 Good afternoon, everyone. Again, this is Rachel Perry, and I am the founder and owner of Strategic Data Analytics. I've been performing data analysis work for close to 20 years, seven of which have been for legal aid organizations. So I created the content for and will therefore be introducing you to the data analysis framework, along with Scott Fry, who is our web designer. Just a little bit of background first. The data analysis framework was an idea that was generated during a 2013 to 2015 TIG project. That project was a partnership with the Legal Aid Society of Cleveland Montana Legal Services Association Strategic Data Analytics, the Northeast Ohio Data Collaborative, a local think tank, and Cleveland State University. During the project, we worked on significant analyses of different vulnerable populations in the greater Cleveland area and then also in the state of Montana. Our analyses revealed interesting patterns, some of which were expected. Others piqued our interest, raised questions, and generated discussion regarding what the data revealed and how that new information could be used to improve effectiveness and efficiency. The analyses were fruitful for those two legal aids, but we wanted to make data analysis more accessible to a broader audience. So we developed an outline for the data analysis framework. And then starting last year under a new TIG project that is a partnership between Northwest Justice, again, Strategic Data Analytics, and Scott Fryday Design. We've actually created that framework, the site, and all of its contents. So Scott Fryday is going to give us just a few minutes about the technical details, and then we'll launch into the content. So go ahead, Scott. Thanks, Rachel. So yeah, I'm Scott Fryday. I'm the developer on the project. So I just wanted to quickly touch on some of the technical aspects of the site. So we are using a Drupal aid back in for the site. In the last couple months, Drupal 8 has seen some updates that have smoothed out some of the rough edges that it first had when it first launched. So now at this point, I think organizations might be thinking about when they're upgrading or creating new sites using Drupal 8. And one of the new features that Drupal 8 has is something called REST API. And basically what this allows you to do is turn Drupal into a strictly content management system only so a separate interface can be used to display the content, and then Drupal is used to actually deliver the content to that interface. So this is something that is a little different than a normal Drupal setup, but it doesn't really require that much different techniques for setting it up. So it's something that I have found a lot of benefits for. So my contact information will be available at the end of the webinar. And if you are curious about doing this with your Drupal 8 site, definitely feel free to give me an email. And I will be happy to provide some tips and advice. So that's about it. I didn't want to get too technical in this. So I'll turn back to Rachel. All right, great. Thanks, Scott. So we're going to go ahead and jump right into the site. And so here we are on the home page. And for those of you who want to look at the site while we're walking through it, it's at daf.lsncap.org. So a little bit of basic information about the framework here. It provides analysis instructions to help organizations answer five high-level questions and or 118 more detailed data sub-questions about eligible people, their legal needs, and the services provided. It's simple to use so that even organizations that are just beginning to look at their data will find it helpful. But it also offers a level of detail that will make it helpful to more data-savvy organizations that are interested in more sophisticated analyses. The instructions include guidance on what internal and or external data are required in order to be able to answer the data questions and undertake the analyses. And every example shown uses actual legal aid data. So if we scroll down on the home page, there's a section that I want to call your attention to called Things to Think About. And these are several things that any organization undertaking any data analysis should think about. The topics covered here are watching for data patterns, running every finding by staff, just scroll down, dealing with difficult data, data integrity, and factors that can skew your data. Each of these sections describes the issue. So an example of a real-world example of a legal aid facing that issue and tips for how to deal with the issue. So we'll just take a look at one in particular, or one section in particular. So here in the dealing with difficult data section, there are some examples at the top of uncollected data. So sometimes you do not or cannot collect data about something that is important to your client. The two examples there are information about formerly incarcerated persons and then disabled persons. And in these examples, they describe how some legal aids were facing those issues and were able to use either external data, so outside of their case management system, or proxy internal data to analyze those conditions. So I'm not going to go through all of them, but I would encourage anyone thinking about doing analysis to look at these things to think about and figure out how it might apply to your particular organization and the analyses that you're trying to conduct. So when you're ready to actually start looking at the analysis examples, you can get there in a couple different ways. You could click on Begin Exploring, and you can always, no matter where you are on the site, click on the Data Analysis Framework link up here. So I'll go ahead and do that. And this is really the heart of the site, sort of the