SmartTranscript of House Judiciary - 2025-02-06 - 9:10 AM

[Chair Martin LaLonde ]: February sixth. And, throughout this morning, we are going to be looking at criminal justice data and, understanding from various, entities, stakeholders that deal with criminal justice data data. And we're gonna start with the, division of racial justice statistics of the Office of Racial Equity. And over to you, Susanna. Thank you for for being here. There we are. [Witness Susana Davis ]: Good morning. Thank you all for having us. [Chair Martin LaLonde ]: Yes. Good morning. [Witness Susana Davis ]: For the record, Susana Davis, executive director of racial equity for the state. I'm joined by two of my colleagues today, Tiffany North Reid. She's the data manager for our division of racial justice statistics, and Anthony Jackson Miller, who is a data analyst in the division. [Chair Martin LaLonde ]: Good morning to both of you. Good morning. [Witness Susana Davis ]: Good morning. I'm going to, just thank the committee in advance for its, understanding. We are triple booked as an office this morning for nine o'clock. So I'm gonna be teeing up this conversation with you all, and then I'm gonna hand it over to Tiffany and Anthony because they understand math better than I do. And then I'm gonna head over to your counterparts in senate judiciary. So I'll I'll I'll have to run through about halfway through this presentation, and I I'll I'll give it over to Tiffany. [Chair Martin LaLonde ]: Right. Thank you. [Witness Susana Davis ]: Okay. So I suppose, I will start us off. I'm gonna thank your committee assistant for being gracious and flexible with us. We did get over some visual materials to you all this morning, and I'll be sharing my screen so that you all can see those. [17 seconds of silence] [Chair Martin LaLonde ]: Alright. So do you do you have our new cohost? Is Yes. Yes. Okay. You are. Okay. There we go. [Witness Susana Davis ]: K. I can see your room, so I know that you can see my screen. So let's are you seeing presenter view? I think you are. No. You're seeing me. Yeah. You're seeing me, unfortunately, for you. Okay. So thank you to all for inviting us to share a little bit about our work and our report. The first thing that I wanna share with you actually is that you all received a report from us, and, we did a lot of work to put that report together. And as with many things, we went back and discovered we're still not fully satisfied with how it came out. So what you all will find in the next day or two is that we will send you another amended version of this report. It's important to us that we get it right, so thank you for your patience. I do wanna highlight one thing. In the in the in the brief time since we originally submitted that report to you all, we discovered and also were made aware of by colleagues the fact that there are real issues with a lot of the data we're looking at. And that because of the different ways that you can collate, collect, and cut up the data, we heard from a number of people that they had questions and even concerns about some of the things that they were seeing in our report. It's important to us that we get things right, that we'd be accurate. It's also important to us that we draw actionable conclusions. We think that we have a lot to go on already. We think that we know a lot about disparities in Vermont and in law enforcement and in other aspects of the justice system that are not law enforcement. And so I think that there's still a lot of really good material in there that we've uncovered and that we've processed, but we wanna spend some more time working with it. So please do look out for, an updated version of that report. [Chair Martin LaLonde ]: Great. Great. [Witness Susana Davis ]: We'll talk a little we'll talk a little bit later today about some of the challenges that we did encounter and the things that we'd like to do to help address them in the future. For now, I want to leave you with the main takeaway. This is just the highlights that if you walk away with nothing else from us today, please do consider the following. First, the data landscape in the state. You already know this. It's the reason for the creation of our division, but that the available data are inconsistent, they're incomplete, and they are quite scattered. Our staff spent an unusual number of hours, over the last couple of weeks just revising and reviewing and learning about all of the intricacies and the disparities in data. And I don't mean disparities in terms of outcomes of the people represented in the data. I mean disparate datasets. Let me adjust my display. There we go. We also know that, there's a lack of a shared understanding on what to report. We've seen this in the datasets. For example, when we look at things like traffic stop data, there are a lot of agencies around the state that still don't have a full understanding of what's required. A lot of data fields that are blank, that we are shocked that they are blank, and inconsistency in the way that people are completing certain fields. What that does is it makes the reviewers work more difficult because it's hard to compare apples to apples when we're also getting pears and oranges in the mix. One takeaway for us that was really critical was we, as a team, are data professionals. And I'm I'm saying we even though I'm not a data professional, but I have a team of data professionals. And if even they are experiencing challenges and confusion and frustration using data that are publicly available, how on earth can we expect the public to be able to meaningfully engage? Right? Are we setting them up for success or for failure from the start? The next category of insights I'd like to share is on interpreting of data. We know that because the data may be incomplete or inconsistent, it ends up making it harder for us to draw definitive conclusions. I know that data people and numbers people like to be really precise with language, don't like to make too much, promising based on data. So we are really cautious, but there are certain things that we know, and there are certain things that we can say with confidence. Right? We know that there are racial disparities in justice outcomes in the state of Vermont. And that's not just true in Vermont. It's true in many other jurisdictions around the country. This is something that we don't have to be shy about. We know that it's true. Really, it tends to boil down to what is the degree to which we are going to see disparity. Not a matter of if, but how bad. The next thing is that we know that sometimes, bias, discrimination can be more difficult to capture in data, and it may not be showing up in traditional statistical collection methods. Now that's for a number of reasons. Sometimes people are willfully manipulating the data, but also sometimes it's just a matter of how data