 Hello, and welcome to our workshop on census flow data. My name is Vasiles Routis, and together with me is Oliver Duke-Williams, both of us, based at the Department of Information Studies at UCL. So the structure of this workshop, the first part is on what we're doing now, the welcome and the polls. We're going to move forward with the presentation about what actually are flow data. We'll then demonstrate how to access and download data from the UK data service. And then we're going to have a lab, a practical activity on retrieving data. And then we're going to get some feedback from you about that activity and some general questions and answers about census flow data. So before I start, I'll just apologize because of the interface. I mean, it lacks a bit, you know, the interactivity that we have used recently using other platforms. So you're not able to speak or serve either or anything, so you're only able to type in questions. Oliver will be monitoring those questions, and we will be answering all of them. But still, the workshop will lack some sort of interactivity. And as far as I know, UK data service will cut up with more modern platforms. So the next workshops and webinars will be using more robust platforms to cut up with the latest developments that have been accelerated due to the COVID-19 crisis. So what is the UK data service first? So UK data service is a comprehensive resource funded by the Economic and Social Research Council, ESRC, which is now part of the UKRI, which stands for UK Research and Innovation. It is a single point of access to a wide range of secondary social science data and more importantly, it provides support, training and guidance on how to access and use those data for your own research. So what are census flow data? Sometimes in the literature, you will also find other names for flow data, such as interaction data or origin and destination data. More recently, we're trying to stick with flow data, but we use those terms interchangeably. So flow data, interaction data and origin and destination data regarding UK census is more or less the same thing. So flow data consists of counts of flows between two locations, an origin and a destination. Some examples include migration data to the UK, commuting data, for example, commuting for work purposes, journey to school and movements associated with the second residence. So those flows can be between two locations anywhere in the UK, like in this example from London to Birmingham, or it can be flows within the limits of a city, for example, to go from home to work, like this example in Glasgow. And it can also be flows within the same area because we count those as well, so they don't have to be between two different areas of data. The data are produced at different spatial scales and this makes the processing of such data sometimes a bit more complex. This is an example, a sample of some of the most popular 2011 census spatial scales. So we can see that at the bottom of the pyramid, we have the output areas which consist of more than 200,000 areas. Then we have the lower layer, super output areas with more or less 40,000 units, wards, almost 10,000, middle super output areas, 9,000 local authorities, 400, and there are some others like regions or countries which simply consist of England, Wales, Scotland, Northern Ireland. And in most cases, you can aggregate from the lower to the higher spatial scale. So Flowdata website offers some lookup tables that you can do that, but sometimes the boundaries of those scales are overlapping. So in that point, there is a best fit scenario to mitigate the problem, so not all scales are compatible. For example, wards, NMSOAs are not 100% compatible, so you need to have a best fit approach in order to convert one scale to another. So as well as normal areas, which are the spatial locations, flows can also be associated with various aggregates to the spatial areas, such as for example, overseas, work offshore, or work at home, which we anticipate will become much popular in the next census. And one thing to note is that there has been different handling of cases such as this over time. So those pseudo-spacial geographies were treated differently until 2001, different in 2011 in terms of warding and stuff like that. So there's not a 100% coherence between among the different censuses over time. Flowdata, because of the complexity, are one of the last census data products because counts have to be collated from the three different national statistical agencies. So here is the map of the UK for England and Wales. There is the Office for National Statistics, ONS, or Scotland. There is the National Records of Scotland. And for Northern Ireland, it's the Northern Ireland statistics and research changes in ISRA. So the progress of flow data sets. In 1981, things were a bit simpler. We only had the migration statistics and the workplace statistics. Things progressed in 1991, more or less remained the same. 2001, we still have the migration statistics, but on top of the workplace, we also have the spatial travel statistics, which were only produced in Scotland. And for the rest of the UK, the workplace statistics remained unaffected. And in 2011, we have some more splits on the data. So we have the regular migration data, but a part of that also became the spatial student statistics. When someone identified that the previous location, house, was still unrelated. And then together with workplace statistics, we also have the spatial residence statistics, which