 Hello, everyone, and welcome to this webinar on religion and data in the UK. First, some introductions. I'm Ali, and I'm based at UCL, also presenting our Tom Clemens, based at the Scottish Longitudinal Study at Edinburgh University, and Yassine Kooja, who's based at Goethe University. This is just UKRI. They have various data and method services that they fund, and these are the ones. And then also there's a webinar on language and data in the UK, which is happening next Wednesday on the 25th of March, again, from 3 till 4pm. And you can go back to UK Data Service to see details of this. So today's webinar, Yassine will be presenting on religion in understanding society. Then I'll present on religion data, which is available in the ONS Longitudinal Study, which covers England and Wales. And then Tom will talk about religion data available in the Scottish Longitudinal Study. If you have any questions, we won't have time to answer them as we go along. So if you can type them into the questions box, and then we'll answer them at the end. So over to Yassine, who's going to present about religion data in the understanding society. So my name is Yassine Kutia. Welcome to the webinar on Investigating Religion in the UK. I hope the connection works. My name is Yassine Kutia. I'm a researcher at the Institute of Sociology at Goethe University, Frankfurt Amin. And I will present you some information about how religion is measured and covered in the understanding society data. So my presentation is going to have the following structure. So I'm first going to give you a very brief introduction to understanding society data. I can't go really into depth here. That would be beyond the scope of the presentation, but I just want to make you aware of some of the main components of the UK HLS that you should be aware of to then understand the second part of my presentation where I'm going to cover some of the specific questions that are included in various components, modules, and waves of the UK HLS. Then I'm going to just make two brief examples of papers that use the data on religion of the UK HLS so that you can get some sort of idea what sort of analysis are possible. And then I'm going to end with some final notes on religion and other UK surveys. So let's start. What is the understanding society data? So you may be aware that the UK HLS study is currently the largest HLS study in the UK. It was started in wave one in 2009. And currently there's data available up to wave nine. So each wave usually ranges over two years in which the interviews were conducted, which is good to keep in mind. It's still uprunning, so we can expect to have several additional waves in the years to come. Now the UK HLS consists of four different sample components. So there's a general population sample, the GPS, which has again a Great Britain sample and a Northern Ireland sample that's so far relevant to know as some variables are measured separately for these two samples. And jointly they consist of about 40,000 households, even more, there are even more interviews because sometimes there were several people, of course, interviewed in one household. Another sample component is the BHPS. So that was basically the predecessor of the UK HLS. And at the time when the UK HLS started, there were still some remaining households within the BHPS, the British Household Panel Study. And those were incorporated into the UK HLS at wave two. And yeah, I'm not going to go much into that. There are also some questions on religion in the BHPS, but yeah, I won't cover those. So if you are interested in these, you should have a look at that yourself. Then importantly and interestingly, there's an ethnic minority boost sample, which consists of about 4,000 households from the main, the largest ethnic minority groups in the UK. So that's Indian, Pakistanis, Bangladeshi, Turban and Africans, plus additional interviews with some people from other ethnic minorities. And then there's an immigrant ethnic minority boost sample at wave six, where they basically included an additional 2,900 households of these minority groups and immigrants also to somehow adjust for the attrition of the original sample of the minorities and to allow for really more detailed analysis also based on ethnic minority groups. There are of course also ethnic minority members in the general population sample. And those should usually also be included in any sort of analysis because those are also some of areas of low ethnic density in contrast to the ethnic minority boost samples, which usually only focus on minorities in high ethnic density areas. And yeah, to get the, to make the data really representative, you need to use the weights and include the minorities from both samples to really get some representative analysis. The UKH less has a wide range of topics that are the regular questionnaires that are asked annually, ranging from yeah, health, work, income, education, family life, social life and so on. So that's really good. But then in addition to that, you also have some rotating modules on specific topics including religion. And those are not asked every year, but yeah, in a regular, in regular, relatively regular intervals. And then on top of that, you have something that's called the extra five minute questions that were asked to a subset of the samples. And there you have additional questions, especially related to issues of