 So today's webinar. I am very excited. She is one of our regular presenters and she will be with us next week for these Or not next week. She'll be with us in November for the hundredth webinar celebration It's Michelle. Hey split today. She's going to present on census micro data. She's done like I think three webinars for us on the census data and different kinds of Hi Explore that for the hundred So today is census micro data, which is definitely one that I will enjoy. I'm very excited about this. Michelle has been a data librarian for almost 20 years. She having received her early training through the census Bureau and North Carolina State Data Center. Census data is one of her specialties and she's witnessed changes over three decennial census is now She works constantly with a variety of tools to pull data from numerous census surveys and multiple formats. She is a past president of NC LA's government resources section and she's very excited to do this fifth webinar. There's this You know, I wondered why my shoulder was hurting and now I know it's because you strong arm be into this No, I actually did volunteer. So this is what I promised to talk about today. Um, the what why and where of census micro data. There are a couple things to point out here. Census data is referring to the census Bureau. It's not necessarily the decennial census micro data because a lot of the decennial census micro data is limited at this point. So there's more than just decennial census data and I'm going to spend a fair amount of time on the what and the why and Description about the where I'm not going to try to accomplish getting you out of this session, knowing exactly how to use the micro data tools. First of all, I'm not a statistician and there is a lot of statistics that's involved when you start dealing with the micro data. So don't expect that I'm going to make you an expert in one short hour. And then I hope that I will provide you with the tools that you need to learn a lot more about this on your own. So there are a lot of resources slides at the end of this presentation that I'm not even going to cover in the presentation. I'm just going to tell you that they're there. So later you can go download the slides themselves and you have will have links to all of those resources. All right. So starting off, what are micro data. So they're individual responses. When you go to data.census.gov, what you're accessing there are tables that the Census Bureau has pre made for the process of the micro data tools. Pre made for the public from the micro data. So all of those tables are based on the individual responses. But if you're looking at the individual responses directly Then you have a lot more flexibility in terms of being able to look at greater detail. So One of the first things that you need to know about the individual responses is that they are not weighted. Now I will probably reiterate this several times during the presentation. I am not a statistician. So I really can't explain in detail what the weights exactly are. But there are a good number of resources in the resource slides that will help you understand that concept better. Suffice it for now to say The micro data must be weighted for you to use them properly. And in most cases, you cannot analyze the micro data without either a special online interface or statistical software. So The Census Bureau has a tool that allows you to create tables. But if you really need to analyze the micro data, you're going to have to use Either one of the online interfaces or statistical software to do that analysis. So why would you use the micro data? Well, basically, you can get a lot more detail that can be detailed by topic because Individual variables have a lot more choices within the public use micro data than in the aggregated data. So for instance, you can get Native American population by the individual tribe. You can get single years of age and you can create custom cross tabulations that are not available in the pre-made tables. So in the pre-made tables, you could get a table that's raised by educational attainment. You could get a pre-made table that's income by race. But you could not get all three of those variables together, raised by educational attainment by income. Then moreover, the pre-made tables are limited in scope. They really address the most popular variable cross tabulations that there are. They certainly do not make available every possible cross tabulation. So if you wanted to know income by vehicles available in a household, you can't get a pre-made table that does that. You would have to make that in the micro data. And then you can change the ranges that are available using the micro data. So in a pre-made table, you can get the number of housing units valued from 300,000 to 499,999. But you can't get 350,000 to 550,000. In the micro data, you can create a custom range that will give you that information. There are a lot of caveats to the micro data. First of all, there's less geographic to tail. The geographies that are available in the micro data are called PUMAs. So the micro data for, I'm going to stick with this label for the American Community Survey. The label is public use micro data sample PUMS. So you'll hear me refer to PUMS pretty much throughout the presentation. So the geographies that are available are much larger than ordinary geographies because there's so much detail by topic. The geographies have to be larger in order to protect respondent confidentiality. So the geographies are called PUMS, public use micro data areas, and they are at least 100,000 people. There is a larger geography called a super PUMA that is at least 500,000 people, but you're not going to get anything smaller in the PUMS data than those PUMS. It is more complex to create your own tables. There is a lot that you have to know about the data itself to understand what's possible within the micro data. So if you dive into using micro data, you're going to find yourself reading