are related to second residences. And 2021, we still have no idea, but we anticipate there will be more or less closer to 2011, rather than any of the previous censuses. So we think we should be more or less like 2011. So access to data, until all the data, all the census data until 2001 are available to the public. You don't have to register, you just go through the website, login as a guest without any password, and download the data freely. But data from 2011 is a bit more complex in terms of access, because we have multiple levels of access. And there is a trade-off between spatial detail and attribute detail. I'm going to show what I mean later. And there's also different routes to using the data. So this little cloud, let's say it's census flow data, and it's split between four types of data sets, as I discussed earlier. So it's the migration data, the workplace data, the student data, and the second residence data. Then we have this special level, so we can have local authority districts, middle super-output areas, wards, output areas, and workplace zones. And we saw that between those different spatial levels, there's a huge difference on the level of detail about the areas that they are describing. And a crucial thing, the security level for 2011. So we have the public data, but we also have the safeguarded data. That means that you cannot just go to the website and download those tables. You will have to first register UKDS, you will have to be a member of academia or UK government, and then be able to download that data with some restrictions on how you share the outputs. And we also have the secure data, which are not available on any website, and you will have to get a special license from ONS, go there. I'm not quite sure what's happening now with COVID-19, to be honest, but you need to go to a secure lab and download this data. And then we have the attribute detail. We have the headcounts, which are the simplest ones, because they are just the total of people. We have the univariate and the multivariate. I'm going to explain those in a bit. So the various data sets all stem from questions on the census form. So as we saw for 2011, we have four types of census flow data tables. The migration ones are flows between an origin and a destination, and are based question 21 of the census questionnaire, which was a question, what was your user address one year ago? The student tables, as I said, is subset of migration data. And there are tables for people who have indicated that their address one year ago was a student or school address. The workplace tables are related to journeys to work. There are flows between a residence and a workplace, and they're based on question 40. Your main job, what is the address of your workplace? And then we have the second residence tables. So there are several different sets of flows. First residence to second residence, second residence to work. And all these are based on questions five and six for England and Wales only. And that question is, do you stay at another address for more than 30 days a year? And if so, what is that address? So these are the questions that actually built the census flow data. So there are some significant changes on how the date, the census flow data is structured between 2011 and the previous censuses. So up until 2001, the flow data were grouped into sets, like for example, in 1991, migration set one, 2001, workplace level two, etc. So all these sets were like some supergroups because they contained several other tables. And all these tables within that set were defined by the same combination of geographies. So we had an origin geography, like for example, words, districts, etc. And a destination geography. And in some cases, a few additional non spatial categories like the ones that we described earlier. And this was a bit of a more simple approach because you had less tables, a few sets, and then you could select the table within that set. Whereas for 2011 data, we don't have the explicit supersets that contain tables. So we have hundreds of separate tables and its family consists of them. And more important, those tables within them did not have common geography definitions. So that sometimes make it more complex to analyze those data compared to the past. So there are some easy ways to identify a table just by using, just by seeing their ID. As I said, the tables cannot be grouped into sets as easily as before. But we can get some hints about what that table is, but just looking at the name. So the census 21 tables start with the letter, which can be either M, W, S, or R. And that first letter responds to M for migration, W for workplace, S for student, and R for second residence. And then the next letter can be either F, U, or M. And this corresponds to F for headcount, U for univariate table, and M for multivariate table. Then there is a number, usually in an NN format, so it can be 010203, etc. And sometimes they also have a suffix like A, B, C for further splitting the table up. And this number corresponds to what the actual data is on that table. And finally, we have the geographical specification of the table. So this can be UK, England and Wales, Northern Ireland. And I think that's supposed to, it's going to be either UK, AW for England and Wales and Northern Ireland. I'm not quite sure if we have any tables only for Scotland, but in theory it could be, yes. So for example, to break down that table. So we can see that the M corresponds for migration, F for flow as a headcount, the simplest type of table. 