get collected. Remember, we tend to follow a sort of westernist derived, system of doing data science, And sometimes those systems or the the the research methods that are used may not necessarily be incorporating all of the different demographic groups or situations that that are out there. And then, of course, the evergreen problem of statistical significance. One of the things that our office really struggles with for years now is that, looking at data that are disaggregated by race is really difficult because we are often told that the n number is too small or that the, data subset is not of statistical significance, which is mathematically, perhaps correct and understandable, but socially, philosophically, ethically, deeply insulting. Because we're talking about, people in particular, if we're talking about people of color in Vermont, we're talking about tens of thousands of people. Right? To to to argue about statistical significance over our outcomes in the state tells us that when it comes to things like health, safety, education, workplace, that in each of those aspects of life that we are clamoring to exist on charts. Even though we know our experiences, we know what's happening to us, we know what are the push factors pushing us out of Vermont, and what are the difficulties that are keeping it that are making it hard for us to remain. So things like statistical significance is something that is often used as a way to invalidate otherwise really, really salient data, and that's something that we are still grappling with. And then we also have to consider historical context, social context, etcetera. Now what you'll hear from us about future goals include things like data governance council, dashboards, and generally, having insight circles so that we're hearing from more people and more players than the few who you tend to hear from them yourself. As I mentioned earlier, the exercise that we went through with our annual report was a really, really good and stark reminder of the limitations of the material we're working with and the processes that we're working through. [Chair Martin LaLonde ]: So before before you go on, Susanna, could you take a question? [Witness Susana Davis ]: Absolutely. [Chair Martin LaLonde ]: Yeah. Go ahead, Angela. [Member Angela Arsenault ]: Thank you. Thank you, Susanna. I I was wondering just about the statistical significance, what you're bringing up. How is how is that determined? Like, who decides what is statistically significant? [Witness Susana Davis ]: That is a question for a number person. I will ask if you are willing to defer that question to a member of my team. It's I think that the I think that the answer generally is that it depends. It depends on the broader sample size, what percentage that smaller number represents, and whether we can have a degree of confidence in the accuracy of of those data. For example, we could take a sample size of a certain number of people in the state. But because of the way that the demographic breakdown is in the state in terms of, you know, who who lives in Vermont, it might be the case that you have nobody from any other groups represented. You may have only dominant group members represented in that sample size, or you could take a sample size of the same number of people, but let's say you did it in a different geographic region where there is more diversity. Maybe now you have an outsized representation of different groups, and neither of those is gonna be reflective of the broader population. So I think that's one consideration when it comes to who determines what's statistically significant. And I think it also depends on which datasets you're using. For example, I know that when it comes to things like census data, there is a thing that the government now does called differential privacy that I think is it kind of wreaks havoc on population data for jurisdictions that have really small population groupings such as municipalities. Think of places like Vermont, Alaska. I'm not gonna go into the details of differential privacy, but I'm just gonna say it's one out of very many, many things that can sometimes skew the numbers and make it so that we can't necessarily trust them, and therefore, that's where questions of statistical significance come about. But, again, I know you have other data professionals coming in after me who I think can probably give you a better answer to that. [Member Angela Arsenault ]: Yeah. And thank you for that. And maybe my my question truly is in cases of where we cannot say something is statistically significant, how can we be more mindful of its significance? Yeah. [Chair Martin LaLonde ]: You know? [Witness Susana Davis ]: The first thing that I'm gonna say is that when we talk about statistical significance, that is a quantitative data determination. But there is a whole other body of data known as qualitative data. It's powerful. It's valuable. And despite what some people will tell you, it is absolutely critical for us having a complete understanding of an issue. So I would say that in cases where there are questions around the statistical significance of something that we're seeing, we should absolutely be coupling that with qualitative data to help us build a complete picture. We may only have two people in the state who've experienced a particular outcome, but if it's an awful outcome and if it's true and it's happening, we can still act. That's one piece of it, is look at qualitative data. The second piece is, we can also look to other jurisdictions that are similar to Vermont who may have more robust or more, thorough datasets. So I know that Vermont is a charming and very, very special and unique place, but there are a lot of ways in which we're very similar to other parts of rural white America. And so I think being able to borrow what we've seen from other jurisdictions also helps give us insight into some of the things that we're seeing when we can't necessarily trust the numbers or rely heavily on the numbers we're looking at. And then I think the third thing is, not just relying on your advisers, entities like the RDAP and others, but also just looking at the history in Ramallah. Right? So for example, we may squabble over the degree to which there are certain disparities, but if we know that year over year over year there are disparities, we can act. So I think that taking all of those things into effect, historical trends, similarly situated jurisdictions, and also qualitative data, those are some of the strategies that we can use to overcome the hurdle of statistical significance. [Chair Martin LaLonde ]: Thank you so much. Thank you. [Witness Susana Davis ]: Thank you. I'll spend a couple of minutes just talking a little bit about the division. And in in specific, what I wanna do is just remind the committee about what are some of the things, the the