the technical documentation a lot. You will see as I go along how little I know about PUMS data. And I will point this out on purpose because you need to know it takes some work to really be able to use the micro data effectively. I was kind of surprised to learn this last point that the PUMS data is actually less accurate than the regular data. And that's because, there's actually a couple reasons here. One is that the micro data is edited in order to protect confidentiality. So for instance, in the income data, there's some top coding and bottom coding that goes on so that you would not be able to identify individuals from the PUMS data. In even what is a large geography, people whose income is very high and very low might could be identified from these data because they're so detailed topic wise. So there's that kind of extra editing that goes on with the PUMS data. And then in the second place, the PUMS data is actually a smaller data set than the regular summary files. The American Community Survey PUMS data is only two thirds of the responses of the full data set. So if there are pre-made tables for what a person needs, they should use those pre-made tables specifically because they will be more accurate. The Census Bureau does provide a few different resources for understanding not only the accuracy of the file overall, but also to compare your own estimates to their estimates. This sounds sort of contradictory, but the Census Bureau actually does create estimates with the PUMS data in order that researchers can compare their own estimates to those sort of pre-made PUMS estimates to see if they're on track in terms of their estimates that they're creating with the PUMS data. There's only a few of those. They only use a few variables for those examples, but they are available within the technical documentation website. So the two ways that I'm going to talk about are the new tool that come out from the Census Bureau called MDAT or MDAT. I have heard Census people refer to it both ways. I tend to just use MDAT. So you'll hear me use that today. And then IPUMS USA. IPUMS is a project of the Minnesota Population Center at the University of Minnesota. And they do a bunch of microdata projects, as the name implies. Their acronym is actually slightly different. It stands for Integrated Public Use Microdata Series, and IPUMS USA is specifically all of the microdata from the Census Bureau related to the Census doesn't really make sense because it also includes the American Community Survey. But those two surveys are the only ones that are included in IPUMS USA. We'll talk more about content in just a second right now. So here's a rundown of the microdata content within the Census Bureau's tool, MDAT. So it's focusing on the most recent data and just two different surveys. It doesn't even include decennial census microdata at this point. It only includes the American Community Survey, the Community Survey for Puerto Rico, and then the current population survey, which is an annual survey started after World War II to help quantify economic and work issues, economic and employment issues. You can see that the smallest geographic level for the community surveys is PUMAS, and for the CPS is state. For the basic monthly CPS it's county. Now the reason that the March supplement is limited to the state level is because the current population survey is really designed to be used at the state level. So you get into a little diciness if you try to use the CPS at the county level, but there are some data available at that level for the basic monthly data. I'm not going to go into exactly the details of the CPS because that's not what we're working on today, but we should do another help webinar on that. Not me. I don't use that data very much. Anyway, so here's some caveats about MDAT. It is still in beta, and that's important because for one thing, right now it does not include the margins of error. The margins of error are super important for any ACS data because the more detailed data you get to whether that's by topic like groups of people by race or income or whatever, but also by geography. It's the topics part that's important to us here today. So one of the race categories available in MDAT has 68 different categories for race, including all of those individual Native American tribes, including different origins for race by Hispanic origin. So there's 68 separate categories there. And then educational attainment has 25 different categories. You can look by grade how many people completed first grade through high school. And with income, of course, income is an infinite variable. So with no margins of error, if you cross tabulated those three variables, the individual categories would be tiny. The margins of error would be huge because they have to be with such sparse cells. And you wouldn't know what those margins of errors are because they're not included right now in MDAT. I don't have any indication at this point of when they might think that the beta status might end. All I hear from them at this point is we're continually adding new features. So I expect it to go on for a while. It might be as long as a year. I would tend to think if it were going out of beta soon that we would have heard a date about that. So then the last, not exactly a caveat but more of a note. The Census Bureau says that estimates created with the PUMS data will be different from the same variables in a premade table. And that's because they have a different universe. There's a smaller number of responses in the public use micro data than there are in full data set. But the Census Bureau says those differences won't be meaningful. And I think I mentioned before that the Census Bureau has calculated some estimates with the PUMS data so that researchers can compare their own estimates with those. So that's linked in the resources slides. All right, so IPUMS, and I will say I've worked with some folks at the Minnesota Population Center who work on this project and they