center of it that will get you everywhere you need to be. It has four sections, the top section being the data questions, the next being the analysis types, and then we'll scroll down and see the next section is Data Resources and Partners. And the final section is the Data Analysis Framework Matrix. So you can really see it in one spot. We're going to look at these sections in detail. So to start up here with the data questions, again, these are the five high level questions. And when you click on these links, they'll take you to the much more detailed sub-questions that relate to those areas. Now I want to make sure you understand what these questions are referring to. So when we ask who is eligible, the data that we'll be looking at there has to do with the poverty population in your service area. So who is in your community and eligible for your services? And when we look at the question who requests the system, an easy way to think about that is we're really going to be looking at intake data there. When we look at who do we help, we're going to be looking at who do we serve compared to who are we unable to serve. How do we help looks at the level of service. So were we able to provide extended service or was brief service or limited service appropriate? And then finally, when we ask what resources are required, there are lots of different types of resources of course, but we in this site are going to be looking at hours worked. So let's go ahead and jump in. We're going to start with who is eligible. I'm sorry, let me back up. I want to keep going through the sections on this page. We'll get back to that shortly. So let me scroll down and go through these other sections. This section with the analysis types gives you brief descriptions of each of the types of analyses that we show on the site. So the snapshot description is visible right now. It says snapshot analyses measure counts or percentages for a given period, usually the most recently completed year. If any counts or percentages are unexpected, comparison trend or spatial analyses may be necessary to better understand the reasons for the unexpected results. And likewise, if you click on any of the different types of analyses, you can see the descriptions. We'll read some of those as we go, so I'm not going to read them right now. Then scrolling down to the data resources and partners section, data resources provides advice about which data are the easiest to access and the most informative. So I want to show you what's in here because I think there's some really useful information. So we'll click on that site and there's two links from here. Internal, case management system, and then external data sources. When we click on internal, we get a list of some suggested internal data fields that are good ones to start with. You have so many fields in your case management system, so it can be hard to figure out where to start. So this gives you some ideas about different case and client fields that might be interesting to start with your analysis. Then if we go to the external data sources, there's lots of really good information here. There's some information about why comparing internal and external data is helpful. There's information about using the US Census American Community Survey with some important links that describes what the American Community Survey is and how to use it. And when I'd like to really call your attention to, there's some information about when you should use the five-year estimates, the three-year estimates, or the one-year estimates. And that's described in the link here. Then there's a table that I think could be very useful for lots of organizations. It is sorted by vulnerable population topics and then gives you a list of some suggested or recommended US Census ACS tables, along with a link to a site there with the census information and not only the data, but it also includes articles and different analyses that Census staffers have conducted. So there's lots of topics here and lots of references to the different tables that you might want to use. Finally, at the bottom, there's a list of what we're calling data aggregators. These are just some examples. These are organizations that aggregate massive amounts of data and do some of the analysis for you. These kinds of entities exist all over the country. So again, these are just some examples you might want to look for some similar resources near you because they may have done some interesting analysis that would be relevant to the questions that you're interested in answering. And then finally, there's some other sources of raw data. They aren't listed specifically because they'll vary by location, but these are often sources of good, needy, external data as well. So if we scroll back, oops, sorry about that. I'll go back to this section and look at the partnerships. So the partnership links provide potential connections to organizations with access to data and analysis capabilities. It can be especially helpful for you to partner with an organization that specializes in research and data analysis when you're performing large and or sophisticated analyses projects. So the link here, there's a link here that's to a document that provides a list of 628 potential University Social Science research partners from around the country. The document is sorted by state. And so I would encourage you to take a look at that, download it, and look at some of the organizations or university departments that are in your area. And then back to the same area. Origins, this is more detail about the background, how this all started, the original TIG project. And so if you'd like to read more about that, you can find that out here. So then finally, the data analysis framework