatalities right and so that's a that's a data that that's a set and the how the how the city wanted to use that we want to have nobody dying on a bike right and so there are a lot of different ways you could achieve that and then and measure it how is the city using this data as far as you know or how are their plans have you heard from other department heads how they might use this to make changes um yes I can answer that I just wanted to go back to the idea of like no traffic violations because the other thing that this holds is how many traffic crashes the police responded to and so if you look um on the trends tab so up at the top again all right this got like zoomed in a little bit I did but so if you click on trends you can you can filter this is all um incidents right now but you can filter out and see that while traffic um enforcement has gone down crashes are relatively stable they haven't they haven't increased a ton um so the way to do this is um on type of call so actually over where you were before there's okay um you if you click select all it'll unselect everything um and then um you can look down and I think it's motor vehicle oh yeah it's already open so you can you can just yeah they'll show all all types um but you can filter into just crashes too so this tells you that these are all here um so you could you could spend some time looking at this and you could say 2016 there were 4800 traffic stops and there were less crashes or there were more crashes or there were right yeah and then in 2022 there are 400 traffic stops and yeah we could figure out did that affect crashes yeah but anyways your question was about how the city uses this data um and the police data is used pretty constantly you know both for traffic crashes there's also um an annual report that comes out once a year looking at racial disparities in policing in burlington so that will be coming out in a couple months um and that uses a lot of this police related data um both for traffic stops but also looking at arrests and use of force um I just have so the other piece of this was I noticed the racial equity data coming soon yes that seems like a big um big question mark especially given the city's cough that um racial um you know that we're in a pandemic in I think in 2021 they they called a public health emergency public health emergency around racial injustice and so talk a little bit about that I there there's not much to talk about yet because I what we um are working on and need to work with the racial the REIB director on who is relatively new is what should go in there and I think that's still an open question of what data should we be collecting what data should be looking at to understand that problem better and that's we we haven't filled out that section yet um but we are planning to so that be that gets down to again that governance the question of like how is this who's deciding where this data is collected and then who's maintaining the data who's making sure that people are filling it in yeah I mean and that might be more of you know probably for the likely for that section we might be using census data which then is already collected and we just have to pull it together but I think the the the question for that section is what should we be highlighting what should we be looking at so there's in in many of these cases obviously like traffic stops or the permitting there are data sets that already exist that are pretty robust that are um quality controlled so to speak and that are used um fascinating um there there are a few um other these featured visuals are sort of ad hoc um dashboards that could also be interesting um the wastewater results a lot of people have seen that's a holdover from the pandemic where we were tracking uh COVID in the wastewater and we are actually still tracking that at two wastewater plants so if you want to know sort of where the levels are you can you can see that there the workforce demographics are just looking at um who's working for the city there's a comparison between 2023 and 2019 um and that gets a bit into demographics across the city yep and then the c-click fix yes data we'll go back through we'll go back to that other page here did I lose my it seems like you could get lost in this for a while yes there's there's plenty to look at so c-click fix um if you don't know is an app you can use to report issues usually with physical infrastructure in the city you know potholes or you know trash on the trash on the green belt that sort of thing and so you can use this this dashboard is just putting a place where you can look through and um find out what the city's been responding to what people have been asking the city to respond to um and how those have trended over time and that could be a way of saying you know we need more code enforcement yeah possibly or if you know if you have a sense that people are putting out more trash on the green belt than ever before you could you could look into that you could um search for that in your in your neighborhood this is another one that's interesting because you know I think I started using c-click fix when it first came out in 2015 yeah but I would guess that that user base has increased since it was first released so it the trends might reflect or increase in users though I I'm not sure I mean I I'm happy to hear that you use it I hope I would love more people to use it because it is used both um you know its primary use is for the city to respond and fix problems of the city um but it's also useful in my job to understand what's happening across the city and using that data to inform policy yeah I mean that would be interesting to know that would be another level of data that maybe is here how many users are on c-click fix and where are they located yeah we don't we don't have a ton of data on that because most users are anonymous on c-click fix um so we don't know exactly who is meaning you could publish you could have multiple you could have multiple yeah I'm not sure we know sort of individual unique users we do know a lot of city employees use this within their work too yeah um so that makes up some of the user base I mean it's interesting just anecdotally it's not like I used it at time but I used to look at it and I noticed that the city's response has um been much more um perfunctory or um and I think you know there's there's could be an exhaustion like the excitement of this cool app and we can interact to like okay all right already I know about the pothole all right yes I do think there are there are plenty of duplicate requests where lots of people respond to the same yeah the same issue yeah so interesting um well I think we're way past time at this point and I really appreciate you coming in there's so many other questions and I think probably a big one is what's the sort of overall cost to the city to keep maintaining data sets like this or to require them or you know but it sounds like um we have enough to talk about in another program okay and see do you find this work interesting and yeah stimulating yeah yeah it's really fun I mean I I love making the dashboards and building these visuals and um it's and like I said before it's great to hear when other people are using it um and I encourage people to go and check it out yeah and so that's we found it easily by just typing in btv data hub into google you can also find it at data dot burlington bt.gov and it'll take you there data dot burlington dot vt.gov we'll probably put that up there for you okay well um thank you all for tuning in and I hope this was helpful and interesting to you and informative um town meeting tv I'm Megan O'Rourke thanks for watching