 Hi everyone. Welcome back. So let me start the next session, which is on program indicators and how to use program indicators for data analysis. But before that, let me share the word of the day, which is spike wax, again, another vaccine name. I'm also copying this text to the chat. So today we are going to discuss under the program indicators on how to use program indicators for analysis of tracker data. So mind you, this is not a session which will focus on how to configure a program indicator, because that will be taking place separately in our tracker configuration academy. But here we will discuss how to use the existing program indicators that have been configured into your system or tracker data analysis. So the objectives of the session, we will describe what a program indicator is, and we will also describe how program indicators are derived. We will also understand the difference between the event and enrollment program indicators. Then we will try to understand how program indicators can fill tracker data analysis gaps which are present in other visualization tools. For example, if you can remember, in the event report I mentioned yesterday, there are some gaps. And we will discuss how program indicators can fill these gaps. Then create visualizations using program indicators derived from tracker data. This is what we will be doing during the demonstration. So program indicators are computed variables based on data elements attributes or constants used to aggregate individual level data. So as I briefly mentioned yesterday, we are using program indicators whenever we want to calculate or compute something which might be a combination or which can be a standalone of a variable based on data elements attributes or constants. And these are mainly coming from tracker data. So probably like to give you an example, like if you have a, if you want to kind of count a number of track entities who are having a disease, right? We can do that using a program indicator. For example, let's look at this example. We can create a summary count of the number of AstraZeneca first doses administered by going through this program and counting the events which meet this criteria. So for example, when we look at this table below, we will see like here we can see we have a column for dose number, right, which mentions the dates of first dose and second dose. And then we also have the last column as vaccine name. So because we have this as a line list, even using the line list, we can count, right? So for example, if we use the filter for that data element, we can count. But like if we want to get this as one single value, we can use a program indicator to get a calculation of say the people who have received AstraZeneca vaccine for the first dose as a single value. So this calculation happens automatically without we have to do anything like what we did yesterday using the event report. So we don't have to use the analytic tools in DHIS to the analytic tools apps in DHIS to make this calculation. So that's what the program indicators does. It does the calculation automatically and then give us the exact value. And in some ways it is similar to using the pivot table output style in event reports for some of the functionalities or displaying options within the chart in event visualizer. So for example, yesterday when I was doing the event reports and today morning when you did the event visualizer, there were like some of the tools which are available in those apps, which can substitute that previous example I showed you, right? So there is a bit of an overlap, but this program indicators are more powerful and we will discuss like what are the gaps program indicators can feel, which cannot be met by the existing tracker and its analytic tools. And this program indicators are more flexible because they allow us to fill several limitations. We have covered when we were trying to analyze tracker data directly using event report, event visualizers and maps. So let's see some of the limitations that we faced while we were using this analytic tools such as event report, visualizer and maps. So for example, when creating summary data, we had difficulty in combining or displaying data from multiple stages together. So for example, we were not able to create enrollment type pivot tables in event reports. So in event reports, we could create enrollment type line this, but we were not able to create enrollment type pivot reports, which are pulling data from multiple program stages, right? And then also, we could not create enrollment type visualizers visualizations in event visualizer and maps. So this is one gap, which we had in the existing analytic tools, which the program indicators can fill. But having said that, there are like, I mean, you have to also understand there are advantages as well as few disadvantages of using program indicators. So the advantages are they offer more flexibility in creating summaries of events and enrollments in tracker data. They can be used in tools, users are more familiar with, say for example, most of you must have used the data visualizer and maps applications in DHS2 because you already use DHS2 for aggregate, right? So most of the users are familiar with those tools. And this program indicators is a concept that you can use in aggregate tools as well, right? Not only in tracker tools. So for example, you can browse through program indicators and do some analysis using the data visualizer. There we have the pivot and also the maps. They offer a number of advanced functionality, including Boolean logics, like if statements, relationship counts, differences in dates, etc. So there are like a lot of other functionalities, which the program indicators can do. But there are some disadvantages as well. One thing is each program indicator needs to be configured and this can be potentially time consuming because like for each, say, combination now, you know, like for example, you might think like we are in aggregate, we have this for data collection, we are having, we are using the category compost so that we can, you know, like if you want