ethnicity and immigration, which were asked to a subset of the sample and yeah, which allow you to go even deeper into some issues, especially related to religion. And I'm going to say a bit more about that in a minute. So if you want to do analysis, if you want more information, here are just some references to user guides, the general user guide and the user guide to the ethnicity and migration research based on understanding society. And also the website is very helpful. So yeah, there's much more information there that yeah, you can look up yourself. So let's move on to what sort of questions are included about religion in ViewKitchen S. So there's the religion module, which is the main module on religion. And this one includes some basic information. So you have information about the religious background of respondents, then information about religious participation, so attendance, and then some information about what I call now religious belief. And these questions were basically included in the adult questionnaire. They are stored in the file and in Dress. So in Dress is the individual level file, which includes most of the information. There are also different other files, which you will notice if you download the data set. And then here is basically a placeholder. It stands for the various waves, which are labeled as A, B, C, D, and so on when you download the data. So these questions were asked to all respondents in wave one, wave four, and wave eight. The next one is planned for wave 12. And in addition, these basic questions were also asked to new entrants in the extra five-minute sample. So let's go a bit deeper into that. What are the specific questions that were asked? And first, it's important to note here that if you have the data set open, you'll notice that there are different variables that measure religious background for the Great Britain sample and for the Northern Ireland sample. I'm just going to go briefly through the questions of the Great Britain sample with some of the basic descriptives so that you get an idea of what's covered there. And if you want to know about Northern Ireland sample, you can have a look at that yourself, but it's basically the same question, more or less the same questions. But they're measured with different variables. So the OPRLG item is basically just the question, do you regard yourself as belonging to any particular religion? And at wave one, as you see, that was asked to everyone in the data set. And you have about 27,000 people answering yes and then about 21,000 answering no. And that's a filter question because based on what respondents answered here, they either get the question, which religion where you brought up in. So in case people indicated that they do not belong to a religion, then they were asked which religion where you brought up in. And then you see here on the right-hand side the various religious denominations that were used. So also as first option, no religion, then various different Christian religious denominations. Also Muslim Islam as a denomination, and then a few other religious groups. And yeah, that's basically what was indicated. Here is then the question for the respondents who did indicate they belong to a religion, then they were asked to which religion they belong to. And again, that's more or less the same list except of the no religion answer category. And again, most people answered here, of course, some sort of Christian denomination, but now you see already that the case numbers for Muslims is higher and also for the other religious groups, Hindus, Jewish and Sikh and Buddhists. You see here in the upper, in the most upper row, there's minus eight and these are basically the respondents that are covered in the variables for Northern Ireland that I just mentioned earlier. Then the other two questions, religious participation is pretty straightforward, it's a standard question on attendance. How often, if at all do you attend religious service or meeting with answer categories once a week or more, less often but at least once a month, less often but at least once a year, never or practically never or only at weddings, funerals. Then there's a question on religious belief, which is how much difference would you say religious belief makes to your life? Would you say a great difference, some difference, a little difference or no difference? So that was basically the religion module. Now we have the extra five minute questions where you have some additional more specific questions, items on religion and let's walk briefly through those. So first of all, the extra five minute questions consist of three components. So people who belong to the ethnic minority boost sample or to the additional boost that was started at wave six, they were asked these additional questions for all of respondents who belong to these different samples. Then it has a general population comparison sample, which is basically a random sample of the general population sample, which is kind of handy to make comparisons. And then you have ethnic minorities in low ethnic density areas in the GP samples that are also asked these questions to really make the data representative. Within these extra five minute questions, there are again barriers rotating modules and the modules that include questions specifically on religion are on the one hand the ethnic identity module, which is included in the six