the tasks and deliverables that you all are expecting from us. I'll do this kind of quickly. Oh, just kidding. First, I wanna talk about the mission and the purpose. So both of these, of course, come from the statute. So this is the mission and purpose that you all have bestowed upon us. Of course, our mission, according to act one forty two of twenty twenty two, which, of course, puts the governing statute of three VSA section fifty, fourteen ish, fifteen, sixteen, is to collect and analyze data relating to racial disparities with the intent to center racial equity through those efforts. Of course, the purpose of the division more explicitly is to create, promote, and advance a system and structure that provides access to appropriate data and information and that ensures privacy interests are protected and principles of transparency and accountability are clearly expressed. This is really important, because it's not just about the nuts and bolts work that we're doing, right, that's creating, promoting, and advancing a system, but also includes certain ethical and philosophical, markers in here. Right? Transparency, accountability, privacy. These are super important. And we're seeing those not just reflected in the division's work, but also in, the ways in which we wanna influence other work happening around the state. For example, when we think about privacy, transparency, and accountability, one of the first things that comes to my mind is our AI advisory council. It's one of many examples, but there's a lot of intersection with this work and with, what the division is asked to do. We have, as I mentioned to you before, a manager. We also have one analyst, one vacant analyst position that's been talked to recruit for, and then one temporary analyst position. This is for a grant funded project that we're a part of. So we've been extremely grateful to have that second analyst on board. That's Anthony who is on the call. And, also, we are planning for a near future when we no longer will have, will have that position, particularly because that grant has a focus on equity, and it is federally funded. And we know that right now, the shenanigans coming out of DC are making it really difficult for us to be confident that federally funded work with an equity lens is going to continue to be funded. So those are some of the uncertainties swirling around our staffing situation. What we do as a division, and, of course, all of this is pulled from statute, we work collaboratively with state and local law enforcement agencies to collect data. We collect and analyze data, and then we conduct, info sharing and gap analyses. Maintaining an inventory of technology assets, maintaining a data dictionary, developing a strategic plan for justice technology, developing agreements between agencies and memorandum of understanding, reporting monthly, of course, to two different work groups, the RDAF and the Racial Justice Statistics Advisory Council, reporting annually to you all, check, establishing, maintaining, and implementing a program for managing records, analyzing data to identify different stages of the justice system where you'll see racial bias, organizing those data to present it, and then of course presenting it to our advisory council. Developing and adopting data governance policy, which includes establishing systems for standardizing data collection and methods to make the data sharing easier. Recommending evidence based practices and standards for others, developing public use data files, establishing which data are to be collected, and which agencies, state and local, have those data. Later on in the presentation, Tiffany will tell you a little bit about some of the other statutory authorities, that are granted in our in our enabling, act that will also have to do with how we're going to go about, establishing those standards and systems. Okay. I'll stop here and hand it over to Tiffany. I'll give you all my thanks. I am getting a note that I am almost up in in senate judiciary, so I do apologize for needing to leave. But, again, I thank you for for hearing from us, and I look forward to more discussion. [Chair Martin LaLonde ]: Are there any questions for me before I go? Any questions? I think we'll save them for our folks. Thank you so much for for being here. [Witness Susana Davis ]: Okay. Thank you. I'll drop a screen share and invite my colleague, Tiffany, to resume it so that she can continue sharing with you. Thank you all. [Witness Tiffany North Reid ]: Thank you. Thank you, Susan. Okay. Let me start sharing my screen. And I just want to thank you for allowing us to be here to speak about, the work that we're doing, and addressing the report as well. I think it's, it's a, it will be a good conversation for us. Let me share. I'm new at this, so give me a moment. [Member Angela Arsenault ]: This is typically the most stressful part of the entire [Witness Tiffany North Reid ]: Certainly. And I can understand why. It is. I've heard great things about your group though, and I know you're doing great work, so [Chair Martin LaLonde ]: I appreciate you. We [Witness Tiffany North Reid ]: do appreciate you. We have, let me, you know, and certainly I would start at the very end. That's okay. I feel like Susanna already did a really wonderful job at setting the context that we're working in, some of the challenges. And I also just wanted to make sure we have level setting. I know we often throw around terms and language that, you know, can mean different things to different people. And I think as we're approaching this work, yes, we have the statistical language there in the background, but we're also grappling with a really, really tough topics that sometimes is hard to capture in a way that's meaningful. And with that said, one of the one of the things is we're we kinda move into talking more about the report and more of our work. You know, we we've been having the, we've been having conversations and many conversations around statistics, around history. One of the terms that we think about often is disproportionate impact. How that's being defined can differ depending on who you talk to clearly. You know, and I think this provides some kind of bounds around how we approach some of the things that we are looking at or trying to research. So basically, and you can read along with me, just it occurs when groups are affected by a policy, a practice or a system, at a significantly high rate. And that may go back to, one of the earlier questions. And how do we define that? And I won't go too far left around statistical significance, which is really important. We'll kind of get back there, but is affected at a significantly higher rate than other groups, often in a way that's not aligned with population. And so this is some of the challenge that we come up against is how is this or not reflected in