to combine two variables and get an output and we want the variables for each of the combination, it is automatically done. But for program indicators, it is not like that. You have to configure program indicators for each combination. So it can be kind of time consuming and initial configuration can be a bit complicated. And users can define a requirement for a program indicator. But once it is defined, you can't modify these filters in real time that you did yesterday with event reports. So if you can remember, in the event reports, we could decide, right, say for example, if we wanted to filter the data pertaining to a particular case ID, we just had to insert that case ID into the filter. But this kind of flexibility is not there with program indicators because these are preconfigured. And if you want to add or change a filter, you will have to inform a person who has maintenance application rights in DHS to instance. So it's not too flexible when it comes to end users. So there are two types of program indicators when we are working with tracker data. So we have event program indicators, and we have enrollment program indicators. So what do we mean by event program indicators? So these indicators are evaluated per event within a particular program stage. So events kind of like calculate each of the events within that program stage, just similar to what we explained yesterday. Whereas enrollments evaluate across an entire enrollment within a particular program. And this is kind of a repetition of what I mentioned yesterday. When we are using enrollment program indicators, it uses the most recent event within a program stage in its calculations. It's just as same as I explained yesterday when we were doing the visualizations with event reports. So let us also compare the event versus enrollment program indicators. What are the differences? So here what we are trying to do is like we are trying to count the number of PCR tests requested. And if you can look at these two examples that are shown in the screenshot, in the first one we have for the stage two, which is an app request, we have one event. So which is highlighted in gray color, we have one event. But in this second example, for the same program stage, we have two events. So when we request for an event enrollment program indicator, the output is going to be different. That is what we want to hire. So for example, if we ask, if we configure an event program indicator, which wants to count the number of, which will count the number of events. So here, irrespective of these two events being for the same person, because there are two events, when we use event program indicators, the output will come as number two. And when we use enrollment program indicator, the output will come as number one. Is that clear? This is kind of recapping what we explained yesterday. Okay. So let us look at the enrollment indicator example. So what we are trying to do is we will try to see number of hospitalized cases with the positive COVID-19 test result. So here, this entire sentence we can break down into several components. First thing is number of hospitalized cases. That's one thing we want to analyze. And it also has a filter. They should be positive, they should be having a positive COVID-19 test result. So if we are going to develop this using an enrollment type program indicator, what it will do is it will combine data from multiple program stages. So it will check here from the stage one, which is clinical examination and diagnosis, which is a non-repeatable stage in our demo instance. It will check whether the person is having yes for the hospitalized data element. We have a data element called hospitalized. They have all the patients who are fulfilling who will be counted will need to have yes value for this variable. And then for the lab result, which is coming from the third stage, there it has to be positive. You can see here it has to be positive. So patients who fulfill both these criteria are the ones which are calculated, which are counted. And here, the difference, because like yesterday, we could not get an output like this. We were able to get an output, which is a line list kind of output. It was a list of patients we were getting, not like if we wanted for this, I mean like here in this table that you are seeing as the example output, the first column is the areas, say like if it is a province. So for the province, how many people were there who work COVID-19 confirmed and hospitalized? This kind of output we could not generate yesterday using an event report because the event report enrollment analytics did not allow pivoting. But here now we can. Okay. Right. So with that, I mean, we come to the end of the presentation and we are going to move to the demonstration. Any questions up to this point? Right. If there are no questions, let's start the demonstration. So what I will do is I will quickly share my screen. So it's okay. So I have logged into our demo instance and what I'm going to do is like for this initial demonstration, I will use the aggregate analytic tools which are, which is basically the data visualizer and the maps application. So let me open the data visualizer application here. And then let us open an existing favorite item called underlined conditions last six months. This one. So when I click and open this, this is what I'm seeing. So what do we see here? In this visualization, we are seeing right people who have received the vaccines and who are also having underlying conditions. Right. So underlying conditions is one. So let us see like what were the data items which are collected underlying conditions. Right. And it is looking at the entire country for the last six months. Okay. So here we are getting aggregate output. Okay. These accounts based on event data or rather tracker data. Okay. So this is something that we were still able to do yesterday using event reports. And you remember like event reports application allowed us do pivoting. Right. Based on a particular program stage. It did not let us do pivoting when we want to combine multiple program states. But