year rotation. So it's asked at wave two, wave eight, and then at wave five additionally for new immigrants in 16 to 19 years old. Then there's praying frequency, which was asked at wave four and wave eight. And then there's religious practice, which is also at the six year rotation and was asked the first time at wave four and then will be included in wave ten, which is an upcoming wave. So what are the questions that are included here? So first of all, you have questions about the importance of religion to the sense of how people are. And I want just to note here, these are always two variables. And again, the same logic applies here as for the religious background variables. There's one variable, which was only asked to people who answered that they belong to a religion. And then another variable, which was asked to people who said they do not belong to a religion. And then, depending on what they answered, the reference here was to the current religion or to the religion that people were brought up in. So yeah, you have and how important is your religion or the religion you were brought up in to your sense of who you are with answer categories from one to four. Then as a second, you have, do you feel proud of your religion? The religion you were brought up in with three answer categories? Yes, no, neither, yes, or no. And then you have, how do you feel when you meet someone who has the same religion as you or the same religion as you were brought up in until you had five answer categories from one very happy side? Very unhappy. Then there's a standard question on praying frequency, which was included at way between four and eight of the extra five minute sample. So here the question was, apart from when you are at religious services, how often do you pray every day, more than once a week, once a week, at least once a month, only on specific holidays, less often or never. And then finally, you have the religious practice module, which was so far only asked at way four, where you really have a whole range, a whole set of items that cover whether the religious belief of the respondent affects various behaviors. So eating, drinking, bearing certain types of clothes, marriage and dating, education, decisions for charitable giving and helping, investments, friends and jobs. And here for each of these items, people could respond with either yes a lot, yes a little or not at all. So that's the extra five minute question. So these are basically the main questions that included in the UK HLS. There's some other questions that also somewhat address religion, particularly in relation to discrimination when searching for a job or other forms of victimization and harassment where religious often asked, whether respondents are usually asked whether religion might be one of the reasons for why they were discriminated in the past. So there's much more on these issues, but I didn't include that here now because I didn't think that is something that measures religiousity or religion of respondents self. But yeah, if you're interested in that, you can also look at, find some more information on the website. So now just some research examples. I'm going to go really quickly over that. So there's one paper that I wrote together with Lucinda Platt, which is called Labor Market Entries and Access of Women from Different Origin Countries in the UK. And here we used the item religious beliefs. So how much difference would you say religious beliefs makes to your life in order to explain ethnic variation in labor market transitions of women. So we basically used this as an independent variable to explain labor market transitions. And yeah, the nice thing about this variable, of course, is that it allows you to kind of compare really just between different religious groups, which is often problematic if you use such measures as attendance, where you have really different guidelines regarding how often people ought to attend in different religions. Another paper which uses the UK HLS is by Shannon Kogan. And they basically use the religious affiliation, the religious background measure to create a neighborhood level measurement of religion. So yeah, to the share of people from the same religion in the neighborhood or the share of Muslims and also the share of non-religious people. And they look at how this affects individual life satisfaction. So that's also a very interesting approach that is possible with this data that I wanted to share with you. There are other publications that use measures of religion that I didn't include here, unfortunately. But yeah, you can also look that up at the website and search particularly for studies that use religion here. Then finally, to end the presentation, I want just to make some references to other surveys that also include information about religion. There's the British Social Attitude Survey, which was asked in most years since 1983 and usually covers just some basic information on religion. But in the years 1991, 1998 and 2008 and 2018, it uses the module of the ISSP. So it goes much more into depth about religion. And then there's the Citizenship Survey, which was running from 2001 to 2011, which also includes information on religious affiliation, behavior and perception of religious prejudice. And yeah, you can search the UK Data Service website to find more surveys that include information about religion and yeah, any data set that