02 is the table number, and UK means that that table covers the entire UK, United Kingdom. So as I said earlier, for 2011 census data, we have three levels of user access. The first one is the public one, and the public data can be downloaded through UK data service, but also directly via the Office for National Statistics, or the NOMIS web, which is an OS, ONS related website about labor statistics. And those data use the open government license, and you don't have to register all again in order to access and download them. Whereas safeguarded data are available only via UK data service to members of academia, local and central government, NHS, and UK parliaments and assemblies via the end user license that the users need to accept when they register with the UK data service before they are able to download any safeguarded data. And finally, we have the secure data as well. The secure data are the most, as the word says, the most secure data because you cannot download them from a website. You will have to apply to become an approved researcher via the approved researcher scheme at ONS using what is now called the ONS secure research service, SRS, which until recently was known as the Virtual Microdata Laboratory. Regarding the safeguarded access, again, you do not get an automatic access if you have, even if you're, in theory, eligible for the safeguarded data. So if you're a member of academia and you have an ACUK email suffix, you will still not be able to download any safeguarded data unless you first register with UK data service and then all the safeguarded data will become available to you. So how the system currently works is by checking the email suffixes. And this is an indicative list of some of the permitted email suffixes. So it's ACUK for academia, GOVUK for government agencies, NHS for NHS, and parliament, Scottish parliament, Welsh government, Scottish government. So yeah, if you have any of these email suffixes, you mean that you can get the data as long as you have registered prior and accepted the end user license with UKDS. Sometimes this approach causes some anomalies because there are some rare instances that some government or academic bodies stuff working there do not have the appropriate email suffix like ACUK or GOVUK. For example, they can just have a dot com. And those rare cases sometimes cause these references because the email suffixes here is an agreement between ONS and UK data service. So it's difficult to overcome. So if something like that happens, usually you have to go straight to ONS to grant your license before downloading any safeguarded data. So back to types of census flow data. I promise that I explain a bit more what's the difference between the headcount, univariate and multivariate tables. So as you can see on the top, the flow headcount, as for example the MF02 UK table is the simplest form of flow data because it consists of totals. So we have the origin destination and then just the number of persons that were included in that flow. Then we have a bit more complex table, the univariate one. And these are tables that relate to one single variable. So we can see instead of having a total, we have a family status variable. So we have lots of rows about describing the different family status and then a total of persons corresponding to each one of these rows. And then it's the multivariate tables, which are the most complicated and bigger ones and the more detailed ones. And those are cross-classified, cross-classified one variable with another. So in this example we cross-classify family status with sex. So you can see the level of detail increases significantly from the previous types. So we have 30 different cases for this particular example because each one is a different case. So we can select people aged 65 and over that are not in a family and are, for example, specifically female or we can exclude a gender, not a gender, a sex for this particular example. Another thing is that 2021 will probably change some of the approaches on that so I think we will no longer have identified sex as the question will be rebranded as gender and we will have more inclusive answers. So that's a good thing moving forward. So the security classification. So as I said, flows, headcounts are the simplest type of forms and the less detailed ones. So in the less detailed geography as well, like for example local authority to local authority, all of the 2011 tables are made public. So you can just go and download them without any hassle of registering or anything else. But as we move to more lower level geographies that have more areas and become more detailed, then we can see that most of those tables are safeguarded with the exception of some workplace statistics in England and Wales that are public. All the other ones are safeguarded from wards to middle super output areas to workplace zones to output areas. Next one is the univari data sets which are a bit more detailed than the headcounts, but less detailed than the multivariate ones. And the security classification here you see that we also have some secure ones. So the output areas to output areas and the output areas to workplace zones are all secured. You cannot download these from UK data service. You will have to apply to become an accredited researcher on ONS, go to their lab and download and use the data there. But still the higher end of the spectrum of geographies, like for example in local authorities, you can see that for some variables