matrix. This is a visual depiction of the structure of the tool itself. It's a point of entry into the analyses and their instructions and examples. The types of analyses are listed in the colorful boxes on the left. And then the five high level data questions are listed along the top. So you could think about the question that you're interested in answering and check out different types of analyses that might help you answer that question. But we're going to actually go ahead and enter the tool up from the top because I want to make sure that you get to see the detailed questions. My guess is that once you use the tool more and more, you might go straight here and jump into the analyses from this spot. But let's go ahead and start with the data questions. And here now we're finally going to go ahead and click on the who is eligible data question. So this takes us to the detailed sub-questions that fall under this eligibility category, such as how many people are eligible? What are the demographics of people who are eligible? If we skip down a few, how has the number of eligible people changed? Where do eligible people live? And you'll notice that there are links then to the different types of analyses that would help you answer these questions. And I will note that this is really just the beginning. These sub-questions are designed to sort of prompt your curiosity. You may have other specific questions relevant to your clients or your area. And hopefully some of these will help you think of those questions. So let's just go ahead and say that we're interested in taking a look at a snapshot of eligible people in our area. So we'll click here on the snapshot button. And this is what the example analysis will look like for every example that we have. So what you'll see are four sections, the example data question, multiple analyses are possible, data sources, and then example analyses steps. Those will be the four areas of instruction for every example that we have. You'll also notice that there's a lot of information on the page that helps you sort of anchor where you are and move around if need be. Up here are the five data questions. So if you wanted to move around and check out other questions, you could click on any of those questions along the top. Likewise, along the left in the colorful boxes are the different types of analyses. If you wanted to check out a different type of analysis to answer the question about who is eligible. And you'll also see that some of the sub questions are listed here again to just prompt you to think about the kinds of detailed questions that you may want to answer. So for this particular example, our data question is how many people with specific demographic characteristics? And in our example, we'll be looking at seniors, those who are 65 years of age and older. So how many of those people are eligible for our services? Multiple analyses are possible. They include things such as, what proportion of our service area's population includes seniors? What percentage of our area's seniors are in poverty? These numbers can be compared to internal data regarding the percentage of our clients who are seniors. Should it match the proportion of the poverty population who are seniors? Does a high proportion of seniors in poverty make that population a higher priority for our organization? Then scrolling down to the third section, data sources, this suggests two possible tables from the American Community Survey and gives you a link to the spot on that site where you can find these tables. And then the example analysis steps begin. And in this particular example, we've shown some screenshots of what it looks like when you're working with the American Community Survey in the Interactive Site American Fact Finder and gives you some pretty detailed instructions on how to find just exactly what you're looking for. So, excuse me. So in this particular example, once you've opened the American Community Survey and gone into the advanced search, it tells you to click on the geographies box. And when that then pops up this next screen, it suggests that you figure out the geographic type that you want. In this instance, we selected state. And then we were able to select from a specific state and we selected Montana. Then we clicked on add to your selections. So now anything that we pull in will be limited to that geography, that state. And then the screen then allows us to type in a topic or a table name. And above we'd provided some specific table names. We went ahead and entered in this particular table, the B-17-024, which is the age by ratio of income to poverty level in the past 12 months. Again, if you're unsure about whether to pick the five-year, three-year, or one-year estimates, there's a link to the site that will explain that for you there. Then the steps for conducting the analyses are pretty basic and high level. But there are some tips that folks have run into that we want to make sure folks are aware of. So for example, sometimes when you download data from the ACS site, some of the data that you need to be in the format of numbers downloads as text. And so this calls your attention to that and suggests that you highlight those relevant cells, right-click, and select convert to number. This is the kind of thing that if you don't think of, it might make it really impossible for you to conduct your analysis. So little tips are included such as that sort of thing. Now at the end of the instructions for all of the examples will be a visual that is an example analysis. The instructions don't go into great detail describing the steps in Excel. But that's because of the wide familiarity with Excel. There are other examples using analysis tools that are not as well-known and those will