I mentioned, you can also find on the website of the UK Data Service, including understanding society and access to data from there. That was it for my part. Thank you for your attention. So yes, I'm Allie and I'm going to talk to you about religion in England and Wales using data from the Office for National Statistics Longitudinal Study. We're going to start off with a poll. The question is what was the most frequently reported religion among the England and Wales population in 2011 after Christian? And you need to be aware that Christian includes Church of England, Catholic, Protestant and all other Christian denominations. And there are four choices. Hindu, Muslim, Jewish or no religion. Okay, so 67% said no religion and 30% Muslim. Okay, thank you. So the question might seem quite simple, but it's not. You might have changed your religion. So what religions you put down, your initial religion or your new religion? Also, you might have a religion, but you might not practice it. So do you then give your religion or do you put another answer? And also, what about minority or unofficial religions? You'll find out what proportion of people in 2011, the highest proportion and what religion they reported after Christian in 2011 later on in the presentation. But the presentation is going to, first of all, tell you about the Office of National Statistics Longitudinal Study. Then it will talk about religion in the study, dealing with what questions were asked, when they've been asked and how it can be answered. And then briefly, I'll just give you some examples of previous studies that have used religion in the ONS-LS. So the Longitudinal Study is a 1.1% sample of individuals in England and Wales and they're selected if their birthday falls on one of four days in each year. And there is census data from 1971 to 2011. The censuses in England, well, in the UK happen every 10 years. So an individual's census forms are linked from 1971 to 2011 for as many years as they were present. And this gives up to 40 years of data on study members. The data also includes information about people in the same household as the LS member at the time of the census that is being completed. So when a Longitudinal Study member is a child, the information on other members of the household might be about their parents and their siblings. But later on, when they're in their 20s and 30s, the information about other people in the household will be on their children and their spouse. And the data is also linked to some vital events. So that includes births and deaths. Members leave the study through death or through emigrating out of England and Wales. And they enter the study through being born on one of the four birthdays that are used to select the sample or through emigrating into England and Wales and having one of the being born on one of the four days that's used to select the sample. There are two sister studies to the O&S Longitudinal Study. There's a Scottish Longitudinal Study, which Tom is going to tell you about next. That's a 5% sample and the data is from 1991 onwards. And then there's the Northern Ireland Longitudinal Study. That's a 28% sample and the data is from 1981 onwards. And both of these studies have more data than the O&S LS does. And there is also information about all three of the studies available on the COALS website. So to use the Longitudinal Study, there are two access routes. You can either do it in person at a secure setting. And there are three of these secure settings in England and Wales. One's in Wales in Newport and there are two in England, one in London and one in Titchfield in Hampshire. Alternatively, if you can't get to one of the secure settings, then you can submit status grips or R code and they can be run remotely for you and then you will be sent the results. But no data is transferred out of the secure setting until it's been sort of cleared for disclosure checking. In addition to that, all researchers using the O&S LS need to apply for research accreditation or already be an O&S accredited researcher. And if you visit the Celsius website, you can find more information on this and how to apply for it. You will also need to undergo secure data training and that all has to be done before you will get access to the data. And before any analysis will be run for you and any results will be sent to you. So, in 2001, that was the first census when religion was asked about and it was asked in question 10, which asks, what is your religion? There were eight tip boxes, starting with none and ending with a writing box for any other religion. The question was voluntary, so there is also a non-stated category in the 2001 religion variable, which is RELP-naught. You need to be aware that the Christian tip box includes Church of England, Roman Catholic, Protestant and all other Christian denominations. However, some people may have ticked any other religion and then written, for example, Catholic in the text box, in which case in that RELP-naught variable, they will have been coded in a category called Catholic. And you need to be aware that the people in that category won't necessarily be all Catholics because the vast majority of them will have ticked the Christian box, first of all. And it might be that people who write Catholic are much more sort of regular, goes to church, being more staunch believers, but equally, it could be that they read the