the data? Are we collecting the right data? Are we asking the right questions? And that's where a lot of the gaps can happen because, you know, we may be looking at a dataset, but maybe there's a variable that wasn't collected that actually paints more of the picture, of the experience in Vermont. And so that's what we're trying to get to a place of. Are we do we have the right data? Are we collecting the right data? Are we able to analyze it? Are we able to analyze it at a high level, not just descriptive statistics, but are we able to get into your predictive statistics or, being able to predict or, analyze relationships in a at a at a higher level. And that's what I'm hoping we we can get to. I think the statistical significance conversation is a really good one. And I think we have a lot of work to do in that area. We we have, we have some things kind of coming along that where we wanna have these conversations, not only among agencies, but also with communities, and just get to thinking about some things that we might not often think about. And I will say that in academic circles, statistical significance is a really big conversation. And I know that from when I, you know, did my training so many years ago, that was the gold standard. There are now talks in epidemiology and biostatistics about what are we missing if we are limiting ourselves to looking at this data in such a way, or we're trying to reach this threshold, what might we be missing? And that kind of goes to, that's another conversation, but often what you see in journal articles, we're only publishing things that are statistically significant. So we might miss out on that other, you know, those other couple of articles that just don't make it there because they don't have that, you know, that that p value of zero point zero five or zero point zero one, but we might be missing something in the process. You know, so, you know, just keeping that in mind. And, you know, we are operating and understanding we I will put a caveat here. We're operating in a space where we have a backdrop. We have a reality that there have been findings of disproportionate impact for certain populations. Nationally, yes. But and also in Vermont, that suggests disproportionate impact. Let me back up. Across different metrics. And so I have that in mind. I also have the caveat that that doesn't, I don't think that should define our trajectory and how we approach these questions, because I would like to, I would like to believe that we've made progress and I really want to figure out how do we track that in a better way because the reality is that, yes, we can identify these, you know, disproportionate impact or where there are issues, but maybe we might also be making advances, but we're not able to capture that in the data that we have. And that's a big problem. Let me go to the next one. Annual report. Here we go. So we're gonna we're gonna be we're gonna keep this at a very high level. But please, if you have questions, we're happy to answer. As Susanna mentioned, we will be sending you with some additional information and, you know, between today and tomorrow because I I think there is the right word is I think we need to be able to convey our experience, but also we we want to make sure that we in presenting information results, we have to put that into context is very important. And I think also, as much as we say we present things as being tentative and we couch it and say, you know, we we only, you know, we looked at this because we couldn't look at this, It can still come across, to stakeholders and people in the community or people we work with in a way where it can be interpreted maybe in a way that we would not want it to intend, especially because there's a lot of great work underway for people who are addressing issues of equity. And so we want to be mindful of that as we present and also making sure that we're presenting on the best data that we have, and that reflects us, I think reflects better, you know, the reality of what's happening in Vermont. And so that's what we're striving for. So the annual report, so I wanna take a step back. So we, you know, we did, we looked at a few datasets, and I think that's important for a few reasons and we'll, I'll kind of get there. So the main thing so what we're trying to do and I think my vision and, you know, I've looked at a few models around the nation. I don't think anyone's getting it right, completely right. But there are some examples of things we could be doing. I think we're, we're ahead of the curve in so many areas and I think we can continue to, you know, push limits in terms of, how well we collect data, what we're able to do with our data in terms of policy, in terms of interventions. There's so much great work that can happen. So my mind is like data infrastructure, data infrastructure, because in order for us to fix challenges or to do the kind of research we want to do, we really need some [Member Karen Dolan ]: we [Witness Tiffany North Reid ]: need some advancements there. We need some, and I think you all have probably had a thousand too numerous too many conversations about that. But I just want to say that that's the larger picture we're trying to get there. And we have a few things in mind and solutions that we're trying to implement that we hope will land and be helpful to, Vermont and to our partners that we've worked with who've already been in this space. Transparency also will be important. I think not only for us and that's why, you know, we're doing things in such a way, we want to do things in such a way where we have transparency. I think, operating within this moment is, also an opportunity, I think, for self reflection because I think as we run these analyses and we do this work, we need to make sure that even if we look at things differently, how we might analyze something, that we do have the transparent understand how we got to there. And that's what and that takes a conversation that we need to have. And so that's what we're working toward. Please speed me up if I need to. I can be a little bit long winded, so I don't I want to try not to, offend anybody. [Chair Martin LaLonde ]: It's been fine. Thank you. [Member Coach Kevin Christie ]: Thank you so much. [Witness Tiffany North Reid ]: We want to collaborate with agencies. We have some things coming. We're preparing a data governance council and we hope to have representation from those key agencies who are holding this data because they know they know the issues with their data. They know, what can be shared. They know best practice. We want to tap into that. We're new. We're new. We're still in development. So we still have a lot of people to talk to, but that is the goal as we're preparing the State of Governance Council in collaboration with ADS is that we need people in the room talking numbers