provide more detailed steps. So this particular analysis looks at the share of the poverty population in Montana that is made up of seniors and then also the percentage of seniors who are in poverty. So two different but important ways to look at poverty for seniors. So the left-hand side shows that among everyone in poverty in Montana, 14% of those folks are seniors. Whereas the right takes a look at the whole senior population and shows us that 31% of seniors are in poverty. So if we think back about some of the questions we've looked at, this might prompt us to say, to wonder does the fact that one-third of the seniors are in poverty make that particular population a high priority for this organization? It's something that they would have to answer but it certainly provides some useful information to help them answer that question. So now let's say that you're still interested in understanding your eligible population but you wanna look at a spatial analysis. So you can scroll up to the top and we could click right on the geographic distribution and I think that eventually again, once you use the site, that is what you would do. But I'd like to make sure that we take a look at those sub-questions first. And so I'm gonna go ahead, like you always can do from anywhere in the site and click on the data analysis framework that takes us back here to the heart of things. Click on that eligibility question and now I'm going to take a look at one of the spatial analyses. In this case, the geographic distribution analysis. And the sub-questions there are, where do eligible people live? Where do eligible people from different demographics live? And where do eligible people from your defined groups live? And up here, we suggested that defined groups would include the vulnerable populations that you as an organization have identified as high priorities. So we'll go ahead and click on the geographic distribution and we're taken to an example much like the one we just saw, except we can see we're still in the who is eligible section but now we're looking at the geographic distribution. Our question here is what is the distribution of our eligible population throughout our service area? There are multiple analyses possible, such as where are the areas with high poverty and low poverty rates? Where are the areas with high and low numbers of people in poverty? There's a note, keep in mind that looking at numbers and rates can result in very different patterns of poverty and thus patterns of eligible population. You can look at broad trends by county or municipality or trends closer to the neighborhood level like census tracks or block groups. Are the patterns of poverty as you expected or are there any surprises? And any areas you expected to see different poverty levels than the data they show. Then for data sources, this uses strictly external data, one of the same tables we looked at before the S-1701 poverty status in the past 12 months. And then for this example, we have two examples that we'll look at. Now the first one, example one, does the distribution of the eligible population vary by community? This was done using GIS mapping to look at data for Northeast Ohio and this was an analysis from the original TIG project for which we worked with a professor from Cleveland State University. So he used ArcGIS, a mapping software that most legal aids don't have. So the instructions here are very high level. The idea being that an example like this could be shown to a social science partner in your area with ArcGIS capabilities and that person would know how to replicate this type of analysis. But we wanted to make sure you had some examples that wouldn't require going to external resources. So the same question here in example two, does the distribution of eligible population vary by community? In this case, we use the American Fact Finder built-in mapping. So if you follow the instructions from above about how to get at the poverty data, these instructions here show you that from that area, from that section of the American Community Survey website, you can actually create a map and it's as easy as clicking on create a map. And you can play around with the formatting. This particular map shows the levels of the poverty rate in the state of Wyoming, the darker green being the higher levels of poverty and the lighter color being the lower levels. I should have mentioned above, likewise the darker areas here show the higher concentrations of poverty and the yellow showing the lower poverty rates. Okay, so now let's say that you're interested in another data question. You're interested in who requests the system. So we can go ahead and click on that question and it will take us to the sub-questions. This time around, we're gonna hone in on the trend sub-questions. So we'll take a look at some of them. How has the number of people requesting assistance changed? How have the demographics of people requesting assistance changed? How have the number of people requesting assistance from your defined groups changed? How has the number of people requesting assistance with certain categories of legal problems changed? Et cetera, you get the idea, lots of ideas. So we're going to look at a trend analysis and just a reminder that the description of the type of analysis appears at the top of every screen too. So this tells us that trend analysis scrutinized changes over time in client conditions, reviewing trends over a five year period or longer when possible. Spikes or dips that appear in trends might confirm what an organization expects or raise additional questions worthy of investigation to better understand the unexpected change and determine whether it calls for proactive steps. Our example data question here, how has