question, saw the text box at the bottom, and completed that without looking at anything in between. Similarly, there are people who write atheist rather than ticking none, although this is less clearly a subset of none than Catholic is a subset of Christian. And since the question was voluntary, none of the values are imputed. So the next slide shows the numbers of people for each religion in 2001, in the nine standard census categories. The missing codes, minus eight and minus nine, were used for LS members who didn't answer the question. And that was about... That was about seven percent. The most frequently reported religion, though, was Christian at about 72 percent. And then the next category was none, which was reported by about 14 percent. In terms of religion, though, Muslim was the next highest proportion of LS members, and that was about 3.24 percent. So as you can see, the sort of next highest religion after Christian in 2001 that was reported was Muslim, but in terms of the next highest category, it was none. The next slide looks at religion in the 2011 census, and it was question 20 in this census, and again, it asks what is your religion? There were eight tick boxes starting with none and ending with a writing box for any other religion. The question was voluntary, and therefore, there is also a not stated category. And that was... And that's in two variables that there are in 2011, RELP-11, and then RELP-GP-11, which groups the variables into the nine census categories. And again, the next slide shows the proportions and numbers of LS members in 2011 and their religion, the religion they reported. So in this case, minus nine, which was no code required, covers students who weren't at their time address, and then minus eight was just an unacceptable text response. The people who didn't answer the question, because again, it was voluntary, they were all coded under minus six, which is missing. That was again about 7% of the sample. So it's quite similar to the level in 2001. Again, the highest or the most frequently reported religion was Christian, but that had decreased to 63% from about 72 in 2001. And then the next most high category that was reported was none, and this had increased to 23% and then 2001 it was about 14. Again, though, after Christian, then the actual religion that was reported, the most frequently reported one was Muslim, and it was about the same at 3.2%. Longitudinal data allows us to look at changes in religion of the longitudinal study members that you can't see by using cross-sectional data. So this table shows religion reported by longitudinal study members in 2001 and in 2011 for those members who responded to the question in both censuses. The central diagonal, which is highlighted in yellow, shows people who reported the same religion in both 2001 and 2011, and the rows show what happened in 2011 for each religion in 2001. So the orange, they might be cells, but they might be kale-pink. They show what happened in 2011 for each religion, sorry, they show what happened for people who were Christian in 2001 and what they reported in 2011. Likewise, the blue cells show people who reported they were Buddhist in 2001 and then the religion they reported in 2011, and the green cells show the same for Hindus. So this table shows the number of LS members who responded to the religion questions in both 2001 and 2011, and this is about 369,000 LS members. The yellow cells on the diagonal, they show the people who gave the same response in both years, so they gave the same religion, and that covered about 306,000 LS members. However, about 63,000 changed religion between 2001 and 2011, and that was about 17% of the sample, and the majority of these who changed religion were Christian, and that was about 73% of the people who changed religion between the two years. So the next table shows the same data, but using the percentages. So here for each religion in 2001, the most common religion in 2011 has been shaded in dark orange, and then the second most common religion has been shaded in a lighter color, so that kind of light orange pink color. If you look at the diagonal ones, then it's Muslims who are the most likely to report the same religion at both time points, and that's about 96% of LS members who responded to the religion question. And the least likely are the other group, and that's about 24% of people. So that's the one diagonal cell which is not shaded in dark orange. Hindus, Sikhs, Jews, and Christians also tend to report the same religion. Of those who stated a religion in 2001 and changed their response to the question in 2011, they usually changed reporting other or none. In contrast, for those LS members who tick none, they see no religion in 2001 and change religion. They usually change to Christian, and that's about 20% of them. So that is this number here. Okay, this is a list of current and previous research using ONS longitudinal study data that also looks at religion. There's one looking at continuity of non-traditional religious affiliation, and that's by Williams, Dennett, and Shelton. Then Sylvie Duburt has looked at the fertility of ethnic and religious groups in the UK, and Rob Henderson and Hattersey have looked at regional patterns of teenage births in relation to social factors in education, and the social factors include religion. There's also a study on mortality in religion, and you can find out