say that's how they pronounce it is not with a long I but the short I, IPUMS. I don't know why. Anyway, there's a lot more data here. So for the ACS, they're not just going back to the start of the official ACS data in 2005. They're including all of the test data back to 2000. They include the three year data, which were only available from, what would that be 2005, 2007, from 2007 up until 2013. That's the range of when the Census Bureau was producing the three year data. Yeah, I'm not sure if they have test data before 2007, but they might. CPS goes back to 1962. So there's a lot more data here. Notice that it goes down to the county level for the March supplement data. Now, that's misleading. Again, the current population survey is designed to be used at the state level. So going to the county level is dicey to begin with. And then because this is POMS data, they are, I guess, redacting as a word suppressed is a better word. And they're suppressing a lot of those county level data to protect confidentiality. And I emailed the Ifam's project to find this out. It's a rate of actually 55%. There are only 55% of counties in the March supplement data. So it's not a reliable thing. You won't be able to get data for all counties, even though it goes back so far. All right, so the caveats for Ifam's. First of all, you're downloading the microdata records. When you're using MDAT in the data.census.gov interface, you're actually going to get an aggregated table at the end. They're not going to force you to download the microdata records, but Ifam's does force you to download the microdata records. Now think about this. I live in a small town. I don't know actually how many people live in my town, but it's at least 10,000. So right there, 10,000 records for my little town. The Puma for my area is actually my entire county. And the county includes at least one other, you know, not big city, but a city. So there are tens of thousands of people in that one Puma. So you need to be prepared for the fact that when you download data from Ifam's, it's going to be a very large file. You're not going to be able to use it most likely in something like Excel. You're not even going to be able to open it, except the first, I don't know. The record limit in Excel used to be 5,000 records. So I think it's gone up since then, but I don't think it's up to 10,000 yet. So you're going to need something else to use it in. Second, it's a blunt instrument. With the MDAT tool on the Census Bureau site, you're going to be able to select an individual Puma. In Ifam's, you can't do that. You select a specific geographic level, but then it gives you all of the geographies in that class. And when I say all the geographies, I mean the whole country. So it's not even just the state. So be prepared for that. Okay, waiting. This needs to be addressed. I am really not qualified to explain it in detail. Basically, when you're using Census data, there's two main weights. The household weight versus person weight. When you're using the MDAT tool, it automatically applies what it thinks is the appropriate weight. So if you have selected all household variables, it's going to automatically apply the household weight. If you have selected all person variables, then it's going to automatically apply the person weight. But depending on what you're doing, you might actually pull some household variables and some person variables. And you want the household variables to be weighted like people. So then it gets a little more complicated. You can change the weighting in MDAT, but you have to know that you need to do that. And you have to know which one to apply. So again, I can't really explain that. But there are a lot of resources linked in the resource slides, as I've said before, that will help you get to information about how to understand that. Now, Iphems takes a hands-off approach. They don't try to apply an appropriate weight. They just give you both the household and the person weight, and you get to decide. But again, you need to know that you need to do that, and you need to know which one's appropriate to use in your particular situation. Okay, so these are the steps in the MDAT tool to produce an aggregated table. You select the dataset and vintage when you start, then you select your variables, then your geographies, then you can view your data cart and see what you have and decide if you need anything else, if you need to make changes. Then you go to the table layout stage where you decide which variables you want to appear in the columns versus the rows, who actually construct the table so that it looks the way you want it to. And then after that, you can view the table if you want to. You can just download the table into whatever format you particularly mean. To compare to IphemsUSA, I will say here, IphemsUSA has an online analysis interface. I am not going to try to demonstrate that for you because I don't know how to use it very well, but I will show you how to create an extract which would just download the data for you. So this is a very parallel system to the MDAT steps. First, you select what they call samples, which involves the particular survey that you need and the particular years that you need. Then in the variables area, you select your variables. You also select your geographic levels, but you have to know that they're there. They're listed under household. You have a choice between household and person in that interface. And then again, you can view your cart to see what's there. And when you go to create your data extract, you have to log in. Probably should have actually put this into a slide earlier. I was going to mention it earlier and I forgot. Both MDAT and Iphems are free to access. There is no cost to you regardless of who you are. If you have access to the internet, you can access these resources. The sort of catch with Iphems is that you must register. Iphems is pretty much maybe not completely supported, but probably in majority supported by