the number of people requesting assistance from our organization changed over time? And again, multiple analyses are possible. How do the trends in intake numbers compare to the trends in eligible people over time? What have been the annual changes in intakes compared to the annual changes in the eligible population over time? What percentage of all eligible people have requested assistance in each of the last five years? The data sources here use a combination of intake data from your case management system. And there's some instructions here about the kinds of fields and filters you would want to apply. And then that same external American Community Survey poverty data. This particular example describes how to combine that internal intake data and that external poverty population data into one chart that looks like this. So, and it specifically suggests creating a combination chart in Excel with intakes on one axis and the eligible poverty population on another axis and using a different format. So again, this is called a combination chart in Excel and you use a primary and a secondary axis to depict these different pieces of data but that allows you in one place to look at the trends. So we can see for this particular area that the eligible population has been on the increase and intakes in general have been on the decrease with a slight increase in the last year shown. At the bottom of the page is an example of an intake dashboard. Again, intakes and who request assistance are the same thing in our site. This example uses an open source program called Microsoft Power BI that can easily use data from Excel among other sources to create really nice visuals including multiple types of analyses. I know a lot of legal aids are interested in dashboards and this is a really easy program in which to create them. If you're interested in more information about how to use that tool, I presented a tutorial on this program at the January 2017 TIG conference and if anyone wants me to send it to them, my contact information will be available at the end. All right, so now let's take a look at one of the examples that analyzes who we help. So we'll click here on the data question and again, we're taken to the sub-questions. And this time around, we're going to hone in on a geographic concentration analysis and so we're gonna look at those sub-questions including, does the geographic concentration of intakes match the concentration of served people? Does geographic concentration of intakes match the concentration of served people with specific demographics? Let's go ahead and take a look at the example. So geographic concentration is not a concept that's familiar to everyone so I wanna talk about what it is. The description says that geographic concentration analyses compare geographic concentrations high or low of multiple variables to determine how the variables and location impact each other. It uses something called location quotients which can sound kind of intimidating but this example shows you how you can create them pretty easily. So in a nutshell, let's look at the question. The question, how does the proportion of people we serve from different parts of our service area compared to the proportion of eligible people in those same areas? So in a nutshell, we wanna compare the share of the poverty population by county to the share of the clients we serve. So for example, what does it tell us if one county makes up 10% of the whole area of poverty population but only 5% of the clients we serve are from that county? There's a mismatch there and maybe that's appropriate but maybe we need to understand why there's that mismatch. So location quotients compare those different percentages and determine whether the service provided is lower than expected, close to expected or higher than expected. The data sources for this particular analysis include closed case data from your case management system and again, there's some details about specifically what to include and then again, poverty data from the American Community Survey. So this example uses Microsoft Power BI which I mentioned earlier in relation to the dashboard that we were looking at. I did provide more details knowing that most folks have not yet used this software. Still, I think following these instructions you would find it actually pretty easy. I'm not gonna go through them word for word here but basically you gather external data about the share of the total poverty population that resides in each county. Then you gather data about the share of all served cases by county and then you divide those two percentages. So you end up with numbers that range from 0.1 to three or more and anything under 0.75 means that there were fewer cases served than would be expected based on that county's share of the poverty population. So if we look at this map from Microsoft Power BI, the red areas are where fewer than the level of served cases was fewer than expected based on that area share of the poverty population. So you'll see a few of those areas and then anything that's yellow is in the expected range. The amount of served cases in that area matches pretty closely to that area share of the poverty population. And then anything in the green area which is above 1.26 is showing us that there are more cases served here than would be expected based on the share of the poverty population. Note that the areas in white are areas with fewer than 20 cases and they were omitted from the analysis because of the potential data skew with such low numbers. So even though I think this probably looks like it would be pretty hard to create, I promise you following these instructions and again, they're quite detailed but they're intended to be. If you follow them closely, you