this was again something possible yesterday using event reports. But here the difference is we are trying to design this table using a pivot in the aggregate data visualizer application. Okay. So that's what we are trying to do. Okay. So let me demonstrate you how we are going to do this. So first of all, what we are going to do is to click on this new. So everything refresh. And now I have to select the parameters. So first, what type of visualization? So I want to create a pivot table. I assume I did not explain too much about the data visualizer application because this is something you already know from our aggregate training programs on aggregate analytics and the DHS to fundamentals. Right. So we have selected pivot from here and next we have to select the data dimension. So for data dimension, we are selecting the data type as program indicators. We usually we are familiar with selecting indicators and data elements, but we are going to do something different here. We will select program indicators. Right. And then after doing that, we have to select the program. So the program is going to be COVID-19 vaccination registry is one. Right. And we are searching for the program indicator all underlying conditions. This is one. So here once I selected this, I have to select the period. So let me select last six months from here. Done. And organization unit. What I'm going to do is I will select the country, but I want the country to be selected at level two. Right. So that I'm not looking at the entire, not only looking at the entire country, but I want the country data, which are disaggregated at level two, which is a provision. Okay. So after doing, so I will just click on hide. Then let me see whether I have all the layout set properly. So if you can remember the previous layout, so like, is this the same thing that I'm going to get? Because there I can remember the periods or the months were appearing as columns and the organization units were appearing as rows. So here it is not properly configured. So what I will try to do is I will try to move organization units to the rows and periods to the columns. After doing that, let me click on update. And here we are getting the same table that we, that I showed you before. Okay. So we didn't try anything too new. Only difference that we did compared to our usual aggregate analytics is to select the program indicator, which has already been configured, as opposed to selecting a data element or program or indicator that we usually do when we are using this aggregate data. But one thing I have to mention is like when we are trying to do this, okay, they have selected enroll type of a program indicate it's not an event type of a program indicator. Let us see whether, I mean, what will actually happen if it is, if you get as an event type program indicator. Okay. So to do that, let me duplicate this tab. And let me quickly create, let me see what is there. It's not been selected. So let me quickly select the program indicator. I will select COVID-19 vaccine registry. And let me search for the underlying conditions. We have two programming. The one at the bottom is the underlying conditions for events. So I'm going to select that because previous I selected the other one. And the periods, let me select last six months. The organization units now at level two. Okay. Done. Let me rearrange. So with your units here, period here, and I click on up here because it has to be preventable. I click on update. And this is what I'm getting. Okay. So if you can compare these two, this is the enrollment type. And here we have the event type. You can see, let us focus on the April column. So here for the VNT, we are having for the April, in this first visualization that we initially developed, it has 23 as the value. Whereas for the second one, which is based on event analytics, the value is 33. So it is always the values that we are getting in the sales in the event table is always more than what we are getting in the enrollment table. Why does that happen? What is the reason? Is my question clear? So I'm comparing these two tabs. Here we have the enrollment. Here we have visual events. So the values that we are getting in events is always more than the values that we are getting in this table for the enrollment. What's the reason? Anybody? Yes, Amit, you have unmuted. It's a very simple question anyone wants to answer. So in enrollment, one person count only once as it is repeatable stage. So when that person get multiple service or repeatable stages, then it's called multiple times in events reports. Right, exactly. So it's like this, because here the enrollment, like only it only takes into consideration one event from a program stage. Whereas in the event type of a program analytics, it will count all the events which are present for a given program stage. So that's why the final calculated value is going to be more than what you always get for enrollments. So hope that is clear. But one thing we have to keep in mind is what I have showed you now is something that was even possible using event reports yesterday. So let's see certain things that we were not able to do yesterday. So now one, yeah, so let me like keep the same example and let's try creating a chart because we have seen how the tables work. Let's design a chart using a programming so that I can take this forward and explain you a few more concepts. So what I'm going to do is click here, file, new it. And then I will select a chart and let's try to open a chart whole symptoms and health out. This is the one person with CBS symptoms and health outcome this year. Okay. Right. So let's see what is getting displayed. What can you see here? So here in this chart, it's a line chart. That's the first thing. And on the x-axis we have what we have the individual. And then we have on the y-axis the numbers. So here we are looking at COVID-19, the patients who had COVID-19 symptoms present. And then on the blue line, it is about COVID-19 symptoms present and resulted in death. Red is COVID-19 symptoms present and recovered. So to produce this, we are using