like sex, AIDS and method of travel, these tables are publicly available. All the others are safeguarded. And this same is with the wards and MSOAs. Sex, AIDS and method of travel is public. All the others are safeguarded or even secure. And then the multivariate data sets which are the most detailed ones. So the only public ones are the sex biades in local authority level which is less detailed ones. And some other bars are safeguarded and then all the rest are secured. So you can see that there is a relation between the level of detail of the data sets, like multivariate, univariate and flow count and the level of detail of geographies. And this is to protect people from being identified by the census question. So these are precautions and measures that have been taken to secure any kind of leaks of identities of people participating in the census survey. So that concludes some generic information, some general information about census flow data. What happens beyond? Within the current situation, most likely 2021, 2022 census will be the last traditional census as we know it. Probably from 2031 and beyond what we call admin data will be used to replace census. So those admin data can be data from the NHS, data from mobile phones used and everything that the government probably can have access to it. Always, you know, respecting or there's always the issue of privacy in those terms. But this is something that hasn't become concrete yet. So we don't exactly know how this go ahead, but there's a huge pool of information available. So the government thinks that census data on something that soon will become something of the past. The first batches of 2021 census flow data are not expected before late 2022. And that is because of the complexity of the flow data, although we think that because now most of this census collection is online. So it will be significantly speedier than the previous censuses where everything had to be digitized. The papers had to be digitized before being processed. So for 2011 it took more than three and a half years for the first 2011 batches to become available. And now it would probably take a year and a half, two years at the most. But because of the COVID-19 situation, emergency, we expect harmonization issues because as probably most of you know, Scotland has decided to postpone the census to 2022 whereas all the rest of the agencies will still go ahead for March 2021. And this will cause some issues, especially for flow data and especially between cross border flows from Scotland to England and Wales, for example. And it's also a matter of the quality of the data because probably in 2021, March, most people will still be working from home. Whereas 2022, we hope that more or less we will have to move on from this COVID-19 thing. And so the data between those two censuses might not be 100% comparable. So we expect to see spike of working from home for England, Wales and Northern Ireland, but for Scotland it might be a bit different in 2022. So all those are issues and discussions are being made at the moment on how to mitigate those problems that we anticipate to face in a few years when we process the data. So an example of using the data. So a table called MM01, which is as we described earlier, M for migration and M for multivariate. So that means that it's quite a detailed table, allows us to look at flows at world level from all this outside the UK. So we can look at the international migrants coming to the UK. So processing that, downloading the data from our website, putting that into a database. We prefer PostgreSQL because it allows much more advanced on special data than other types of databases like MySQL. But this is just some personal preferences anyway. So what we did, we downloaded the data, load them in an SQL database and re-aggregate it on the basis of origin and flow size. And then we grouped by origin. So we could find how many words had X as the most common country of origin. And then we look at the number of the countries. So that's the final poll before we move on demonstrating how to download data. So what you think is the most common origin country for migrants from outside the UK in 897 words, which is the absolute most frequently seen most common origin country. So what you think is, which country you think is the most, for instance, one is Australia, China, India, Poland or Spain. Most of you identified Poland followed by India, China and then last Australian, Spain. But the reality is wrong answer for most of you. And congratulations to the 9% that identified Australia. So Australia is the most popular country in terms of migrating to having the largest number of words, migrating in the UK, followed by Spain, USA and then Poland, followed by the other European Union, accession countries, Germany, France, India, other Middle East and then the bottom China. So you see sometimes the census data can show us unexpected results that can surpass our existing knowledge or sometimes our biases. So this is an example of what census data can give us with a very simple and basic processing of the data available through our website. And this is just the words that Australians have preferred in the UK. This has been visualized using QGIS with the data that were produced in Postgres SQL. These are the words for Spain and the USA. So yeah, using maps like this, it becomes easier to see where international migrants go across the UK and it has an impact on public policymaking. Okay, so what I'm going to do now is share