and trying to figure out how do we get this right? And just sharing information because we're often working in silos and we realize that. So collaborating with agencies. Yes. Did, was there a question? [Chair Martin LaLonde ]: No, I think we're good. [Witness Tiffany North Reid ]: Oh, I'm sorry. Okay. Collaborating with agencies, you know, to strengthen the data infrastructure. I also am very personally passionate about elevating the the topics around Lifecourse and data equity, Lifecourse in particular, because I I want us to be looking upstream and I know Susanna talks about this significantly. You know, if we're looking at someone who's eighteen or twenty, you know, that person was a child at some point in time. So how do we how do we get into the space of thinking about prevention? And maybe we don't want to be able to statistic with a statistic. Let me say this, statistically significant. We shouldn't necessarily be able to predict. And there are some, there are some, and I'm not Vermont, but there are some studies of places where you literally can predict the amount of, number of eight year olds who, or the percentage who might be incarcerated, right, ten years later. I don't think we should If we can do that, I think we can, we can probably, do something different in terms of, you know, making change there or figuring out what are those, what are those, risk factors? What happens along the way? So from infancy, you know, we're thinking about issues of parental incarceration. How does that, have collateral consequences, right, for the children involved, for the families involved? When we think about then maybe next stages, maybe we think about school environments or we think about what happens, you know, for at risk or marginalized youth. And then we think about, you know, not saying that everything is a complete line and it's completely connected, like one has to be in front of the other, but there's kind of like these really critical points that we need to be thinking about across the life course, right? Also data equity, right? Data equity and we have some trainings coming up which we're really, really excited and eager to launch, because it's not something that has often been talked about in the past. We've had a few workshops. I don't know if any of you have been able to attend where we've talked about data equity, but it's it really reframes how we think about data. And in a way that when I learned data, it was, you know, we didn't talk about data equity. We talked about how to collect data. We talked about the data life cycle, etcetera, etcetera. But how do you think about who gets to collect the data, who gets to look at the data, right? Who gets to select what kinds of data points are important? And then interpreting the data. So, I mean, even just if we overlay the data lifecycle from collection all the way to analysis to reporting, there's someone or people operating along that circle, but who's not included in that in that conversation. Right? Or who might be prioritized in that conversation? Who's missing from the table is, you know, I think it gets us to thinking a little bit a bit more about what data equity is and why collect data? What are we doing? And I think there are so many people in state government right now who are working their tails off. And maybe they don't want to add another data metric. I get it. But I think if we start to have those conversations and people understand why this is important, I think it maybe makes people feel a different way. And maybe we do this in a different way where it's not necessarily a burden, but it's actually something that benefits, Vermont. So that's that. Let me speed up a little bit because I know I'm so I think in the process, we want to be developing dashboards. There are already some really great ones there, that we point to other work that's already been done. So we don't want to duplicate the work. We want to be able to on our website in the future, we want to elevate. Like we want to point to CRG. We want to point to, the different, you know, dashboards and resources that we have available. We're not trying to recreate the will. And we you know, I often feel like, the work that we're doing is very compatible with other folks who are in this space and just figuring out how do we elevate that work and then also innovate, I think, in our own space and kind of coming from ORE, we have a very unique perspective and we're just trying to figure out how to get that right. And so we have been talking with our data collaborators and like, how do we get the dashboards right? How do we get them up? What's important to talk about? And I would say also elevating that life course, that kind of continuum, to keep it in people's minds and maybe other people get ideas about what needs to happen based upon what we can present visually. [Chair Martin LaLonde ]: Before we have a question here. I've learned from here. Yes. [Witness Tiffany North Reid ]: Thank you. [Member Karen Dolan ]: Hi, Tiffany. Thanks for sharing. Just on this piece on collaboration, because I'm definitely hearing the piece that a robust system of data collection is key to this. And you're saying how there are some systems that already exist. I'm wondering if you can speak to what are some of those agencies that you're collaborating with, and if you can just speak to those collaborations or are things working? Do we need to help with any of it? Like, how how is that working? [Witness Tiffany North Reid ]: We that's a great question, And we have a lot of information to share. And I'm I'm I have to I have to speed up my because I I get kind of passionate about these conversations and I think I'm taking a little bit too long to get to the point. But you make a you this is a great question. So right now, so we I want to highlight first, the grant that we have from Building Bright Futures to focus on issues of early childhood, which I think has made us, you know, stretch in different ways as we're kind of connecting the dots. That's a that's been a wonderful collaboration because we also get to see another topic that also kind of intersects with us and we get to share information and we get to think about not only the wonderful things that are happening around early childhood, but, you know, how do we have a future forward thinking? Like if we have children who may be at an early age, how do we, [Chair Martin LaLonde ]: how [Witness Tiffany North Reid ]: do we redirect? How do we make sure that we get those children on the right path? So it's making, I think we're having a conversation and I'm looking forward to seeing where that work goes. And I like I said, to echo Susanna, I'm very thankful that we do have representation in that space. We have a funded data analyst who's on the meeting as well, who is doing a lot of work. And I won't go into too much detail in case