more about these studies from the Celsius website. So to access the data for the ONS LS, you can do that through the support unit, and that's Celsius. I'm one of the USOs, the user support officers there. And you can also find out more about accessing the data from the Celsius website. And Tom, who is going to present next, he will tell you about how you get access to the data to the Scottish longitudinal study. So that's the end of my presentation. Thank you for listening, and I just need to acknowledge that the data that's been presented uses the Office of National Statistics longitudinal studies, and that although it contains ONS statistical data, it does not imply the endorsement of the ONS in relation to the interpretation or analysis. So now it's over to Tom. Okay, thanks very much, Ari, and hello everyone. My name is Tom Clemens, and I'm part of the Scottish longitudinal studies here at the University of Edinburgh. So I'm going to be fairly brief, I think, because a lot of what Ari has covered also applies to the SNS. But there's a couple of key differences, I think, between the Office of National Statistics longitudinal study and what we have here in Scotland. It is the SNS, the Scottish longitudinal study, is based on the sensors in the same way as the ONS longitudinal study. We have a slightly larger proportional capture of the Scottish population, so we capture 5.5% of the Scottish population based on 20 semi-random birth dates. This gives us around 270,000 SNS members and around 505,000 household members associated with those SNS members. And similarly to the SNS, this data allows us to cover or examine a wide range of questions, ranging from demographic and economic questions to health, housing, migration and fertility and so on. Importantly, this large population capture that we have allows for quite detailed geographical information, quite detailed geographical analyses, and users are able to provide their own lookup tables for their own specific research projects that can be linked to the SNS. But apart from a few key differences then, the SNS functions in very much the same way as the ONS longitudinal study. So one area that the SNS is slightly different is that we have some extra datasets over there in SNS that we're quite fortunate to have. The main data that we have is the 1991, 2001 and 2011 census. And linked to that we therefore, we also have information around births, deaths and marriages. We also have migration information, which is used in one of the studies that I'll go on to show as an example later on in this presentation. But we also have access to school education information from 2007, so that includes people's census and attainment information as well. We've also linked to environmental information as well as a set of other kind of geographical variables. Quite crucially as well in Scotland, we're very lucky to be able to link the SNS to wider NHS Scotland data, so we're able to link to things like hospital admissions data and other health datasets, quite a wide range of health datasets, which obviously can be studied alongside the data that we already have in the census records, including religion, for example. So I just want to kind of highlight with the slide the fact that all three of these studies are quite similar. Allie's touched on this already, and it's just to note that there is the possibility of doing comparative studies, and that's quite a powerful resource, quite a powerful resource in particular relation to studies interested in religion. And one of the other projects that I'll go on to discuss later on is a good example of this. Okay, so moving specifically to the question of religion in the SNS, we have our religion collected in two sentences. So the religion questions are asked in 2001 and 2011, but not in the census in 1991. Similar to the SNS, the questions on religion were not compulsory, so there is the possibility of selectivity in the response to these variables as noted by Allie earlier. Okay, so I don't spend too much time on these slides simply because they're very similar to what Allie just presented, but here are the questions as they are asked in the actual census. Crucially, in Scotland, there are two key differences in the religion information that we have in Scotland compared to England and Wales, the first being that we ask two questions, so we don't just ask what religion do you currently practice, but we also ask what religion or religion determination you were brought up as, which is quite an important additional piece of information that we have in Scotland. And we'll go on to discuss how that's been used in some of the studies in a second. So both questions are relatively high valid response rates in the 90%, so that's great. But the second key difference that we have in Scotland is that we have separate categories in the Christianity group, so we've separated Church of Scotland and Roman Catholic. Obviously, Scotland has a history of sectarianism and it's an issue that's one of the important in Scotland, which is why the census here in Scotland is differentiated between those categories. So that's been a focus, so a lot of research that's used this information in the SLS already has been to study the issue of sectarianism in Scotland. Okay, so here is the numbers for the religion variable in 2001. So just to give you an idea of the numbers that we have available, you can see that