my screen and I'm going to demonstrate on how to download data using the Wicked Tool that we have at the UK Data Service. So I'm going to use Google Chrome, which is one of the most popular web browsers. So let's head to the UK Data Service main website. Oh, just another note, not only the UK Data Service is soon to use more robust tools for video conferencing, but also the website soon will get a revamp. So in the near future, expect it to look much more neatly, much more, you know, much better. But we'll look different on what I'm going to show you. I'm not quite sure when the launch is, but it should be fairly soon. Okay, so this is the main UK Data Service website. There's lots of information around news, how to deposit data, how to find data. We will only interesting for the census data at the moment. So if we scroll down, we will see some of the most prominent data types offered by UK Data Service and census data on top of it, because census data are quite significant for the UK Data Service. So if we click on that link, we will move to the main census support page within the UK Data Service website. Here is the data catalog where you can search for specific datasets, some latest tweets, various information. And then we have a quick access panel at the bottom right. And you can access other census data, such as aggregate data, micro data and boundary data. But for today's workshop, we're only interesting for the flow data, so we click on that link. This brings us to the next page, which is the main UK Data Service flow data information page. With some very generic information about flow data. And then we click it, that link will bring us to the main website where we will be using to download and access flow data. Just a reminder again that if you want to download safeguarded data and you haven't done... And you're eligible, you have an ACK for example, or a Gov UK email because your staff or a student or a researcher at a government body or higher or other education before downloading safeguarded data. If you haven't done so, you need to register. Click register here, put in your details, accept the end user license. And only then you will be able to log in and access the safeguarded data. Right, so if we click on the Wicked link here. So we're getting to the main Wicked UK Data Service, which is the portal for accessing flow data. So there is some information and then here, this panel here is the main route to access this kind of data. So the three options, Wicked. Wicked is the flexible query builder. So you will be able to subset, create labels and stuff like that. Because this is the most commonly used route to access flow data. But we also offer some other alternatives such as the Wicked Downloads. Which you can download bulk tables as they were generated, as they were produced by O&S. And this is more particularly useful if you're an advanced user, for example. And you want to download a very, very big table. Then this approach might be best because you can get all the data, the bulk data, which can be hundreds of gigabytes. Or not hundreds, but sorry, exaggerate a little bit, but it can be like two or three gigabytes. And then you can load it into your preferred system. For example, Excel will not work for such big tables, but there are alternatives like SQL, like R, like SPSS, or like Python. So those are the most commonly used O&S data. Software and programming languages that more advanced users use in order to process the data and reach their research objectives. For this workshop, we'll only handle some small data. So here, I'm not quite sure if you see the pointer. If you see the left-hand side of the screen, there are some links, some quick links that you can access some extra information. So you can, for example, click on the available data. And then you get some other types like sample data sets, migration data, committing a journey, second residence. You see those are the types that described earlier. And for example, if you click on the migration data, you get a whole list of tables that are migration related. So an interesting thing is that if you notice, look at how many 2011 tables are for migration. Whereas sets that have several tables included inside them and releasing the tables explicitly on their own. So UK census flow data. And for example, if we click on the UK local authorities 2011, we get some information and we can generate a list of the tables. And we see here because if we download the bulk data, we won't get the description like the area. We will only get this part of the area code, which usually doesn't make much sense. But Wicked offers lookup tables that allow you to match an area to its area name. This is not always the case though, because for example, for output areas, we have 200,000, more than 200,000 areas. So there's no way to put a label on that. So you can see that you will have to deal only with codes. Okay, so let's go back to the homepage. So we'll start by the simplest way of retrieving data from a Wicked website, which is the Wicked downloads. So if we click on that link brings us to the login page. So if we want to download safeguarded data, then we need to follow this route, login using our institutional account. And this will only work if we have already registered with UKDS. For the purposes of this work, so this is not needed because we're going to work with public data. So we're going to click on the starter login, whereas no username or password is required. Okay, that brings us to the flow data