he would want to talk about some of that. Building Bright Futures, preschool development grant. I would also say that when we're thinking about the juvenile justice space and the criminal justice, we are building those relationships. I think I have been very, I have wanted to be very cautious, and that's probably why you don't see me very often, because we have a lot of, I think behind the scenes work that's happening. So the major partnership I think that you all facilitate facilitated was is our our our work with ADS. So to build that data lake. The the you'll see more of that work happening. It's often my frustration and my, what I try to move through is that because when we're working in a data space, there's often this thing that's happening often, you know, that no one gets to see for a while. And I have to work through that. I have to work through that because I think good work is coming visually. So the work with ADS is awesome. And you will see things, through that collaboration that will be happening, in collaboration with the Data Governance Council that we're bringing together. And so I'm kind of talking it a little bit around cause it's it'll probably make sense. We have been talking with data collaborators, in the agency of education. We have been talking with folks, at DOC. We have been gathering data from different points. I don't want to call it a collaboration yet because I think collaboration is a very strong word that I very much respect. And so we're trying to get to a point of collaboration where we have people at the table. We are literally like continuously talking with each other. We're planning, we're supporting each other. And that's the space we're trying to create. With the Data Governance Council, we intend to have representation from criminal legal folks. We intend to have representation from the early childhood folks. We will have, of course, representation from ADS. So we have the data folks in the room. We have the IT folks in the room to figure out what's the problem with the data and the platforms and all that good stuff. But I see those as our key relationships. And I think, and of course, you know, of course, CRG, is someone that we respect in this space, who we want to be a part of, what we're trying to do, and who can also guide us in ways that we may not even know at this point in time. I know we're missing some. I'm going to keep going because I know I can be a little bit long winded, but does that give you a sense of we have a lot of work to do, but I think we also have a lot of collaborators and partners. It just feels kind of like we're, you know, it's, it I feel like it needs to be consolidated. Like, we need to get all the people in the room and not necessarily we're having siloed, kind of conversations or things we're working on together. And it's kind of like over here and over here, but really bringing things together. [Member Karen Dolan ]: No. That's helpful. Thank you. It just gives context of where you're at on this journey of collecting data, presenting that picture. [Witness Tiffany North Reid ]: Yes, we're early. We're early. We have a lot of work to do. We have a lot of work to do, but I'm very hopeful that we'll get more done this year, especially as we have, I don't know what my computer did something. Let me move on because I know I'm being, probably taking maybe too much time. We, you know, I think this this kind of says some of the same stuff. We're trying to get to ensuring data integrity and just continually engaging stakeholders from agencies to communities to the public and basically toward addressing disproportionate impact. I will jump into the report and we just have about maybe five slides left and I'm going to move a bit quickly. I think I saw I'm sorry, did I see Laura? No. Okay. Well, so I'm going to, so our initial suggest so initial assessments and you can define it how you want to. Like I said, this is a very, we've relied upon the data that we've seen before. And certainly we have references that show disproportionate impact. We have been able to and we continue to work through the data sets that are publicly available and I and there's a reason for that. And I just want to kind of share my thoughts on that. Initial assessments, very, very high level. Yes. We compare to when we think about that definition, you know, the demographics of Vermont compared to the numbers that we see for some of these metrics, is really important to think about. Now whether we're going to have the data or the statistical significance remains to be seen, And that's what we're working toward. And I will say with that, that is part of the reason because we we are, we do have concerns and our partners do have concerns about the publicly available data. We do need to amend that report for you all. But at the same time, I think it's very imperative that we continue to work with those datasets over the next few weeks. And I will I'm committed to improving those datasets in collaboration with the data holders, which are of course VCJC, DOC, AOE. And I think also DPS should certainly be a part of the conversation because they bring great information. So we're committed to doing that. I think we also want to have the backroom look like what's behind that dataset, like what's the actual data and that's in process. I also want to be say, I want to say that we have been hesitant to, to do that until it's right. And I, my, my, my, my general kind of initial, thoughts on that is that it should happen under the Data Governance Council. Because I want us to be looking at this data in a way that, it is protected. We have all these bounds in place, so we don't have issues of, you know, revealing anyone's, you know, personal data. We want to make sure we're doing it right and we have all the right people in the room when we go to pull that dataset from DOC. That's the identity. We want to be doing it right. And I also want to say that's why the public why I feel that the public datasets are so important for the work we're trying to do. So I don't want to brush those off. Right? I feel like it's a first step. If people if someone in, you know, agency of education or someone in DOC wants to kind of get a sense of what kind of data, you know, let's say DPS has, right? It does help to have publicly available datasets to just think about what kind of data is available, like what kind of questions might we ask. And maybe they don't want to go through an MOU process and actually get the actual data, but maybe just get a sense of the data or maybe just run a few numbers. I would like to see that improved because I think if we can, find a process and we'll talk more about that, if we can figure out how to get it right, and also making sure that we are protecting folks' privacy. I think it helps us as we move toward improving the data infrastructure, but also