we've got these separate categories for Roman Catholic and Church of Scotland, and those are the two dominant religious groups in Scotland, as you can, perhaps not surprising. But you can see as well that despite the relatively low proportions over the population for some of the religious groups, we've still got relatively high numbers for some group analysis if that's what you're interested in. So that's what religion you belong to in 2001. Now I mentioned before, obviously we have this additional variable which allows us to examine how you were raised and what religion you were raised, which also allows us to see people who have changed their religion from childhood into adulthood. Now this table looks a bit sparse and that's simply because a number of the cells are too small. So for reasons of disclosure control and confidentiality, we've had to restrict the output. So those blank cells will have numbers in them, they're just very, very small and we weren't able to release those. But what we can see from this slide is just an indication of numbers of people who have reported a different religion in adulthood than that of which they were raised during childhood. So most people stick to the same religion from childhood into adulthood. Some switch between some of the Catholics who are here, Church of Scotland and Roman Catholic, but the vast majority are consistent with their religion. So moving on to the religious questions in the 2011 census, so both of those questions relate to the 2001 census. Moving to 2011, the question on the religion for which you were brought up is no longer asked. So that's an important thing to remember. If you're looking to use the SLS, we don't have information on religious denomination you were brought up in 2011, but we do have the standard question about your current religious denomination. And here are just some numbers for the different groups in the 2011 census for reference as well. Okay, so I wanted to just illustrate three research examples showing how the SLS religion information has been used in a number of studies. For most of these studies, they're focused on the issue of sectarianism in Scotland. And I think this particular example illustrates the real value of the information we have in Scotland, particularly around the fact that we have household information, so we're able to see inter-sectarian partnerships and get an understanding of changes in inter-sectarian partnerships over time. So this study was interested in examining how partnerships between, particularly between Catholic and Protestant individuals has changed or has changed with age. So you can see that most, if you look at table three here, most religious partnerships are between the same religious individuals, so Protestant, Protestant, Catholic, and the vast majority of these partnerships, or vast majority of partnerships in Scotland are to people with the same religion or practicing their religion at all. What we can see if we look at figure four is that a lot of us is checking here over time. So particularly among younger individuals, we can see that there are great many, there are less, as proportion less couples that are homogamous, and there are more relationships between different people of different religious denominations. So, and the study might tell us something about the changes in the degree of sectarianism and increasing integration of different religious groups in Scotland, which is quite an important and powerful message and something that's made possible by the data we have here in Scotland. I think another real advantage that we have with the Longitudinal Studies in general is the ability to conduct comparisons across the different developed administrations in the UK. So this was a study examining the potential for differences in mortality risk between different religious groups but comparing those differences between Scotland and Northern Ireland. And the really interesting thing about this study is that Northern Ireland in comparison to Scotland has enacted a number of important pieces of legislation designed to decrease or remove or deal with discrimination, particularly sectarian discrimination, which hasn't obviously occurred in Scotland. So we have quite a nice natural experiment comparing how this type of legislation might have impacted on mortality differences between different groups. What we can see here is that actually Northern Ireland has the mortality differences, you can see it in these red boxes, but the mortality differences between Protestants and Catholics in Ireland is much less noticeable than the same comparisons in Scotland. Of course, this kind of analysis really takes advantage of the similarity between the data sets, the Longitudinal Studies data sets in the UK. Okay, the final example then is as I guess a more simple example but just interested in how religion influences migration. So taking advantage of the fact that we can compare people's locations between different censuses and we can use this to look at how religion plays a part in people's migration patterns with a number of interesting conclusions from that analysis. So that's just a real kind of quick brief intro to what the SLS, what the possibilities are of SLS data, particularly religious variables in the SLS, and yeah, I think that's all I have to say. I'll pass back to someone. I'm not sure who for questions I think. Thank you for listening.