downloads page, the bulk downloads page. So we have the categories here. So if we log in with a safeguarded permission, then there will be lots of other categories here. But the public access only allows us a few tables to be downloaded. So if we click on that, this is an expandable list. We get some generic information, like when those data were released, the name of the publisher, official national statistics, and then we get a list of all the public available tables. So if you see on the table names, they follow the structure that we described earlier. We immediately able to identify that this is a migration table with headcounts that covers the entire UK. This is a second residence table that only covers England and Wales. This is a student migration table with multivariate type that covers the UK. This is a workplace headcount table and so on. So if you hover your mouse on top of the download, you'll get the size of the file that you want to download. And if you click here, you get some quick information, some extra information about that table. So we have the table title, table population cover and geography that is also available on the main list. But we also get the description of the columns within that table. So we know that the first column corresponds to areas usually residence. The second column corresponds to the country of address one year ago. And the third one is the total persons. And if you notice, for migration tables, the origin is the second column and the destination is the first one. So country of address one year ago is the second column, whereas the area of the residence, which is the current one, is the first one. This only happens for migration tables and this only happens for the bulk download tables because those are left intact from ONS. But if you use the other access route wicked as I'm going to demonstrate here in a bit, this problem is mitigated because the correct order is restored. So if we download that table, okay, so it was downloaded. You see it is a zip file. So if I open that and extract the file. Okay, so these are the contents of the zip file. If there is a text file that has some extra information to help you, there are some metadata for that table. And then the real deal is the CSV file. CSV file can be opened in Excel, SPSS, or it can be imported in R or Python for further processing. So if I double click on that, Microsoft Excel should be able to open that. Yes, here it is. So look, so the first label is, as I said, the area of usual residence, which is the origin and destination, and then the total number of flows in the persons. And this is relatively small table that can be easily loaded in Excel. And you notice that, again, those codes do not make much sense. So if I go back here to the available data and open the supported geographies for the local... Where is the local authority merge? You see, we get, so 95AA, 95AA corresponds to Antrim. So, I mean, you see that it's something more complex, at least for novice users to do that. So that's why I will have Implemented Wicked, which can do all this work for you out of the box, instead of having to use a table, download the lookup tables that I saw, you know, and then try to find ways to merge those together so that areas like this make sense and have their own unique area name. Okay, so we are going to go now through the main route of access, which is the Wicked table builder. So when we click on that, the first page just gives us some summary of current query, but because we just started that, there's nothing. So we have selected zero origin, zero destinations, zero data items. So what we want to do is first is select a dataset to work with. So that brings us to that page, the data selection page. That gives us quite a few options on how to find the table that we want to work with. One of the probably easiest approaches is to, if you're not familiar with the data, is to use the Table Finder. Table Finder is an external service provided by Novice Web in cooperation with the UK Data Service. And if you click on that, you will see that you are transferred to Novice Web. And then from here, you can find all the origin, destination, or flow data, as we say, available. And it provides filters. You can filter out tables by their availability option. So we can select only the secure ones, or only the public ones, or only the safeguarded ones. And then we can also filter those using variables. So if we want to find tables, safeguarded tables that are related with, let's say, hours of work. Okay, so we get those three tables, the workplace, univariate10.covers.uk. So you can just click on the special level that you want. You will be presented with a pop-up. And then you can click on the Wicked Interface because that table is safeguarded and it is only available through Wicked Interface. So if you click that, we're back in our website. The landing page with some options on how to download the data, on how to select or read the data and move forward. And it also has the geography supported for this table, the break down of the variables, et cetera. So that will give you a good hint whether this is the table that we're looking for or just want to skip and find another table. So this is some demonstration of how to use the table finder. But there are also other routes. The quick selection one just selects the totals from the entire list. So if we click on the quick selection, you will see that there's a whole lot of tables. The grey day out once means that these tables are