data integration. How do we follow someone across from, let's say, an arrest to corrections to release, right? We don't necessarily, we're not necessarily positioned to do that. And that's why I say we are trying to get to a place where we can do that and so we can track where inequities may be happening, but also let's track progress because we may be making progress and we just don't know how to capture that for our state. I will also say, and Susanna has talked about all this. Oh gosh, let me hurry up. Yeah. So I just want to say that I think the publicly we still want to continue to work with those datasets even as we're looking at the other data. And I'm very thankful that our data partners are working with us and making that our data partners have are working with us and, making that available for us to have a deeper dive and really see what they see. And then also understand how do we, translate or convey that over some of that, into the public data set. So there is some consistency. Okay. Limitations and caveats. I'm going to quickly say, you know, we've kind of talked about this, you know, we want to address and help and support issues around like disparate data files, inconsistent headings, missing suppressed data, you know, data glossaries, personnel issues because we really need data capacity billing. We need staff data capacity billing around the state, not only for us. I think, you know, Susanna mentioned that, you know, we, you know, we have staff that we have and we're very grateful, but we also know that we also need more capacity. There are so many things that we're being tied into and we don't have enough people. But in a good way, that's a great thing because that means that we're needed. So, but we just have to get it right. Recommendations. So these are our last two slides. [Chair Martin LaLonde ]: So before before you go ahead Absolutely. On on that, there's certainly places where the legislature can help and certainly making state statutes less vague is one of them. And and I'm not asking you to point those out right now, but but certainly would like to follow-up a little bit on that, if if that's an area that we can be of help in addition to, advocating for you folks getting the right resources. But just for future, I just wanna note that for you. [Witness Tiffany North Reid ]: That we certainly need your your support and your help in that area. Susanna has a lot to say about it and one of our other colleagues who's not here right now. But we even though we are on the data side of it, we're we we you know, we try not to dabble too much in the policy side of it, but there are statues around data that we we probably can we can get a lot of work done. [Chair Martin LaLonde ]: So Good. Alright. Well, we'll look forward to that conversation. [Member Coach Kevin Christie ]: Thank you so much. [Witness Tiffany North Reid ]: Oh, what did okay. My screen. Okay. Sorry. I'm going to make this quick in case because I know there may be questions and I don't know, my colleague may have comments too. So I'm just going to, run through this. These are some of the things that we have the ability to do, in terms of improving data governance, that we're committed to. And I think this exercise has been, so important. I'm often in the data numbers world and wanting to be in the numbers world and being frustrated that we sometimes don't have the data that we need. And so this feels like a bit of a solution that, I know Susanna, was planning to do some work around and especially now understanding having dealt with some of the, the public datasets, which I still think are important because they kind of can tell us what's happening with the real datasets. And it's not necessarily anyone's fault. But that's just the nature of administrative data. We're not it's not necessarily being collected for the purposes of research and we understand that. So, you know, how do we how do we understand that but also improve what we have at hand so we can do those really nice analyses that are that are, will support the work. I won't go into detail here, so I know we're going to have a follow-up around, statutes and, some of the things that we hope to do. And so I'll leave that for another conversation. And these are just these are just the things that the additional things that, we're working on that we touched on in our report and in this conversation. So how do we enhance data transparency, accessibility, and quality, in collaboration with, partner agencies like DPS, like DOC, like AOE, developing those dashboards, which will take some time because I think we need to get it right, in order to to foster that that trust. And then also, how do we support the data staff capacity building in the state, also with us and collaboration across the agencies. And we've we've already, you know, we're already starting those conversations. We are working with VCJC to look at some of their data, also understanding that we have limitations because it will be a lot of data to look at. And there will be issues of data quality and things that take tremendous amounts of time and staffing. So we get that we but we want to be supportive and we have already started that collaboration. The other thing is strengthening the data governance and integration. So we just wanna make sure our our datasets is complete as they can be. So when we actually go to do our data linkage with ADS and our data lakes and all that good stuff, we actually have data and we don't have these missing, you know, holds and we have to eliminate this data because we don't it doesn't match up. So I think with the goal of data integration in mind, we have to make sure that these individual datasets, first of all, are, we have to do an assessment and that's the work we want to do through the Data Governance Council. We understand that sometimes there may be data we need that might be on paper still. So how do we make that assessment? Like how do we help agencies? Like, how do we make this a more streamlined process so that if, you know, an agency needs to ask for information, that it is not so much of a lift for our partners because they're already burdened, I think, in working so hard in other areas. So that's what we're trying to do. This is our last slide and I'm going to stop there and I'm going to, also ask, you know, if yes. I see there's a question. [Chair Martin LaLonde ]: So no. I don't think the unless there's a question from somebody online that I'm not seeing. If you could end the share screen perhaps, then we can make sure that I'm not missing somebody online's coach. So go ahead, coach. Unmute yourself if you have a question or a comment. There you go. You're still muted, coach. Let me we there we go. There we go. [ ]: Sorry about that. [Chair Martin LaLonde ]: Yeah. No. No. Go ahead. [ ]: I I want to say, for the new members of the committee