can pretty easily create a map that looks just like this. And format it in any way that you think works best or that you like best. So now we're gonna keep moving through the different types of analyses and now we're gonna look at how do we help? So again, here we are at the sub questions and we haven't yet looked at a comparison. So I think that that's what we'll do. We'll look at questions such as how does level of service differ by legal problem among different demographics? How does level of service differ by legal problem among different substantive groups? Okay, so we've got a quick question here which is regarding the, has strategic data analysis used the US Census data ferret tool? And if so, is this a tool you would recommend or any feedback on it would be great. So basically, have you used the US Census data ferret tool? I have not. Okay, neither have I. So I cannot comment, yeah. I just Googled it and looked it up some but I was not familiar with it before this webinar. Yeah, I have not, I'm sorry I can't comment on it but I will definitely look at it after the call as well. Yeah, so also feel free to email me that question offline or I'll get together with Rachel and we'll write up something super short for the blog over if it's useful and how it could be used, something of that effect. You have stumped us, which is very rare in these webinars, well done. Great, okay. So let's go ahead and take a look at the comparison. And again, a comparison analysis, comparison analysis review linkages between two or more variables and uncover information about client conditions and data relationship. As with all of these, we suggest that when unexpected data relationships are discovered, investigation is warranted to better understand those linkages and determine whether they indicate the need for client service and advocacy work that simultaneously targets multiple conditions at once. And then also just a reminder that the sub questions pop up here as well. So our question in this particular example, how does, how do levels of service differ by legal problem among different racial categories? And there are some details about different types of analyses to consider, which legal problem codes have the highest percentage of extended service cases. Within problem code groupings, the clients with different ratios have similar percentages of extended service. We're looking here at intake and close case data and especially looking at information that helps us to determine whether the client was served or not and the level of service. So this shows a pretty basic table but with some really powerful information about the levels of service provided, brief versus extended. In this column it's brief and in this column it's extended. These sections show the problem codes and then breakdown by racial category is here. So some pretty basic information but some powerful information as well that deserves some attention. So for example, looking at the yellow highlighted cells here in the food stamps category, we'll see that 60% of Hispanic clients received extended service while 38% of white clients received with food stamps cases received extended service. So why is this? Is it purely because of language issues or is there something else going on? It'd be important for the organization to understand that. And then as another example, looking at the red circles, we could see that 55% of all food stamp cases received extended service while only 6% of the private landlord tenant cases received extended service. So these kinds of differences might be easily explained by the organization but they might warrant further investigation. So let's keep going and look at an example that looks at the hours, the resources in this case we're looking at hours required. We're going to look at another geographic distribution analysis in large part because I want to show you another tool that you might consider using. Our question here is does the average hours per case that we spend on cases vary by client zip code? Are certain types of cases that require more time or less time coming from specific geographic areas? Do we tend to spend more time on cases or clients who live near our offices? And this example looks at closed case data from the case management system. And this example uses CARDO which is another free and relatively easy program to use but because I think it's pretty new in our community, there are pretty detailed instructions here. Again, I won't go through all of the details but they describe subtotaling hours worked by zip code, cleaning up the zip code data. That can be a bit of a chore but achievable and then using CARDO to map those data. The instructions also include information about formatting the visual to make it more understandable. So the map here shows the average hour per closed case in 2016 by zip code for zip codes with more than 10 cases closed. Yellow areas show an average of four hours per case while the dark blue areas show an average of 24 hours per case. So the organization will want to take this information and understand, try to understand why the difference. They might want to layer on top of it some information about particular types of legal problems and how they are spaced out geographically to try and understand if some particular types of legal problems are concentrated in certain areas or if there's something else that's causing more hours per case in certain areas than in others. So that's actually the extent of the examples I was going to walk through today and we can open it up for questions or I can show some more examples. What do you think, Brian? So far we failed to cover the one question that we had but I think the examples are really the most practical things that can come out