safeguarded. And so you'll have to log in using your academic or governmental account in order to be able to access them. And the blue ones are the ones that the guest account can have access in. But again, the quick access one only gives you the totals. You can select specific variables. So if I go back again and select by data set and table, that will allow me to select the subset of the table. So let's go, for example, on migration data and select you see the options here. I will have the latest census 2010-2011 and some previous censuses and some other data estimates that have produced. So we select this one. And let's use the first one, which is the multivariate migration table that covers the entire UK at the local authority level. And it is about internal migration by aids, by sex. So if we click on that, that will get us to another page where we will be able to select specific subsets. So if we click here, you see now we're not only able to select the entire totals but we are also able to select a subset of that. So let's compare the migration between people aged between 16 and 49 and the older folks 75 plus. So we select those. A wicked uses a traffic like metaphor. So you see now I select the data and that link, that icon here became green and gives us a hint that we now need to select geography origins. And geography origin consists of several other options. The quick selection again gives us the opportunity to select all the supported geographies at once. But we can also go through a list to select geographies individually. For example, just local authorities confirm. And then we can, as you see, click on the areas that we're interesting in. You can change the order, etc. But there's also other methods like the typing box where, again, local authorities where you can just type a name. Like London, submit. So it identifies the city of London within the UK local authority list. So that's our first election. So now we have selected one area as our origin. But you see the geography is still red here because we need to select our destination as well. So now we have other options. We can again select all the available geographies. We can again list. We can copy the selection so we can duplicate what we selected as origins and make it the same as destinations. We can type in, we can use post codes or we can use an interactive map that gives us more robust methods of identifying the areas that we want to include. So I'm going to go briefly on that. And this interactive map only supports 2011 data. And it doesn't support 2001. So if you click on that, it goes straight to a big map showing the UK because we selected London as an origin before. The application wants to confirm that we want to include London in this election. It will click yes, so we selected that. So as you can see down here, we're in the origins view. So we can switch between origins and destinations. So if we call the local authorities, let's find it. So you see City of London that was selected before. You can locate the area map. Here it is. So that's one of our selections, more of those. So for example, we can create a circle and select all the areas that are the circle touches. And this is measured in kilometers. You see the radius as I try to make it bigger or smaller. So let's, for example, select 30. Okay, you see now those 10 different colors. So we select those. We can also select, we can also remove using that icon. So we want to remove that. And another handy thing is that you can, for example, go on a higher level of geography. And let's say that we want to select all local authorities that fall within that region. So we're going to use the Magnet tool and click on Scotland. And then we'll get a list of all the available lower level geographies that are compatible with a higher one. So now we want to select all individual local authorities within that region, which is Scotland. And we wait for a few seconds. Okay, and there it is. It selected that. So if we go back to the UK local authorities, we see that all the local authorities within Scotland were selected. And I think we haven't selected anything for this nation. So I'm just going to put that in Northern Ireland creators and select those local authorities. Okay, so that was a very quick demonstration. You can play around with that if you want. But if you are not unsure on the exact location that you want to select for your research, I think that tool can become quite handy. So you can see all the selections here. This is for this nation. You can set some of them, you can locate them on the map, or you can even remove them from here. Okay, so what we want now is to go back to Wicked. So we will click Save and Return. Okay, the data has been exported. So if we click here, we'll see that now we have 43 origin selected and seven destinations selected. The geography traffic light is green. So that gives us a hint that we're ready to move on. So we're going to finalize. This page gives us an opportunity to also have a look at the summary of the query. So these are the specific areas that we selected together with the dataset. And we're ready to go and run that query. Okay, so that was done in less than half a second. And we continue to the output pages. So we're given two options now. We can either tabulate the data and download them, or we can use some basic analysis like finding correlations or some distances traveled. This is not something that this workshop will cover, because you'll get a bit more complicated than it already