program indicators. And what can you think of the architecture for what constitute this program indicators? So are these program indicators? So let's take each one of them. What can you say about these three program indicators? Are they taking data from only one stage or more than one stage? These three program indicators. All the stages, all three of them? I mean, what do you think? So here I'm not sure whether the answer is for this one. Like all three program indicators are varying from all the program stages. Is that so? Anybody else? Right. So it's like this. Yeah, Arif, you want to answer? Sorry, my internet was not set. Yeah. You can also type in the chat. It's no problem. Okay. So what I wanted to highlight is like here, especially these two, the symptoms present and dead, as well as symptoms present and recovered, the symptoms present is coming from one stage, right? Okay. Which is the stage one, which has clinical exam and diagnosis. And then this health outcome, it though recovered is coming from stage four. So they are for these two data, for these two program indicators, these kind of combining outputs from two program stages and try to plot this. Okay. And we are actually getting here an aggregate output. So this was not something which was possible using event reports because we are looking at aggregate outputs across multiple program stages. So this can only be done through a program indicator. Is that clear? Right. If so, let's try to quickly design this visualization. So I'm going to click on new. Okay. And from because I want to draw a chart, right? I will select a chart and the chart type is line. And then under data, I'm going to select the program indicators program is going to be case based surveillance. And I'm going to search for the, for the three program indicators. So I will just type symptoms. Okay. Here we have the three. I will select symptoms present. I will click symptoms, present and death and symptoms present and recover. And then the period currently is last 12 months. I want to have years this year. And the org unit, what I want to do is I will select the country, but then I scroll all the way down and filter the entire country and get the data disaggregated by level two, which is provincial. Then of course, I'm going to change the layout a bit because I wanted all the org units in X axis. I'm going to move it to categories. The period, of course, I don't really mind where I don't want it in the visualization. So I move it to filter the data I want to see. That's it. And I click on update. And this is the visualization that we saw before. Okay. Is that clear? Let me also do something else from here. What we can do is we can like, because now we have a chart, let's try to see from the same data visualizer. What happens if we made this pivoting? So I'm going to select the pivot here and click on update. Let's see what we can. There we go. So we can, from using the data visualizer, we can easily change the line chart to a pivot table just like this. So now that we have the individual figures in a very descriptive way, and we can even download this. So we don't actually stop there. We can also do something else, which is we can convert this visualization, which is now a table to a map. So to do that, what we try to do is we can click on this icon here and click go all the way down and click open as map. So when we do that, it'll try to open this as a map, but one issue we are having is it will ask like, there are three data items in this visualization. Do we want all of them or only one of them to be displayed as the map player? The issue is because all these are coming from same org units, even though we select all three of them, it'll overlap on one another. So we won't actually see it properly. So let us only select one of them. So I will select dates, and I will click proceed. There we go. By default, what it will do is it will convert those data and put it as a thematic layer on DHS to maps application. So using the same application, we can actually shift between the chart and the pivot table, and we can transform that data to a map as well. I hope it is clear. Any questions? If there are no questions, we'll quickly try to do the exercise one, which is in the learner's guide. Let's try to take five minutes to do that. If you are not able to finish it, it doesn't, I mean, like it's okay. We will try to go to the move forward in the demonstration, and you can practice it later. So let's take five minutes and try to do it. There's a question. We have subunits and levels, but is the difference? So it's like this. I mean, it's more or less the same, but subunits we are defining based on the logged in user. So for example, it's very relative thing. So if you configure the user to be at district level, and we have saved the visualization to use one subunit, so whoever who logs in, he will see one level below. But whereas if we set something as level two, only the people who have access to level two will be able to see, because others, if they don't have access to level two, it will not really work. So if you want things to be very relative, and be flexible based on the logged in user, using subunits is much better, because it'll always work on a level one or two levels below the logged in user. That's the difference. So let's take five minutes and try to finish the exercise one. How many levels you can have? You can have any number of levels. It's not a problem. It's about DHS to configuration. But having too many levels is can affect the performance of your DHS to instance when it is growing large. So you have to be careful about how many levels you want to have in your configuration. Any more questions? So let's meet in around four minutes. Welcome back. I assume some of you may not have been able to finish the activity yet, but what we can do is we will try to finish the demonstration because some of these activities are kind of going back and forth between the different analytic tools available. So probably like if you go through the entire demonstration and then try to attempt it, you might find it more easy or more