of this. So I would be happy to see one or two more examples if you've got more that you're willing to share. Sure. Because we've got another at least 15 minutes here. Sure, happy to. Okay, so you know what, let's look because this idea of looking at the resources is one that we recently added to the framework. So I think maybe we'll look at a few more examples in that area and maybe move around. So in this particular example, we looked at how many hours did we spend last year on cases from different legal problem code categories and what is the average hours per case that we spent on cases from different legal problem code categories. So I'm just gonna go ahead and scroll down to the visual and show you again their instructions on how to create this kind of analysis. But here's an example that breaks down the data into five different substantive areas, shows the average hours per case for those areas. And then likewise shows the total hours for that particular area for this particular legal aid. And so what's interesting I think to look at here is the differences in these averages and to figure out if you understand why those differences exist. In this particular case, they're not terribly different, those averages. But let's just say this prompts more curiosity and we wanna look at a comparison. All right, so we're in the comparison section and again, I'm gonna scroll right down to see now these visual, but suggest that you read through those steps in order to be able to recreate something like this. Now this breaks it down by served and not served. I think some more interesting questions maybe arise here. One being why is the time that we're spending on cases that we don't end up serving increased? And how do we feel about spending almost one and a half hours on cases that we don't actually end up serving? Does it maybe mean that we're providing service sometimes and not giving ourselves credit for it? Or what does that mean? And then likewise, sticking with this, let's look at some trends again related to resources, scrolling down to the analysis. And this looks at one particular, one particular area, one particular substantive area, family law cases, and it shows the average hours spent on free service cases, extended service cases, and cases not served. And one thing that I might wonder about is the amount of time spent on the brief service cases. Is that high or is that appropriate? Should some of those have been actually closed with extended service? It would be something for that organization in that particular practice group to think about. And of course, you can do this with any substantive area. Let's see, let's show some more, show another snapshot of who do we help. And hopefully by moving around like this, it gives you a sense that you can move around through all the, I guess, 25 examples that we have here and get lots of ideas of your own. This looks at what are the demographic characteristics in this particular case, poverty levels and language of people who requested assistance, so intakes, and received it in the past 12 months. What about those who did not receive service? So we've got a few visuals here. We can look for this particular organization at the share of those served based on their level of poverty. So for this organization, 45% of their clients were below 75% of poverty, 34% were in the 75 to 124 range, et cetera. You can look at that same breakdown by language for this particular legal aid organization, 86% of those they served were English speakers, 9% were Spanish, excuse me. And then another visual shown here are some charts that show a breakdown of the served and not served counts and percentages by poverty level over here and then by language over here. So there can be some interesting comparisons when you look at the levels of service, when you break clients down in these different areas. And then finally, there's another example on this page of a dashboard, in this case, a closed cases dashboard using Microsoft Power BI. Brian, do you want me to keep going through all the examples? No, I think we're good at this point. I think we've got enough. We are actively seeking feedback on the data analysis framework. We have one more final webinar coming up here in the fall where we're going to be going over this. If there is anything that you would like to see changed, updated, that type of stuff, please let us know. We've still got a little bit of development funds in our budget to continue to improve this. We thank you all for coming out and please feel free to get ahold of any of us here working on this project. Scott is here and can help people with the Drupal side of things or the technical aspect. Rachel is the strategic data analytics guru here and I am kind of the jack of all trades working on lots of different technology initiatives through alicentap.org. Thank you for coming out and we look forward to hearing any of your feedback. You're going to get a survey after this is done. We take those very seriously. If there's any feedback you can provide to us in the survey, we greatly appreciate it. Any final parting words, Rachel or Scott? No, just thanks everyone and I would encourage you to play around with this site and hope that you'll find when you do that you too can perform these kinds of data analyses pretty easily. I would just say thanks. Thank you. I hope to see some of you out at our can you work remotely which happens to be very relevant to the individuals who are on this panel. As I know that all three of us have worked remotely on several different projects. That is being put on by Just Tech. Tomorrow it is the first time we've worked with them to put on a webinar series and I look forward to seeing how it goes. We've got some great speakers on there.