is. So I'm going to go straight to tabular output, but in your own time, if you want, feel free to play around with some of the analysis tools that are available on Wicked. So we click on the tabular output. And here we're given some more options, some of the most important ones, the output layout, origin destination matrix. So that generates a data output with origins as rows and destinations as columns. This is handy if the table is not very big, because if it's enormous, then that might cause trouble and most likely won't be able to load it into basic software like Microsoft Excel. But we're also given other options like the origin destination pair list. So this just puts all columns next to each other. So we have one column with the origin, one column with the destination, and then the totals for each variable that we have selected for now. Just select the destination pair list for this example. And we can also change the labels produced. So the LA label will give us the actual names for the local authorities. But if we want to, we can also put the area codes, like those weird numbers like 95AA, because maybe you want in the future to use a more advanced approach and you actually want those area codes to be included. And we click the change labels. We make sure that comma separated values are selected. This is important because HTML's just don't cut it. I mean, it's fine if you just want to demonstrate some very, very minimal flows. But all the other cases, it is best to generate them as a comma separated values because that's the way forward to import the data in any kind of process. So we have a lot of processing tools and software. And I think we're ready. Pair list, comma separated values, change the labels. So preview and download. Okay. This gives us some generic information about what we selected. Preview, which is not very helpful in this form, but we're going to download the data and open that in Excel because it won't be a massive file. So we go to download output data. The default file name is this one. We can change it to whatever you want. I think that was something like that. I'm not quite sure about that name. Again, save. Okay. So this is straight into CSV and not a zip file. So we opened that in Excel. And you see, we now have both the local authority name, but we also have the codes because I explicitly select them there. And the same one goes for the destinations. And then we have the actual data. So for example, we only selected four variables, I think, six with the totals. So we have people 68, 16 to 49. And all six, we have 649 only male, 649 only female, 75 plus, all six, 75 plus, six male, and 75 plus female. So it's very obvious even with an egg die that people 75 plus are extremely more likely to have moved houses compared to the younger counterparts. This is of course also because the population of people between 16 and 49 is greater, but it also might mean that other socioeconomic and age-related things like people, older people, unable or unwilling to move a house for any obvious reasons. So that was a very quick information that we can get by simply downloading a very simple kind of data set from Wicked. Okay, so we're going to move on with the practical because we don't have much time. So if you notice on the dashboard of the GoToWebinar, there is a handout section. So if you can download that file and open that and go through these slides and try to download the data via Wicked. And those braves ones, so if you are more advanced in using the data can also run the final exercise which is importing the data into Excel and running some Excel functions that will allow you to identify the most popular countries of origin into the UK. So I think we should give like five, seven minutes and then another 10 minutes for feedback questions and answers and then wrap up the workshop. So in the meantime, I'm going to have to go through the questions and try to answer them as you go through the practical. If anyone has trouble opening the PDF file or whatever, please do let us know. I mean, if you don't have time to do that now, it's fine. We can also do that later and send us any queries. We're more than happy to help you even after that workshop ends. So yeah, let's get started with the handout. So if you followed all the steps and you open the table on Excel, you should be able to see something like that. You can see, as I said, because it's the matrix and not the parallelist. You have all the origins here as rows and you have all the destinations here as columns. And you get the totals because this is a headcount table. And you can see, for example, I know 18 people from Spain moved to Hartpool between 2010 and 2011. And so on. And the exercise will require you to put some Excel functions towards the end in order to find the best, you know, the top countries for migration for its local authority. But probably we don't have time to do that now. So feel free to do that just later and send us any questions that you have. More than happy to answer. Yeah, I think it's time to wrap up. I mean, we're not getting any general questions. So I mean, you can just send us any questions that you have afterwards. So the details, yeah, the emails are here. So you can either email me, the video treatise at ICUK or the Audic Williams at UCL-ICUK. So if you have any trouble or any questions, send us all any of the queries there. And we'll be more than happy to answer them for you. So I think that concludes the workshop. Thank you very much all for attending.