clear. So let's start the next activity. So let me share my screen. So we are already in the maps application. So let's try to open a already saved visualization which uses program indicators. So before opening the visualization, I must mention now you learned about how to use the maps application for visualizing track entity and events data in the morning. So one limitation was that it was focusing on the individual events and the track entities and the clusters, but it was not actually taking into consideration how certain track entity instances or events can be aggregated and displayed on the map. So that's the kind of gap that we are trying to address by using the program indicators. So let me open one program indicator. So there should be one called the COVID case based surveillance suspected cases last 12 months. So let me click and open that one. It's a bit of a heavy visualization. So it might take a couple of seconds to go. So here we have a maps visualization and this map is of bubble type. So in this legend you are seeing here, right? So if the kind of the size of the bubble indicates how significant the value reported was and also you will be able to see a timeline here at the bottom. So what we can actually do now what you are seeing here is a timeline maps visualization and let me click on this play button and see what happens. So right now in the timeline by default it is showing October to November and when I click on the play button you will see it will start moving and then you will be able to see how the visualization changes by each month. You can see now the the suspected cases are becoming higher compared to what it was in the October. So this is in fact a nice visualization that can come in really handy when you are doing a live demonstration and probably like if you can put it in a dashboard and you can present it to a larger audience. So let's see how to produce this visualization using the maps application. So let me click file and then new and then I need to add because I already have the base map. The base map is coming from this OSM light. So I will add a thematic layer. So I'm clicking on the add layer button and select thematic and next under the data tab I have to select which item type. So the item type is going to be a program indicator which is coming from case-based surveillance program and the name of the program indicator is COVID-19 suspected cases should be some COVID-19 suspected cases this one. And the period I will select as relative last 12 months and then make it timeline. Then the org unit I will select low at level 2. So that's done and then let's move to the style of the map. So I will select bubble map and single color legend so that it will be only one. After doing all that what I have to do is click on update layer click on add layer and then it will design the visualization and here we go we have that visualization. So basically what this map is showing it will be showing the COVID-19 suspected cases as an aggregate value in a dot. So one dot per province that's how it goes for each of the months. In a snapshot we we are only seeing the first month but when we click on the play button it will start kind of first wall clip where it will go on showing the bubbles for each of the months. The size of the bubble indicates the value of the aggregate number. So this all happens based on the program indicator we have defined we have configured the configuration of the program indicator of course is out of the scope of this academy but when it is configured you can nicely use it for your analysis. Right so are there any questions about how to use this program indicator for maps? Again like before you ask this question currently we don't have a mechanism in DHS to analysis like from this aggregate visualization figures so let it be I mean whether it is a map or a chart or a table right for to directly jump into a individual cases record that is not possible say for example if you want to like once you click on this dot to have a link where you can open individual records no that is not currently possible. So that's one limitation we still have in DHS too which is of course a nice to have a feature but of course it's kind of a very complicated design process. Any questions? If not we can do the exercise number two in the learner's guide right and we can take around five minutes to do that and we will meet in five minutes. Right welcome back so we are moving on to the later part of my presentation so in the in the in the last section what I want to highlight was a concept that I emphasized in my previous presentation that is about the program indicators can work at individual level to show a calculated value. Say for example if you are obtaining a list of people in a table we can use program indicators for one column to get a particular figure for individual person that is one. The next concept what I mentioned was if you try to get use the same program indicator which has a aggregate type say as a sum or average something at probably a district or provincial level then that same program indicator will show aggregated value based on all the people or all the persons all the track entity instances which are coming under that organization unit. Is that clear? So what I mentioned was the same program indicator if you mention if you use it at individual level to obtain a line list and if you use it at a higher level district or provincial or at a higher level in your organization hierarchy will produce two outputs because at higher levels it will kind of aggregate values of all the track entity instances whereas if we try to obtain a line list it will mention the values for the individual person or individual track entity instance. Is that clear? All right. Any questions? If not let me try to demonstrate something. So let me share my screen. I'm again back in this familiar application which we did yesterday which is event reports application. This is not a data visualizer. This is event report application. So let's try to open a saved favorite item which is the item is contacts by person. So what do we have here? In this table we can see right we have a list of people right again showing whether that person had the signs and symptoms and it shows how many COVID-19 contacts this person had. So let's see how it has been configured. It has been configured as a line list with output type enrollment. So what happens in that case is like it will be showing list of people which takes into account values obtaining across multiple stages. So under the data section we are seeing two program attributes one data element which is signs and symptoms present or not and here PI stands for program indicator. So it is having this program indicator which is called COVID-19 contacts. So how is this program indicator derived? So even though this is out of the scope of this lecture let me quickly share my other screen to show you how this has been constituted. Okay now what you are seeing here is how a program indicator has been configured in the maintenance. So this particular program indicator which is the name of is COVID-19 contacts is with the aggregate type average and the analytic type enrollment. So it will be taking into consideration the enrollment type of analysis and it will create an average when things are aggregated and what is the expression or what is the criteria this program indicator is using to derive its output. This is the relationship count. So how many relationships which has been attached to the track entity instance. So that is what this program indicator is concerned about. Okay so going back to my other screen. So we are back at this visualization. So I hope it is clear this is showing a line list of all the persons and also showing at individual person level how many COVID-19 contacts that person is having. Is that clear? So if so let's try to quickly design this table by ourselves. So I'm going to click on new. And then from the table style I will move a bit fast because we have already done this yesterday. I will click on line list and the output type is going to be enrollment. And the program is COVID-19 case based surveillance and the program stage I will select the first stage which is the clinical examination and diagnosis. So from that let me initially select the two attributes which are first name, surname and then of course the data element which is whether science and symptoms present. And then let me select the program indicator which is COVID-19 contacts. This is the one. So I double click and put them here. The period is this year and the organization unit is CHW, this is the one. And I click on update and I'm getting the same tip. Is that clear? This is what we did yesterday. Now a few other things we can do of course we can like sort this table based on I mean we just have to click on the column header so it will be sorted based on the value of that column. And let me also add another program indicator to this visualization. So what I will try to add here is we have another program indicator which is called case between onset and consultation. So I just double click that and I add it to this selected box. Then I have to click on update. Now we are seeing in addition to the previous columns we have an additional column which is the onset between the consultation and the days between the onset and the consultation. It is also in this table. Now what do we see in this table? Here the outputs are concerned about each of the track entity instance. So each of the person. So it is showing say for example this Ryan Robinson. He had one COVID-19 contacts and the days between onset and consultation is zero meaning he has got admitted on the same day. But whereas this Timothy he has one contact and the days between is six days. So likewise here in this table it is showing outputs at individual level. So now that we are kind of clear about how a program indicator works at individual level we can see how it works when tracker data is aggregated and we are trying to visualize at a higher level in our own time. But before that any questions? If there are no questions let me move let me open the data visualize application. We will try to get the aggregate values of the same program indicators by using the data visualize application. So click here and open data visualizer and let's try to open the favorite item which is average you have something like that. For some reason the saved item is not showing but let me design a chart for you. So what I'm going to do is like I want to visualize a bar chart which shows the COVID-19 days between onset and consultation the last program indicator we used and see how it is aggregated at the level we are trying to see. So let me design that particular chart. So I will select the bar chart as a chart type and then for the data what I will try to do is I will select program indicators and the program is going to be case-based surveillance. From here I will check this one COVID-19 days between onset and consolidation and then the period I will select years is here then the filter will be low ptr we will visualize it at a lower level. So we will select this oak unit number 13 province 13 and in that one let's try to take all the oak units at the level 3. So basically one level down below this that's it and then let me see whether the categories are in particular order so probably because I'm looking at the oak units I want to have it in the category dimension and period I'm not too much worried because it's just one me I'm looking at and data is what I want in the series. So I'm happy with the layout and I click on update and this is our chart. Now what do we see in this chart? It's the same program indicator COVID-19 days between onset and consolidation but here we are looking at individual organizations health the health institutes or health areas which are at level 3 under that province. So definitely these values we are seeing are not related to one burst so what may have happened? Anybody how are we getting this 2.7 1.9 2.1 how do I mean what are these values? Anyone wants to make any guesses? So average value? Yes that's correct so average value because like if you can remember when I showed you how the program