and any folks watching, this is a very special moment personally for me, hearing this report, being our first report, from ORE and this entity. And the thank you, and this is wholeheartedly, goes to our vice chair, our clerk, and our chair for sticking with it. Because if it wasn't for them, this entity would not exist and our committee. And and I sincerely mean that, and I just had to say that for the record. So thank you guys, and our committee because this was a dream of this committee, and I just needed to put that on the record. And thank you all again. [Chair Martin LaLonde ]: Thanks, coach. So question for for you, Tiffany. So the a specific question and then kind of a broader question. The data governance council establishing that, is that something that you're able to do without us having further legislation, or is that going to be something necessary legislation? And if you don't know offhand, if you could pass that question on to Suzanna. [Witness Tiffany North Reid ]: I feel like I know that. I I will pass it to Suzanne just because there could be nuances that are not clear to me. I I think we have the go ahead. We are we are framing this under our work with ADS who is we've, submitted, I don't the ADC project business case, for the work and the it it includes the data governance council. So I think we're inbounds, but I will ask I will ask Susanna to follow-up on that. [Chair Martin LaLonde ]: Yeah. And and the bigger question, I guess, is, there's a lot of stuff that has to be done within your office and the administration. But want to understand if if there are any other things that you need from the legislature in statutes and and and if if there are needs, if there are some if there are priority needs. I assume I probably would have heard of those already because I have reached out to Susanna before the session. But I just wanna throw that out there again and and not an answer that you have to give me right away, but just something that I want you all to definitely think about to make sure that while we are still in session, if there's anything else that we need to look at legislatively. So, other questions in the yeah. Ian, go ahead. [Member Ian Goodnow ]: I think it's to you, mister chair. Just trying to procedurally understand. Is this hearing we're having right now to be in compliance with three b s a five zero five zero twelve b? [Chair Martin LaLonde ]: You have to tell me what three b at one hundred was. [Member Ian Goodnow ]: I'm sorry. Yeah. This, looking at the statute here, on or before January fifteenth, annual and annually thereafter, the division shall report data analysis and recommendations to the house committee judiciary. [Chair Martin LaLonde ]: Oh, thank you. Perfect. To see that. It's not really I mean, it's it's to kind of go over it was initially to go over the much broader, actually. Yeah. We wanted because we have other people talking about other parts of data collection of criminal justice system, including CRG, which did something according to another statute, Act forty. And I did wanna hear a little bit from the court, as well, and and possibly the Department of State's attorneys just on some particular questions of data collection because also RDAP has brought up some data collection issues. So it's kind of broader, but also this was an opportunity to to go over that report. But that report is still in process because of some issues that have come up, with that. [Member Ian Goodnow ]: And I I only ask because I think, the concerns that have been referenced about the report, I'm not really familiar with them. And so I don't really wanna ask a bunch of questions about it if that's not what's relevant. [Chair Martin LaLonde ]: Yeah. I'm gonna just point because I think I think we'll we'll probably, again, have folks in once once the the final report is available, and and we'll go over more depth in the report. And and that doesn't have to happen before crossover either, you know, but sometime this session will will go back to that and and and look at the the outcomes. Yeah. We'll look at what that report says. So but good questions. So and I don't know if I really gave you a very good answer. But so so, Tiffany, so do do you have more on that? And I'm sorry. Did did were we gonna hear from and and I can't recall what the a was for, a Jackson. I know that you've said before, and I apologize. Anthony. Anthony. Thank you. Anthony Jackson, I could probably hear that. Did did you wanna weigh in on this as well? Did you have some points to make as well? Thank you [Witness Anthony Jackson Miller ]: for having us. Sorry. Am I unmuted? [Chair Martin LaLonde ]: Yeah. [Witness Anthony Jackson Miller ]: I think one of the most pressing things for our department is just looking at which statutes govern how certain data's, collected. So we can make it standardized across the state, which may make it easier collecting it when we go and talk to, agency in education, when we talk to DOC, when we talk to the Vermont, VCJC. We just find that sometimes data is collected in different ways because the statutes don't explicitly explain a b c how it's supposed to be done. So when it's collected, we then get it. We have to figure out what's the best way to put it together. And if the data doesn't line up, it causes confusions when we try to take what looks like quantitative data qualitative data and turn it into quantitative data, which will create problems and try to generate reports, which we don't wanna have happen. [Chair Martin LaLonde ]: Right. So so are we at some point gonna have an idea of what those possible fixes are for for those statutes that get more consistency? [Witness Tiffany North Reid ]: Yes. Yes. Susanna, I I think she mentioned, ten AM tomorrow, that we would have those changes back to you. We wanna just have a quick turnaround so you have what you need. [Chair Martin LaLonde ]: Okay. So soon, possibly. But I'm sure that once we identify that, I'm sure there's still a lot of work, I'm sure, that has to be done to try to do that. But that's definitely my understanding when we created the division. That's one thing that we wanted to understand is how to get some consistency in the in the data. And if that's gonna require additional statutory changes, we'll we'll get to work on that. So we'll look forward to that. [Member Coach Kevin Christie ]: We appreciate that. Thank you so much. [Chair Martin LaLonde ]: Alright. Any other questions or any other points that, either Tiffany or Anthony Jackson wants to make? Not not seeing any. I mean, why why don't we take a ten minute break right now before we jump to the CRG folks? And that'll allow you to get yourself

Play Clip Play Full Video Pause

SmartTranscript of House Judiciary - 2025-02-06 - 9:10 AM

This is useful for seeing what happened at the meeting without wading through the whole transcript: