 Okay, welcome everyone. I see that we are 70 people now in. We expect a few more. But we will start slowly on this workshop. So I'm happy that all of you are here. It will be an exciting day. Also technical wise for us. It's the first time we run this for the GIS and Maps topic online. So just a quick introduction. Most of you also follow the webinar we had back in March. So you already know me. I just want to add that it's great that I'm as a developer. So I'm working full-time on developing the Maps app of DHS2. And I think it's a great privilege to be able to talk to our users directly. It's a great way to get a feedback and also today because we would like you to do the exercises on your own DHS2 instance. It will be a very good test of the system of how the Maps app works with different setups. So I really hope that you also use this opportunity to give us feedback both if you discover bugs. I really hope it won't be too many. We do a lot of testing of bugs on our own instance, but some bugs are hard to discover because it's directly linked to different setups. So if you find if something breaks, something don't work, please report it to us. And at the same time, and also if there are missing features, if there are something you would like to show on the map, read, be able to click on everything that would be nice to have, please give us the feedback. And we will try to, if there is enough demand, we will try to add it. It's a continuous effort and we add new features in every release and we try to meet the demand of our users. So I've worked on the Maps app full-time now for about five years and will continue in the years to come. About you, some of you have started to introduce yourself in the Slack channel. I will show you shortly, but if you have access to Slack already, please just add a few lines about yourself. It's really nice just to see who you are, although we can't meet in person. There has been a lot of registrations for this workshop, which is very good. So we have more than 200 registrations. We don't expect all of you to show up. Some of you might sit by the same computer and some might watch the recording afterwards. But from the registrations, there are people from 66 countries which are marked on this map, mostly Africa and Asian countries. So this is the plan for today. I know that you are all in different time zones. So this is the Central European time in Oslo right now. It's 10 minutes past 10 and we are in the introduction. So this workshop will last for three hours until one o'clock Central European time. And then right after from one until two, as long as there are questions will be available also here on Zoom to answer and Slack and answer any questions you might have. So this is a bit different from the webinar we had back in March. So if you would like a more general introduction to the Maps app, I would advise you to look at the demo, the recording from the webinar that we made. We will also have some short presentations today, but the focus is on you as this is a workshop. So you will need to do some work. So we have four exercises plans and one extra that you might do on your own. So we will spend some time on this today. As I just mentioned, we would like you to use your own DHS to instance. So we have made all the exercises in a way that they can be applied to any DHS instance, but some of the features you need to use requires a certain version of DHS to. So it might be if you are an old version that you cannot do all the exercises. And for this reason, we have also created a workshop instance just for this workshop that you can use, which is running DHS to 35. And this is also posted on Slack. So you will find it there. But if you can't do the exercises on your own instance, you can use this workshop instance. As there are so many participants, we don't plan to have any hand ins. So to get a certificate, you need to follow all the sessions. We expect you also to do the exercises, but we want to control the exercises. And to be able to sort of figure out that you have actually follow the workshop, I will say a quote, some words in one of the sessions today that you need to write down and mention in a form to get the certificate. So I will tell you this loading clear, but I won't tell you where. So you need to follow on the sessions between the exercises. Even though we don't have a hand in, we still want to get feedback and I will show you later how this will be done. So this is our Slack channel. I hope everyone have access. If you don't have access, please tell us in the Zoom chat. You should also find some link in the Zoom chat now to the Slack channel. So it's very important that you have this Slack access and open. There is one announcement channel where we will tell you important information from the instructors. And then there is a questions channel for you to ask questions for us and fellow participants. And then we have one channel for each exercise. And while you are doing the exercise, this is not something you need to do, but it would be very nice if you can share feedback while you are doing the exercises. If you, you can show your results or you can tell us if you have trouble doing some of your exercises. And then we, it would be very nice if everyone could just write a few lines about yourself. It's a great feedback for us as well to know who you are and what you're working on and so on. So please do that. I will just quickly demo the Slack. So this is the Slack workspace. And please go to introduce yourself now and just write a few, a few lines. It's also nice if you add a picture and your full name, but it's not required. And then this is the announcements channel. So during the day we will post announcements here. So this is where you will find, for example, the form to get the certificate where you need to fill out the quote for the day. And then we also have these, for each exercise, there is a channel. And I see even some of you were really eager to start. So I posted this last night, all the exercises are here and some of you even replied yesterday. So this is just, I think it's nice how you have, some of you have already started. Post the results, post some screen from the Maps app. This is not a requirement, but it's great for us to get this feedback. So please continue to do this. So let's start with definition. So, so far in this, we have mainly talked about maps. Now we are going to make a little broader view. And we'll talk a little bit about GIS, which maps is a part of. So, so GIS, the previously the maps application in DHS to will call was called the GIS app. We decided to remove this for two reason one is that just GIS is a more technical term and not everybody knows about it by most people know what what a map is. And also because GIS is a much broader term, while the maps up is basically about making maps and showing maps. So GIS stands for geographic information system. So if you're studying GIS, it will often the meaning will often be science instead of system so geographic information science. So I have a master in geographic information science at the University of Edinburgh. And it's defined as a computer system for capturing storing checking and displaying data related to positions on the Earth's surface. What I mentioned here is that on the maps up is only focusing on displaying data for of these four. But so in a way we can look upon the whole DHS to ecosystem of software as a GIS because you will use other like the Android app for capturing events which might have a coordinate you will use our database. We also have a database called whole GIS, which is specially GIS database, where we store the data, there are other tools that will help you to check and validate the data. And then at the end, we have the map up to display the data, and also still in this workshop, we will focus on the display of data because that is most useful for most of you. Also, as I mentioned in the webinar, you have this possibilities or this special capabilities of GIS is that you can add different information layers on top of each other on a single map. So the map is built around this notion. And you can kind of mix this lay different layer type as you like. And this allows us to to see and understand patterns and relationships. That's why we created maps because we want to see, understand why is there a higher, something more of my higher levels in one part of the country or the other. And then you can maybe add another layers to get help you explain why this is happening. So far the maps up if mostly about seeing, but I will show you later at the end of the day how we have added more analyze capabilities into the maps up in the last release. And also, the, tomorrow, we are going to use a GIS program called Q GIS, and that will focus on analyzing data, and also how you can combine different layers with each other to see if they're to see the connection. So now I will hand over to to Austin. Do you need to share something or can I just keep the slide. That's fine. Hi everyone. My name is Austin McGee. I am happy to be here with you today, helping with beyond to introduce the maps application and some visualization and analysis of geographic data, and particularly geographic data for health, which is what DHS to is very good at. A little bit about me my background is in software engineering I've been also working in the NGO space around water, particularly in Tanzania Rwanda and Uganda previously before coming to DHS to about two and a half years ago. So I'm going to focus on building some things like the application platform for DHS to which allows people to extend the DHS to to other systems to adapt to their local environments. I'm just going to go over this very quickly because we're already pushing our time a little bit, but I wanted to give a quick introduction to GIS or maps or geographic data and health, and a little bit of the history of that. It wouldn't be a workshop on GIS or or geographic data for health without talking about John Snow, who is known as the father of epidemiology. There, there's some debate about if he actually is but he has become known as the father of epidemiology, and who who was john snow so john snow was a person in London, and during a major color outbreak. And where this was a in a time before they knew that diseases were transmitted because of microorganisms like bacteria or viruses. And so there wasn't a concept of epidemiology as we know it today. And what john snow did was he looked at through the data that he saw coming in every day from this color outbreak in London. And the common theory at the time for from the health professionals was that this was the color was coming from dirty air basically so they thought that pollution was causing cholera. And so they were trying to clean up the air as much as possible. And in doing so we're tossing a bunch of things into the river, for instance, rather than releasing it into the air which was problematic. And so what john snow was able to do was look at all of the places where there were cholera cases, put them on to a map which was one of the first times that someone had done this, put them on to a map and try to figure out what they had in common. You can see here the map that's on your screen right now you can see that all the red dots are cases of cholera. And you can also see in marked in blue here are all of the water taps, the pumps where people were getting water out of the ground. And you'll notice if you if you look at this visually and this is one of the things that john snow did for the first time in the world was that if you look at look at the pattern here you'll see that all of these red dots in this particular area are clustered around one of these blue pumps. We had no idea that diseases could be transmitted through water through microorganisms like bacteria or viruses. But this information and presenting the data in this way on a map was enough to convince the city council to actually take the handle off of that pump, which then helped to basically eliminate that source of cholera in for this particular They later were able to determine that their cholera is a micro organism that's transmitted through water, most commonly, it's not transmitted through the air. But even without knowing that looking at this information on a map was able to help solve this big outbreak of cholera in London. And so this is something that just to keep in mind as we're looking at a lot of kind of technical things about maps and presenting information is that it can actually be used in a very kind of powerful way to address real real problems in particularly health around the world and it has been for a very long time. So just wanted to give a little bit of a quick history introduce myself and we'll turn it back over to beyond for the next part of the workshop. Thank you. Okay, so These are our maps releases. So I just want to mention this again. I won't spend too much time on it, but just that you know that we have added new features as we move ahead. So depending on your version there might be some features that you don't have support for and we will try to mention that in the exercises and then I mentioned also if this feature is not supported on your instance you please use our workshop instance instead to try the exercise. The first exercise is we start easy it's about checking out the different base maps we support and also use some of the the map tools. So we are very quick recap from from the webinar we had before the the exercise. So this is the base maps we provide there might be others in earlier version but from 235 these are the one we support. So we have a light version, which is the default one, which we recommend as a background for your thematic layers because it don't, it don't clutter your map it may still makes it easy to read and put emphasis on your layers. And then there is a more detailed open street map player, which are much more names and roads and buildings and so on. So if you need to see the detail or find your streets. This is the one you can use. And then in addition, we have some layers from big maps, especially useful is the satellite injury, which is very nice very good for for large parts of the world. One of the tasks is that you are going to find your own town or even your house if possible. So you just need to navigate the map, you can use the zoom buttons here. Or you can also just double click the map or even the scroll wheel soon to zoom in and out. And then you can drag the map around this is as you Google Maps and others mapping application, just the same for these two maps to locate your house. Then we also have the search tool, which is the magnifier button. So we would like you to test this search tool for your own country. And try some bigger and smaller places in your own country and give us some feedback if you're able to find the places. So this will give you some give us some feedback if if the search is good enough. Or we should maybe try to look for for a better searching tool. This is also based on open street map so often if you the name is shown on open street map. It should also show in this search. Lastly, I think that this is the measurement tool, which is this ruler icon, which you can use to measure both distances distances and areas. And we have an exercise for this. And then also I want to explain now the difference between latitude and longitude. Some people, some of you might know this, but we use latitude and longitude as the coordinate format in DHS to the reason is that this is DHS to is used all over the world. And this is a shared global format that can position that position position anything all over the world. So it's a good generic system to use very often your own country might have coordinates in like a national coordinate system. And then you need to translate into latitude and longitude before you can enter the coordinate on DHS in DHS to. So knowing latitude and longitude and the difference between them is a bit important. So latitude goes from west to know sorry longitude goes from west to east. So by the prime meridian that goes through Greenwich in London is zero longitude. And then if you move eastwards, compared to this map, you will have a positive longitude number. And then on this side, you will have a negative longitude number. It goes up to plus 180 and minus 180. And then the latitude goes from north to south. So zero is the equator. So bow the equator. It's a positive latitude number up to 90 at the north pole. And then a negative number below the equator down to the south pole. So, just by knowing in which sort of hemisphere you are, you can sometimes tell the different what is lying to you that longitude so if you are in this part of Africa. Longitude numbers will be positive and your latitude numbers will be negative. And I've added one here for the Victoria falls. Oops. So you can see one number is with a minus. And then that should be will be the latitude number and the postal should be the longitude. So before we have been not very good sometimes we just showed that type to coordinate and I see, and we didn't really specified what should be first and last. So, no, whenever we try to do something new with coordinates, we always specified what is longitude and what is latitude so you don't mix them up. And then if you can't find your health facility on the map. It is useful to so sort of think that it's the, it's the wrong way around. So instead of here, Victoria falls might be placed out here in the, in the Atlantic. And also, you don't need more than six decimals in these in these numbers, because that is the meter position. And for our use cases that is usually enough. So you will also share this slide on the on slack under the exercise so you can have a look at it there. And also to mention there is a loss of when you don't have the coordinates I see some of you type 0.0. And there are even some coordinates like that in the demo database. And since 0.0 it's actually a valid coordinate coordinate so you should rather leave it empty than typing 0.0. So we will do this all over the world so people have named this imaginary place the new island is in the middle of the ocean. This is the only thing that is there. But there is tons of health facilities schools, kindergarten, everything at this location. So if you Google new island you will find this like a whole internet community travel agencies, everything for it. So so that's another place to look for for misplaced health facilities. Lastly, you can click anywhere on the map, and then select show longitude and latitude, and that will display the coordinates of that place. So we have one task for you to find the latitude and longitude of your hometown or home building. So for the exercise, I will just, I will go through them very quickly here. I won't demo it, but I will go through them afterwards. So you are going to try yourself first. So the first is that you're going to find where you live on the map. You can either zoom in or try to search for it using the search tool. And then find the latitude and longitude of your building. Then we would like you to try the search place search and just search for some small and bigger places and see if you're able to find them please give us feedback if if you are not able to find them or or are. In the search field, you can also paste in a coordinates. So we would like you to take this coordinate and paste it into the search field and see where you where it will take you. And also answer which is longitude of those two numbers you should see that after you can't see it from him, but you should see it after when it's shown on the map. Then we would like you to use the measurement tool to find the distance between your town or neighborhood and the capital, or if you are within the capital already from your building to the parliament building. And lastly, if you can turn on the satellite imagery for the showing the parliament building and measure the size of the building these are usually quite huge. And then at the end, if you like, you can post a screenshot of this measurement to to the slack channel. So as mentioned, this is day one exercise. You will find the the exercise I just mentioned here as well. You have 15 minutes. It's a little bit late already, but we will take it back. So you have 15 minutes. So to 10 to to 15 minutes from now, and then we will go through it together. So good luck with exercise. And there was one just quick question that might be irrelevant here, which was about being able to do, especially this exercise with some of the other exercises as well in a version that isn't 235. And I think the answer I put on slack but just wanted to share with everyone here is that, at least for this first exercise. So as mentioned of 233 version. This, this should all be possible in the 233 version so feel free to do that on your own instance, but you're also welcome to use the Academy instance if you prefer. That's correct. Yeah, that's correct. So, so please start on your own instance. I think the learning effect is better on your own instance because that's the one you're going to use next week. But for this first task that I think all of this should be possible to answer. If not, tell us and then switch over to the workshop instance. Thanks. So we have done with the first exercise, and then we will continue on session two, which is covering the boundary and facility layers, which are the two quite easy layers to start with. So the boundary layer is this icon here. It is very useful to use just to see look at your organization units. So of course, this requires that you have these coordinates for your organization units in your database, and we told you about this in the webinar and said you should not if you don't have them in your database. So right now we don't, because this is a bit complex task is only done once for administrators. So if you don't have the boundaries in on your DHS to instance I hope you have. If you don't you have to use the workshop instance, but then please, because this is crucial to do the thematic layers later on. So having these boundaries are very important. So if you don't have if you have a problem with it or you don't have them, please reach out to us and we will try to help you after the workshop we don't have time within the workshop. So in this next exercise you need to select some of your own organization units, and then you use this organization units dialogue. So you can select for the first exercise is just to select the first level, but you can also restrict the selection to one of the districts. In your database. I also saw it was in question when you use the the workshop instance that you only see Sierra alone, and not your own country for data, and that is true so that the workshop instance. It's only Sierra Leone demo data so you need to use that data, because we don't have access to the country specific data. So these are just some made of data for Sierra Leone we have tried to make them sensible but it's no no real data because they are protected. For boundary layer you can also change the style is not a lot you can do but you can, for example put on the labels, so you can also see the name of the organization units on the map. And then after you add the layer you will see it in the left column as we call it a layer card. And here you can toggle the layer on and off which is an exercise you can change the opacity of the layer and reorder them, edit the layer, instead of adding a new one, and also download the layer, which we will do tomorrow. There is also a data table so this can be used for example if you want to count the number of organization units you have on your map, or you can also use it to filter you can search for for a search name, for example. The facility layer it's a bit similar, but that instead of that will only show point facilities useful to show your health facilities, and it will show with icons instead of circles or polygons as for the boundary layer. And the other option here is that you can add a buffer around it with a certain distance for example five or 10 kilometers, here is 2000 meters. So you can see how like for example to see the coverage of the of the health facility. So when you select a facility layer you first need to select a facility group, which will also determine the icons. And then you can also you need to select the facilities in the organization units and you can also style it and add the buffers. So you will now try to do this yourself last just to mention that when you see the facilities on the map. You can right click on the facility and select show information this is a bit hidden feature, but you should then be able to see some data about the health facility. This you need to set up in another app, but you can check if this is already made for your instance. And if not, you can ask the system administrator to maybe add the indicators that you would like to see when right clicking the health facility. So this is the exercise quickly, you should add a boundary layer for from your own instance, this has been supported for all versions so you can use whatever. And then you could use the first level below the national level, and then maybe use the data table to see how many organs there are on this level. Then you can try to toggle the visibility on and off of the layer without adding and removing the layer, but just in the in the left layer card. You should add labels name of the organization to the map. And then to the same map so you actually have two layers you can also add some facilities. So you can select all facilities in one of your districts. And then try this show information. I just show you to see what what which what will show. And then lastly add a 10 kilometer buffer around each health facility. And then you could maybe zoom out on the map and see if there are areas that are not covered by any health facilities. And as previously you can share your answers in the day one exercise to channel. So good luck you now have another 15 minutes. So I will go through the exercises quickly. I've seen quite a few nice answers on slack. So it looks like you are doing good. I see that quite a few of you are using the Sierra Leone instance so it would be interesting for us just to have some feedback why you are not using your own instance. If it's not because it's too old or it's you don't have coordinates data for your arguments or other reasons. So please just tell us in the in the exercise channel to why you are using the Sierra Leone demo. So quickly go through the exercises so the first one was to add a boundary layer so you click on add layer select boundaries. And then for the first level, I will select district level and then add the layer. The map should automatically zoom in to to the layer or units you select. To toggle that. Sure. I believe I'm not seeing your screen. I don't know if other people are. Okay. Sorry for that. Thanks for telling me. There we go. So yeah, the first task was quite easy so I won't repeat it but just to add one boundary layer to the map. When you have added the layer, you can toggle it on and off by clicking the I symbol here to the left. And you can also change the opacity of the layer by drawing the slider. Then you edit the layer by clicking this edit button and then you have some more options here for example to see the data table in this more button here. So now we're going to edit the layer and add some labels. So then you check the labels checkbox here. I often increase the size a little bit to make them bigger and then update the layer and you should see the labels on the map. You can also change the opacity of the base map so the labels are more visible by turning that one down. So then we are going to take add a new layer to the same map. So you can add as many layers you like the recommendation is to add as few as possible but it's there is there are no limits. To add a facility layer, I will select a group by facility type. And then it's important here to select the facility level. So you need this can be named differently for different instances but it's important to select the level where you have your facilities. And then to restrict this to only one district, I can select the Canima district here and then add layer. The next step is to see information about the facility. If you just hover the facility, you should see the name and the type. And then if you right click, you can select show information. And then you should get some data for this facility that you can you can change in the system system settings apps. So this is done as part of the configuration for your system and then available for all of your users. So you can also see some data here, the code of the unit parent units and the groups which unit is a member of. The last exercise was to add a buffer around each facility on 10 kilometers. So let's edit the layer and then select style and then select buffer. And then we increase this to 10,000 meters and then update the layer. So now you should see these buffers around each one. And where there is darker color, there is a higher cluster of health facilities. So now we added for this district. So you could see here that in the north, there are no health facilities. If people are living here, it is more than 10 kilometers to the nearest health facility. Just take it a little bit further. You could then add a population layer, which we will look into extensively tomorrow, but I'll just show you here. So I can add a population layer. I know for this older version, I need to go back to 2010 to have data for Sierra Leone. So try the different years and then add the layer. So one challenge now is that this layer is added on top of the other. So it sort of hides the layers below. But then it's important that you can just take this layer and drag it so you can reorder the layers to put it on top. So we can also place the border is on top of the population layer. And then if we zoom in, we see that there seems to be some small villages here, but not a lot of people living in this area. And it's probably the reason why there are a few health facilities in this area. But this is an example of how you can combine different layers to sort of try to understand why the situation is like it is. So that was the end of exercise two. So let's move on. So the next session is about thematic layers, which is probably the one you use the most. That one and event layer. And thematic layer is to show aggregate data for your organization units. So we have this layer has been there for a long time. It's this kind of the default standard layer of DHS too. And the standard way of showing this layer is with a technique that is called a coroplet map. And until 235, this was the only thematic layer visualization type we supported. And now we have also added bubble map, which I will show you shortly. But coroplet map is a map where you have these predefined areas, which is your organization units that are colorized, you add a color to it in proportion to a statistical value. So this is the preferred method you would typically use to show an indicator. So what is a bit important here to remember is that you should not use this map coroplet map for total numbers. So, for example, the it should only be data that is in percent or for or per capita, because you will compare one district to another. And if it's total numbers, you don't take into the account the number of people living there. So it could look like a region is doing worse than another region, but the reason it's not really the truth, because it's just that the population is so much higher. So we call this data to be to normalize and that is quite important to use this thematic mapping technique. Then another important issue is how you classify the data. So that means how you assign colors to your own unit from a statistical value. So so far we support three different options. One is to have a predefined legend, which is often good to use, for example, for percentage. And then you can also add your own color. So, for example, if you consider about a certain rate or a certain percentage to be bad, for example, you could add this color red and maybe green if something is is well performing. And these predefined legends are defined in the maintenance app, not in the MAPS app. And these can also be used in other visualization apps like the data visualization. And then we also support to we call it automatic classification methods because it's not predefined is something that is done on the fly. And those two we support are equal intervals and equal counts and equal intervals means that we take the lowest and the highest number for all your organization units. And then we divided into a number of classes, which all have the same size. So for example, zero to five, five to 10, 10 to 15. And then each of these classes get a specific color, and then we put all of your arguments or facilities into these classes. Often I prefer to use equal intervals because it might give it will give them most, I would say correct image of your situation, because, yeah. And the other option you which you can also use to sort of enhance the view is called equal counts. And what we then do is that we take all your organization units, and we make the classes based on equal number of all units in each class. Here, for example, you can see the number in small here, but it's six or units in this class. But while here there are either three or four in each class. So that means that on with equal intervals only this one get the darkest color, but using equal counts, we get three or six or districts, having this in the same class and getting the same color. And this might some would say, give a wrong view, because these are actually behaving better and it more belongs to this class than this class. And also remember you can also select the number of classes so in here there are only four, but you can also select more classes. My recommendation is to use maybe if there are pre good predefined legend or equal intervals, but you could test, but just tell that you should just try to, to tell the what is considered is the important and the true story behind your map, while you select your classification. Predefined legend. We won't do this as an exercise because I cry I've created the exercises so you don't mess up your own on instance. So but you can try this on your own to add more predefined legends. And then in addition to be able to create the legends you can also assign a predefined legend to an indicator on data elements. And if you do that that will come up as a default in the maps app so the users using the indicator don't need to to make this decision because it will always already have a predefined legend. Then from 335 we also added a bubble map because we saw that there was a limitation here, especially with the cool with coming up where there is was often a need to map the number of cool with cases. And then if these are not not if these are total numbers it's not a good way to map this using a corrupt map. But then instead to use something we call it a bubble map, but it's often called a proportional symbol map. And the symbol you you size in proportion to a statistical value is often a circle and that is what we use. So here on this map the size of the circle and the color will show the number of BCG doses given. And if the audience you select is area or polygon, we will just place the circle in the middle of that polygon. And in addition show the boundary so you can sort of see where the circles belong. So for thematic layer you select the for the exercise you select the use the indicator, which is often normalized for corporate map and then you select a data element which is often total numbers to use for bubble map. And then under period, we have also added the possibilities to to show changes over time. So, I think this is from 233, you can either show a timeline and then you can step through the time to see the changes from one month to another, for example, or you can have these split view maps, where we have multiple maps on the screen at the same time. And then you can easily compare and contrast the different the different months. So there is an exercise of of doing both. And then under style is where you select between the core plate and bubble map and just remember core plate is always the default. But if you select a data element, you should use rather use a bubble map. And here you also see that you can select the the legend type and the how the classification. So please try and and figure out by yourself and nothing bad can can happen if you just click around. So this is the exercise number three. So the first exercise is to create a thematic map a core plate from one of your indicators. And use the a predefined color legend if defined if not just create one yourself. And then I would like you to try to right click one of our units and see how you can drill down and then drill up again between the different levels. And then exercise to edit the same layer and switch to an automatic color legend and then try both these options equal intervals and equal counts and see how it's affecting the map and maybe also think about which one you would prefer for this particular indicator. The number three is to create a timeline map from one of your indicators and and then navigate between the periods. So this requires 233. DHS to and then switch to split your map for the same indicator. And then you can compare those two. And then please tell us which one you prefer to use. The last task is to create a bubble map. So that requires 235 so use the demo workshop instance if you don't have this version and and use a data element and add it as a layer. And this one you can also download and share on our slack channel. So good luck good luck again I see you are doing good so far in the on slack so so you have 20 minutes on this then what this one is a bit a little bit more complicated so you will have 20 minutes. And then stop the recording. The size was to create a corrupt that map with the default predefined legend. So I go to add layer select thematic selects I will just select and see and the first indicator and see one coverage. And this should be enough to create a map because there is a default period selected which you can change in your system settings what should be the default period. There is the default is also to select the first level below national level for all units. For style, if there is a predefined legend defined for this indicator, it will also automatically be selected. So here we use NC one coverage, and then add layer. So this is the map. What you can do is to right click here so you and then you can drill down and up so if you want to see how is this situation within this district you can click drill down. And it will only show the chiefdoms in this district and you can continue down all the way down to facility level and see their performance. And then we can drill up again, all the way up to the level where we were at the beginning. And next we can change to automatic legend. So we edit the layer, go to style and select automatic color legends. And then we can select between equal intervals and equal counts, which I explained. Try equal intervals, update the layer, the map will look like this. The gaps like the size of each of these five classes should be the same. And then you can see the count how many all units are within each class in, in the behind it. So there is only one in the highest class. And as you saw also in the predefined legend, I guess if this is percentage, these numbers look wrong. So they said that everything about a certain threshold was colored green. So that is not something you get when you use the automatic legend that that will take the minimum and the highest number and then divide the classes in between. So at least now we see that there is one in the sort of, I guess, best performing class. And if we change this to equal counts, there should be an equal number within each of these classes, but the size of this class will be will be different. So you will see it doesn't divide evenly up. So there is some with three and others with two. So then we want to see this as a timeline map. So right now we are looking at the last 12 months. If we go to period, if we reduce the numbers a bit or the months to only the six months. And then if you select timeline, it will take each of these six months and make individual maps and then add a timeline. So we select timeline, update the layer, and you will see month by month down here. And then you can click on what you will also see is that there are some months where there is no data for some or units and they will show up at black fields. If you prefer to have this as gray, for example, you can click edit style and then show no data. And then you can select the color for these when there is no data. So we take the default gray and then update the layer. And then this will still show on the map, but with no data. So with this timeline, you can click on the play button and it will basically loop through the different months and you can see the changes. And you can also click directly on the timeline to see how it differs from one month to another. So this is one view to see changes over time. The other method is to go to period and select split view maps. And then instead of showing one map at a time, this will split your screen and show all six map at the same time. And these are also synchronized. So when you navigate the map or even if you drill down one level, it will all happen to the same same maps so you can easily compare them all the way. So and personally, I see that quite a few of you actually like the timeline view better. My preference is this of course it makes the map smaller but here you can because you can see the maps at the same time from different periods. I feel it's easier to compare and contrast the different methods, but you have both options and and please use them. The last exercise was to create a bubble map. So let's go to new. And then use instead of an indicator you should use a data element. So we'll select immunization. Long list here. And then BCG doses given. And then as this one is selected by default, we might change this in future version. So if you selected data element bubble map will be the different choice. But right now you need to remember to switch here. And then you can decide to have the same color legend as before, or you can just use a single color. So I will do that. And then add a layer and then this size of the circle will be in proportion to the number of doses given. If you click on it, you will see the value. You can also change the size of this so you can increase up to 50 the radius. So the circles will be bigger so you can see it also depends on your screen size what what will fit. So that was the exercises for thematic layers. Good work. And then I will hand over to Austin. Thanks. I'll show my screen. Okay, everyone should be able to see my screen now. And hopefully you see the presentation. So I'm going to go over the next layer type that we have in the maps application. I'm actually going to go over two in a row, but spend most of the time on on the one that probably most people have more data for and that is for the event layer. The events are for viewing individual things that happen. So an event happens at a particular place. And each event corresponds to a single point on the map for where that event took place. And we'll talk about how to use the maps application to visualize that kind of type of data in a moment. So we have an event layer we have just as we've created the other layer types there's an event option when you add a layer. And the main thing that you'll need to select when you're creating an event layer is a program. You need to select which program you want to visualize. In this case, I've selected inpatient morbidity and mortality. There's a number of people that have passed away in a hospital, for instance, or in a health facility. And there are some other options here we won't get into too much but there are other options for selecting what type of what field of that data that you would like to use to determine the location. In most cases, there's a location associated with the event itself but in some instances you might have your location data somewhere else in the data element for instance. There are additionally a number of options that are very similar so you'll see that this is now the we're looking at the last tab for the event layer creation dialogue. In this tab we have options for styling the event layer and we'll get into that in a minute. But the other tabs here the period org unit and filter are similar to what Bjorn just demonstrated for the thematic layer so I'm not going to go over those as well. You can select the period, the organization unit that you would like to view and the filter to determine which data you want to you want to show and which you want to exclude from those events. For styling the event layer there are a number of options and the main one on the on the left there with the two images is to select whether to group the events together into circles or what are called clusters. Or to show all of them on the map at the same time. And there are a number of reasons why you might want to choose one of these versus the other one of the exercises is to think about that and figure out what why you might want to use grouped events versus viewing all events. But one thing to keep in mind is that in 235 the rendering engine for the maps application which means the way that they are drawn into your browser was changed to make it possible to view many, many more events at the same time. So if you're using a version that is before 235, you might have trouble if you have a lot a lot of events, and you select to view all events option here. In the 235 it should be possible to download a large number of events up to hundreds of thousands even and still show all of those in the screen on the screen. And but if you're using before 235 that might be slow and might be might be hard to use so just keep that in mind when you're going through the exercises. There are a number of other options here the the second main one is styled by data element, which is on the right side, and we'll talk about that in a minute as well. But this allows you to basically select between different based on a data element for a particular event program, you can select whether to color those points on the map, one way or another. So in this case we've selected gender, and we're going to show all of the male events that are associated with a male in blue, and all of the events that are associated with a female in orange. There are many more sophisticated ways to use this style by data element, this is the most simple because it just has two options and two colors blue and yellow or blue and blue and orange, and we'll get into how to use a legend for that as well later. And then we'll you'll note that you can also if you're not using style by data element, you can select a color for all of your points. And you can also assign a radius or a size to those points as well. And the buffer option here is similar to the one that was shown for facilities. You'll notice that it's grayed out here and that's because we've selected group events. So if you select group events, you can't use a buffer because the events are grouped. But if you select view all events you should be able to assign a buffer to each of those points. If you select a style by data element that is something other than a Boolean so it doesn't just have two options or two or three specific options, it's not a not necessarily a Boolean but has a certain values so in the in the previous case we had two options one is male and there's no male and there's no number associated with that. However, if you have a number associated with a particular data element, for instance, age and years so this is the age of the, the person that was recorded for this event. So you can then choose a legend in the same way that you would with the thematic layer that that you aren't demonstrated. You can choose similarly a predefined legend which is defined in the maintenance application. We won't get into too much today but if you have a predefined legend associated with a particular data element or type of data, then you can use that. And you can also create an on the fly or automatic legend, which has classifications of equal interval or equal count, similar to the way you define that in the thematic layer. This is an example of a large set of data this is more than 100,000 events for the malaria case registration program in the demo database. And you can see here that we're also doing style by data element and have selected view all events. So we're actually showing all 100,000 events at the same time when you're looking at the entire country of Sierra Leone. And we're turning changing the color for each of those points on the map to be either blue or orange depending on the gender of the person that was recorded for that particular event. Remember that this might be very slow before 235 so would recommend not using this view all events option for very large sets of events, if you're not using 235, but you're welcome to use the demo instance to test that as well. If you select group events, we have what's called donut clusters which was introduced, I believe in 235 as well. And correct me if I'm wrong on that. I believe those 235 I'll go back and look at the look at the chart. And this is similar to the groups that we saw in the first slide where you had a number, which is the number of events that happened in a particular region. This allows you to see from from the national level you can kind of see the groupings of events in different areas, but you don't see the individual point of exactly where each of those events happened. And this is much more performant in situations where you have a lot of events. So it means that if you have 100,000 events, you don't need to download all 100,000 and show them all on the map. You instead get a much lighter amount of data that you need to download to show where in each region you have a group of events and how many of there are there. You can do this in all versions back to maybe even before to it was before 230 that the grouping was introduced. But the style by data element being applied to a group of events was introduced in in a later version in 235. And that basically means that instead of just seeing a black circle with 215 in this example, you see 215 but with different colors around the outside and those colors represent the legend that we see on the left of our screen here and shows of these 215 events, how many are associated with a person between zero and 15 years of age between 75 and 90 years of age and all the other categories that are associated with with our legend. This is called donut clusters and we'll we'll get into a little bit more about that later. This is just another example of a predefined legend with a large number of events and style by data element, sorry, an automatic legend with style by data element and a large number of events and in this case we're viewing all events. So I mean these two that there's a bit of a difference in what type of what what you're able to see from a from a very high viewpoint so looking at the entire country, you can see different information or interpret different information when you see all the points individually versus when you see all of the points grouped together. Another feature of the events layer is the data table, which is similar to what Bjorn introduced as well. But it's a bit different for the events layer because you have a lot of individual events rather than clusters for each organization unit or sorry, thematic groups for each organization unit. So you get to this data table by clicking the dot dot dot next to the layer on the left side of your screen and clicking on data table that will open up this data table down on the bottom half of your screen. And then you can, as similar to the thematic layer you can or the boundary layer. Sorry, you can type values into these search fields and filter the data in the table, as well as on the map. You can filter here you can also see the information that is being associated with each of these events. So for instance you see the age in years for each of the events that are recorded and what what the age and the gender of that person was who who was the recipient of this or sorry was the exhibiting malaria symptoms in this case. So that's a quick overview of the event layer will have us have an exercise that goes over both event and a little bit of tracked entity layers. A lot of times there's, you need to be careful because you in order to use an event layer or attract entity layer, you need to be collecting point data, or point or polygon data so basically a geometry or a location for the events and the entities that you're entering into your system. So in some instances you might not have any data for the location on these events so it might look like there are no events in the system. But that could be just because you're not collecting the location where that event is happening in your event program or in your tracker program. So if you don't have that that information of where that's happening, we can't show it on a map, and you might need to change the metadata for your particular DHS to instance, in order to collect that information. The second type of layer in this kind of group is tracked entities. It's very similar to events and you can, it can look a lot like an event layer. However, it's going to be using tracker data rather than event data in DHS to and you're going to use tracker programs instead of event programs. And you can see that you select similar things in this in this dialogue, but there are a couple things that are different. This is what an event layer looks like or sorry attract entity layer looks like it's very similar to an event layer in many ways but you're actually viewing tracker tracked entities rather than kind of anonymous events. One thing that you can do in the tracked entity layer that you cannot do an event layer is to visualize relationships. And this is a relatively new feature so it's it's not. There are definitely more things that we could do with the relationship in the future, but I'm going to demonstrate what can be done with the tracked entity layer today. And this tracked entity layer with relationships, I believe was introduced in 230 and relationship maybe we're out in 231 so I'll go back and look at that slide again to see which layers, which which versions were available, had this available. So you go to the relationships tab and you select display tracked entity relationships, assuming that your tracker program has a relationship associated with it or relationship type associated with it. You can select that relationship type, and then you will see not only all of the individual tracked entities, but also align connecting the entities that are related to each other. And this works quite well within a single program so if you have a, for instance, an index case associated with a contact or a positive case associated with a contact in the case of coven 19, or in the case of malaria, or other other types of programs you might have other types of relationships. And you can visualize this also across programs so you can see, for instance, from a malaria focus area which is not a person but it's a focus area that is a tracked entity modeled in DHS to in this case. And that might be associated with several cases of malaria, and you can see the relationships there as well across different programs so malaria entities are a one program in DHS to in this case, and focus areas are another program in DHS to. So I'm going to introduce the exercise here today and you'll have again 25 minutes to look at this because there's quite a bit more to do with event layers than some of the other layers. And then you can again use your own DHS to instance or use the one on the workshop, provided by the workshop, which the login details are there. And that is a 235 inch instance which you will need for exercise part four. And then maybe part two of this exercise, because those part four requires that you have 235 and part two, if you have a large number of events you might not be able to view them all in 233 or 232, because it might be very slow. So if you are looking at a very large number of events and you find that it's very slow. I would recommend using 235 instance that is provided with the Academy as well. To solve the this exercise. First we're going to create an event layer from one of the programs in your instance. I would recommend trying to select a program with a large number of events. And you can use malaria case registration on the demo server which as as we've seen has about 100,000 events associated with it. First, I would recommend adding the layer first you're going to add the layer with the default settings, which has clustering enabled. And then I would like you to zoom in and out on that map to see how the clusters change as you're as you're zooming in and out of the map. Then for part two you're going to edit that layer and change from viewing the groups or clusters of events to viewing all the events at the same time, taking into remembering that this might be slow on versions less than 235 to 34. And so you should add. Yeah, should use it the demo instance if you have a very large number of events. And there are some questions there that you can respond to as well. And then part three is to create a new map with another event layer and select a program this time which has some interesting data elements. If you're using the demo instance you can use inpatient mobility mortality, and then use the select by data element feature style by data element feature to update the colors of the different points or the different events in that layer. And try both view all and group events to see what is different for you. I would then try try the same thing with a different data element or legend and notice what changes. And part four here is going to be to from that same late layer open the data table, filter by one of the columns and notice how the map data on the map changes. Try to create a map that is meaningful so it has some meaningful filtering some meaningful style by data element selected, and then post a saved map or a screenshot of that. Map that you've created to the sorry this should be day one exercise for channel I will change that in a minute. Then finally create if you have a tracker program with geometries and relationships create a tracked entity layer and display those relationships on the map. And you can use the malaria case registration and the index case relationship type on the demo server. So you have 25 minutes for this exercise, I believe. I think that's about right. And feel free to go ahead and get started, I think maybe 20 minutes. And we will be back in a bit feel free to ask any questions you have on slack. Thank you. Byron did I miss anything there. No, I think that was fine. I see some are saying that it's the they don't can't access the data so just remember that for the event layer is only two of the programs where there is actually data. They impatient mobility and mortality. I think it's cool and malaria case registration. So use one of those two. Good point. And there may be cases in your own instances where you need to select different org units or different periods to to find a program with data that is meaningful. And also maybe make sure that your, your DHS to user has access to that event program and the data for the particular or unit that you're looking at. So thanks a lot for the exercise and let us know if you have any questions and do this here. So here we have the event, the maps application in the demo instance, I'm going to go ahead and refresh my page here to make sure I'm logged in. And I am logged in as the admin user, which we is the user that we shared with everyone here today. If you created another user or you logged in as a different user or you're on your own instance, you may experience different behavior with the events layer because of the permissions for that user. But we're going to go ahead and do this as the admin user today, and the admin user for a short time today also didn't have permission for some of these programs which was changed so thank you for the person that went in and fixed that. And you should all now be able to use the admin user to create an event layer with these programs. So I'm going to go ahead and create an event layer here the first part of this exercise if we go here today. Sorry. The first part of this exercise was to create an event layer from one of our event programs, and select one with a large number of events, and then add let that with the default settings, and, and then zoom in and out so I'm going to go ahead and select this malaria case registration, which we know has a lot of events. I'm going to keep the default settings here for grouping grouping those events and visualizing them on the map as clusters. So here we create that layer. You see on the left here we have this malaria case registration layer for the last 12 months. We have lots of events here you can see that these are all have a K or most of them have a K at the end of the number. If you're not familiar that's that stands for 1000. So this means 2.4 thousand events are in this cluster, whereas this one up here at the top has two events. So you can then zoom in and out. So if I just to make this very clear I'm going to use the zoom tool on the right hand side, and I will zoom in. And you can see that the clusters change. So, as we zoom in. If we still had the just the clusters that we saw at the highest zoom level or the looking at the whole country. It's very helpful because you would zoom all the way in and you would see still see two and a half thousand or 10,000 points that are all kind of in in a big group. So as we zoom in these clusters get smaller, and we see more and more detail, and we see more and more where these events are located in this particular program event program. So if we zoom all the way in here, eventually, we'll start to see these individual points so we'll still we still see some clusters. These clusters are much smaller so instead of thousands of events in one cluster we have 235 that type of thing. And then there are ones without any number and these are the events themselves. And so as you're zooming in and out you can see this, these clusters changing. So that when you're at a high level you can see the the general trend of where the big groups of clusters are. And when you zoom all the way in, you can see the individual points. One thing to note also which is another feature of the event layer similar to some of the other layers but you can actually click on this event, you don't right click you just left click. You can see that this has location the age and years the data elements that were collected when this event was was entered here as well. And yeah, that's that's it for that first part of that exercise. I'm now going to jump over to one moment. I'm going to do the second workshop or second exercise of this set of exercises here, and that is to change from group events to view all events. And so that we can just see what's what's different between group group events and viewing them all. So if I go back to this year. I'm going to go ahead and click, edit. So on the left here we have the edit button you can also click the dot dot dot and click edit layer. Going to go to the style tab, and I'm going to change from group events to view all events. That's the only thing I'm going to change here. See, as I mentioned earlier, when you're in the group events, you can't add a buffer, because a buffer on a group of events doesn't make a lot of sense. But if you click view all events, then you'll see that you can actually add a buffer with a radius for those events as well if you would like. We're not going to do that today so we're just going to change from group events to view all events, click update layer. This is much longer, you'll notice so that was something that a lot of people saw right off the bat is this takes much longer to load. This is again viewing 100,000 points at the same time, which could be very slow in some browsers, but from 234 and I said I said incorrectly previously that this was from 235 but from 234. This is a feature called WebGL, which makes this much, much faster and easier to see so many points on the same map. So if you zoom in and out, it may look kind of laggy on or slow or jerky on your zoom screen, but I can assure you that this is very smooth, and you should be able to visualize that in your own browser as well. You can also see from this that we load all of these points once and that takes a little bit of time, quite a bit of time potentially if there are a large number of events, but once they're on the map, it should be easy to zoom in and out and see the data for these individual points as well. So I'm going to now go on move on to the next layer here, some people are next exercise, some people were mentioning the part of the exercise that said why would you want to view events in this way versus viewing them with the grouping turned on the clusters. There are a number of reasons that you might want both so in this case you can see the kind of groupings into an area, maybe in a little bit of a better way than if you had the numbers in the clusters, but it's much less performance so it might be not as fast. And also, it's, it's hard to see how I really how many points are there are over in these areas where there's lots and lots of different points on top of each other. So when you have clusters you can see that there might be 5000 in this area, rather than just seeing a bunch of points on top of each other and not knowing if that's 100 or 5000. So the next part of this exercise we are going to create a new layer with a I'm going to go ahead and create a new map so I'm going to file new. I'm going to add a new event layer, and I'm going to select the inpatient mobility and mortality program. And I'm going to select style by data elements so first I'll keep the grouped events here and I will select. I'm going to create a height in centimeters I think this one has some data. We're going to do an automatic legend off the bat with five equal interval categories, going to go ahead and click add. And you'll see here that there actually are some some bad data in this program. So it's important to recognize this in your own program programs are in your own databases that this is what what's called no island that Bjorn mentioned earlier today. And that zero zero the point zero zero so somebody when they were entering this event data in this program, they put zero zero as the coordinates. But that's obviously not where this event actually took place there's there's nothing there it's in the middle of the ocean. So it's important to note that there might be some data issues in your programs if people are entering the coordinates incorrectly. There might be some cases where someone put switched latitude and longitude when they were entering that data, and they. So it ends up instead of in Sierra Leone. It might be somewhere on the other side of the, this central point, which might be somewhere over in Greece, for instance. So for now we're going to look at just the the points that are in Sierra Leone. And we see that we have a inpatient morbidity mortality base, we're disaggregating based on height, which may not be the most interesting thing but maybe there's some correlation between morbidity and height that we that we want to investigate. So we can see that the vast majority of people in this region, for instance, how are in the 101 to 131 centimeters category which maybe is maybe is incorrect but that this is demo, demo data so it's been kind of created and isn't doesn't reflect particularly. But if we look at this we can see that this is has a larger proportion of the people in that particular height range, and we can zoom in and see more information and click on individual points and get data about the. Let's look at this one here, and get data about the height and centimeters of this particular case. Now we're going to move on to the next exercise. And the next exercise was, oh, we sorry for this for this exercise as well we wanted to also see the same data but viewing it with all events rather than as clusters. So we're now going to look at this with all events. And here we can see the individual height of each of these events rather than at viewing them as clusters. Let's change the legend and see what changes there. I'm going to go ahead and edit this change this to a predefined color legend with height and centimeters as the predefined legend set update the layer. And now we have well defined zero to 100 and then 100 to 120 etc etc, which allows us to have a better a better disaggregation of these height. Next up we are going to open the data table and view all of this data. So let's go ahead and click dot dot dot open the data table. You'll see down here at the bottom we have the data table here now. We can see that we have not only the color but also the age and years the height and centimeters the mode of discharge etc, of these particular people. And we can do some filtering on this data so for instance we want all of the people that are over 120 and centimeters in height, or let's say over 150. So these are all the people that are 150 centimeters and older over and the where they exist on the map. And because we're looking at this data without any clustering we can see all of those points. And if I make this even more restrictive if we have only the people that are over 180 centimeters in height, or the people that are over. There are no people over two meters apparently in this demo data, but if there was people over 180 centimeters in height, we can see that there are not as many as there are total values here. So this might be something interesting if you want to see the number of people with a height and centimeters over 180 in this impression morbidity and mortality program. You could save this this map and send it someone. Now we're going to look at the last part of this and exercise which is to look at the tracked entity layer. Yes. Could we take this in the Q&A section as we're running a little bit out of time. Sure. Yeah. So the, yeah, the last bit was the tracked entity layer and hopefully some of you got a chance to look at that and we will cover that in the Q&A session. Go ahead, Bjorn. I will share my screen. Yeah, thank you. So since the Q&A section is optional, I just want to finish off today, the compulsory part, and then I hope just hope most of you will join us for the Q&A session. We still have an hour to answer your questions. And we will also cover some of the Google Earth Engine and the new stuff in the latest maps in the Q&A section. But for now, this is the quote of the day. So to prove that you have been taking part of this first day of this workshop, you need to write this down in the form that is shared in the announcements channel. I love DHS too. Okay. So to cover, say a little bit about Google Earth Engine layers in the Q&A session just after, we will also have a full day working on population data tomorrow. So that will be very well covered. So a brief summary from today's workshop. So you should have seen that it's quite easy to create maps from your own health data. The next thematic layer is the layer you should use for aggregate data and then data about individuals you have event and track entity layers. Then I will show you in a few minutes for those who are still here, how you can aggregate data from external sources in the Q&A section. So quite a few of you have used the data table and I think that's a very powerful way of quickly being able to filter the data as the map is instantly updating when you are adding the filters. So it's a quite easy way to sort of drill down and see patterns in your data. For example, we have not covered this because again, I didn't want to destroy your instance so we haven't been saving and adding maps to the dashboard, but you can do this as you already can with your other visualization types. So please test this yourself on your own instance. When you create a map, remember to save it and add it to a dashboard. And tomorrow we will look into how you can export data from the Maps app to other applications. And as always, check the user documentation. We try to keep that up to date for all new versions. So please use the user documentation. Tomorrow we have a very exciting setup. So we have the director from WorldPop who will come and present their data sets, which are very high detail down to 100 by 100 meters population data from for most of the world. And what we will do is that we will import this data together with your organization units from your own country or health facilities to a program called QGIS. And then we will combine these two layers and check how many people live within your org unit boundaries or around a health facility. And if time, we will also look at driving distance because just having a buffer around a facility is maybe not so interesting, but you can have even a driving or walking distance. See how many people that health facility is covering. So for those who are leaving us now, welcome back tomorrow. We'll meet again at the same time, 10 o'clock Central European time. And the rest of us will have a five minute break. And then we will meet back here on this same zoom for Q&A. And if you already know some questions, you can feel free to ask them in the questions channel on Slack. So we'll meet again in five minutes. Some of you are the rest we will see. Thank you and see you again tomorrow. Thank you.