indicator has been configured it has been configured to take average as the concept when it is aggregated up in the oak unit high enough. So what it is trying to do is for each of this oak unit it'll take the values of values which are coming for each track entity instance. So each of the track entity instance if you can remember like as in the table before there was like for each TEI each person we had a value for days between onset and consolidation. So they will you know take consideration of all these value and create an average and display it for each of the oak unit right. So that's what has happened. So this becomes really useful because by using the program indicator if you are at the facility level if you are a facility health facility manager we can get a line list and we can see like for each of the patient what actually happened. But if you are district manager or health manager at provincial or national level you are not worried about individual patients but rather you want to you want to see in my district or in my province or in my country what is the usual days between onset and consolidation for patients right. So for this average becomes a very good concept and that's exactly what it does. So if you just capture data at individual level we are able to create these kind of visualizations at national level. So this is what is capable of okay this is what is THS 2 is capable of doing using programming. Is that clear any questions? What is the number before the oak unit? Yeah this should be probably the code oak unit code because this is just coming from these these are not variable values right. So these are the values that are that are already configured when you are configuring the metadata of the THS 2 instance. Any more questions? Right. If not there are no questions at all I mean like they have a very few but I hope you understood because program indicators is a new concept it might be a bit tricky understanding but I went as slow as possible with the given time but if you have any queries please feel free to ask now or probably in the slack. And if you try to so there's another exercise in the learner's guide that you have to do exercise three. So once you do that you will again get a clear idea of how to do it by yourself. Are there any questions? Yeah Arif you are muted. You can also type it in the chat if your network is bad. Right. So let me quickly do a recap of what we learned under this session. So a quick recap of program indicators. A program indicator is derived from individual level event or tracker data. That's how the program indicator is derived. It could be from event data or tracker data. Then we can use program indicators in data visualizer maps as well as event reports. So you can use program indicators in the event analytics applications as well as aggregate analytics applications as opposed to the data elements and attributes which we are only able to use in the individual or event related analytics. There are two types of program indicators. We have one program indicator which is of event type. This will perform an operation based on all of the events within a single program stage. So that's what the event type program indicators do. And then we also have the enrollment type program indicators which will use data from most recent event and this can be combined with the data from multiple program stages and it will produce an output. And then finally what I tried to show you was that program indicators can function at two levels. So it can show data for one event or track entity instance. That's what I did in the last demonstration by showing you a table. And then it also can show a summary of data for all the events or all the track entity instances within the specified org units and a period. So because you did not ask, I will also mention this aggregation happens based on the org unit and the period. Of course, we can set again when we are configuring the program indicator. So usually it is the enrollment date but you can even change that. So how to do that? You can learn when you join the tracker configuration again. So that's it I have to present for this session. Any final questions? If there are no questions, the graded assignments are available for all the sessions that we did today. Please do this. The assignment should be now visible in your model. And also mark your attendance which is compulsory. And please give us feedback. Yesterday we did not get much feedback. So just a kind reminder if you forgot to give feedback for yesterday's session, please do so. And today's feedback will also be available shortly. Please enable the feedback. Feedback for day 4 is available now and the attendance as well as the graded assignments for the day are also visible. So with that I think we can conclude today's session. It took long because we had to deal with a lot of new concepts and show you a couple of new applications. But I hope you can go through the learner's guide and practice all the steps that we demonstrated today and do the graded exercise. And Saurabh, anything that you want to add before we conclude? So are there any questions? If not tomorrow, again now we have one session that will be about custom applications. And then we will have the final examination. So this is again going to be a multiple choice type of exam just like what you have been doing for the graded assignments. So this test and these marks together with the graded assignments are calculated and will contribute to your final grade. Saurabh, are you there? So we are starting again tomorrow at the same time. Please go ahead. I think he's having some technical issues. Are there any questions? I think Arif, you try to unmute. So if there are no questions, again tomorrow we are at the same time at 12 noon Indian standard time. And we will do two sessions tomorrow. First we'll have a custom web application that's again a question that we usually have. And then we have the final examination, which is the Q type examination, which will contribute to your final grade. So that's it. Thank you for joining. Have a great day and see you tomorrow.