grant funding. And so one of the things that they have to do for their grants is provide their grant funders with a measure of their impact. And one of the ways they do that is by collecting statistics from this registration form. So you must submit a registration form. There's something hinky about their registration form. There's a field where they want you to describe your project. And there is a 50 word minimum for your description. So if you put in something like class project, it will not allow you to submit your registration form. I have had this tip directly from Iphems staff. I don't know why they don't just change the form. But what they get all the time to get around this is you type class project, and then you type nonsense characters and spaces until you reach the 50 word minimum. That's one extra little caveat for Iphems. But you can play around with the interface. All you want until you get to the step of creating the data extract before it will prompt you to log in. I should also say I believe that their registration forms have gotten to the point that they are pretty much automatic in granting access, but depending on how many they have in the queue, it might take a few hours or a day or two for your application to be approved. So I always tell people if they're going to be using the Iphems tool, they should go ahead and register at the start in order to give time for that registration to be processed before they actually need to start downloading data. So the other thing about Iphems is it's asynchronous. So when you're using the MDAT tool, you're using it live. And as soon as you finish creating your table, you can view your table and it will load the data and bam, you're done with Iphems. What you're doing is setting up a query for their data system. When you finish your query, you submit it, but then the system prioritizes it by how much it will use in the way of resources. So if you have an enormous query covering all possible geographies across the entire country for 50 different variables, it's going to take a long time for them for the system to process that query. I ran one this morning as a test with two very small variables for one geography, one geographic level, and it took less than half an hour for the system to process it. But you can't count on that if you're using, if you're pulling a lot of data at once. So once it finishes the query, Iphems will email you with a link to download the data, and then you'll be able to download the files. Remember, they're very large, so they're going to be zipped. And then you can analyze the data and statistical software. Of course, if you're using the online analysis system, you don't have to download it. You can select your variables and the kind of statistical test you want to do and just analyze it right there in their interface. Alright, so there are at least three instructional videos in the resources slides that give much more detail about respectively Iphems USA, the MDAT tool from the Census Bureau, and using the Iphems online analysis tool. If you have users or if you yourself want to get into using microdata, I strongly encourage you to watch those videos directly and then dig into the technical documentation. You're really going to need to do that to be able to use these tools. Alright, we're up to the demos. It is 1234, and I am going to try to exit my slides. Notice that my last slide there is the question slide. I encourage people to cite the data, all these tools offer example citations, and then we get into the resources slides. So I'm not going to revisit those, but I just want to remind you that they're there. Okay, now I'm going to move over. Oh, I don't, I haven't mentioned this yet, but this is another thing that's linked in the resources slides. Tiger Web mapping application is a tool from the Census Bureau that lets you look at geographies. So if you, I don't know how well you can see this, I'm probably not going to be able to zoom this very much. Okay, so maybe now you can just barely see that I have selected the PUMAS to show up on this map. It hasn't actually loaded the boundaries. I'm not sure why that is, it should be showing up the Puma boundaries in this blue outline. But having done this earlier, I just happened to know that Orange County is its own Puma, and that's where I'm located. So that's what I'm using right now. When we do these examples. That resource again Tiger Web mapping application is linked in the resource slides so that you can easily look up different PUMAS that apply to a given geographic area. All right, so here we are in data.census.gov. If you scroll all the way to the bottom. Yeah, all the way to the bottom. Here's micro data. So you can click here to go into MDAT, where you can just add MDAT on to the end of your URL. You don't even have to put in the pound sign, you can just use MDAT and it will get you here. So here we're going to choose our data set. I'm going to use the five year PUMAS data. I'm going to leave it with the most recent vintage the 2019 data will be coming out at the end of the year. When I go to next. Notice that this says search is not enabled in beta version. So this is another feature that's going to be activated sometime. We don't really know when I'm going to select grandparents living with grandchildren. And I'm going to select veteran period of service. These two variables just happen to be on this opening page, but you can at least choose a general topic area. If you know that you want to look at mortgage costs, you could switch to that area. Selecting a topic up here does not mean that you select the variables. So if I go to mortgage costs, notice that my data cart says I have two variables selected. And I could choose more things here, but if I X out of this topic. It doesn't affect the variables that I've already chosen. Now I'm going to select my geographies. I want to use the smallest one available which is PUMAS. Then it's just a drill down process like we're familiar with from other census tools, North Carolina. It does use the labels here rather than the identification numbers, which is helpful. But you still kind of need to know which PUMAS covers the particular area that you're interested in. I can't remember. Here it is Orange County Chapel Hill town is the particular PUMAS that I am interested in. You were participating in my webinar earlier this year about and not end that but data.census.gov. You may remember that sometimes data.census.gov will look like it's allowing you to select something in a window like this, but if it doesn't show up down here in your selections it hasn't actually selected it. So I recommend that you check for your selections to be sure that they're showing up down here before you are assuming that you have them and you don't. So now I'm going to go on to my data cart to take a look at what I've got. I didn't point this out in the selected variables window, but if you don't know how a particular variable is constructed, this number of values column will tell you how many categories show up in that particular variable. So if you're looking at something like Hispanic origin or race, these numbers can get pretty large and there will be several variables that have very similar names, but different numbers of categories. So that's one thing. And then if you look at the details, it will show you what those particular categories are. So in this case, grandparents living with grandchildren is a very simple variable. Yes or no, and not applicable. Either the person was less than 30 years old, so is unlikely to be a grandparent, or they're living in an institutional group quarters situation like a military barracks or nursing home. And folks in group quarters are counted through the administrative records of whatever that group quarters is so they don't get really detailed counts that they mostly get how many people are in that group quarters, they don't get all the juicy details. And then for veteran period of service. So this is broken out by category. Gulf War earlier on like Gulf War one versus Gulf War two Vietnam era, Korean era, World War two, and so on. So now if I go back to my data cart, you see these reiterated here and you can see, you know, switch back and forth between the two to see in table layout. It has a listing for the columns here and for the rows here. And notice that both of our variables have been plugged into the columns. That's not going to make for a very good table right. So you're just going to click and drag one to the rose. Didn't work. There we go. So now I have veteran period of service and columns and the, the other one that I chose grandparents living with children in the rose. Actually, I don't like that very much. I'm going to try to swap them. I have them both not in table. That's not helpful. Veteran period of service I want in the rose and grandparents living with children in the columns. So there's a variety of things that you can do here. I have to remember how to do this. If we wanted to make the entire table about grandparents living with children, we could move grandparents living with children to not in table. And basically what that does is it makes the entire table about these things. We still have three of three responses in here, and this would be a better table. If we took out the no not living with grandchildren and the not applicable. We would want the table just to be grandparents living with grandchildren. Yes. So, honestly, I am completely forgetting how to do that, but there is a way to do that. For the moment, I'm going to move it back up to columns. So just to let you know that you can move these variables all around. However, you need the table to be constructed. So if you move both variables in the columns, it would just nest one within the other. And so you would get iterations of say you had the veteran period of service first. For each period of service, there would be, are they living with grandparents? Yes, are they living with grandparents? No. So you would still have all of the information there would just be organized differently. At this point, we do have some values in here. But basically the question marks are because all you're looking at here is the shell of the table, how the table is constructed. You haven't actually loaded data here yet. I think it's showing you zeros because it's easy to show zeros. It doesn't require a lot of resources to load zeros. And it might make a difference as you're creating the table to know that there were so many cells that had zeros. But if you click on view table, then it will load the actual values into this. And so you'll be able to see the actual values. And then at that point, here's your weighting. It's given it a person weight, which is appropriate for what I was interested in here. If you wanted to change the year that you were looking at, you could do that here. That's a really nice enhancement over the previous system data ferret. If you remember data ferret. Yeah, we shutter to remember data ferret. And you would be able to download the data here in CSV file or a JSON file. If you were doing some coding, you might want that from that. All right, now I'm going to switch to Iphems very quickly to show you how this works. Notice that if you wanted to get to the current population survey, you would need to use a different interface in Iphems versus Iphems USA, which has the decennial census and the American Community Survey Microdata. I'm going to use Iphems USA for the demo. You always come to this interface when you enter it. Again, you don't actually have to log in until you submit your query, although you could go ahead and log in here. When you click on get data, you go into the extract interface. So remember I said that your first step is to select your samples here. And this is both the survey and the year. So, all of the decennial data is makes up the majority of the historical pums data. But when you're here in 2010, this 10% that's talking about the decennial census. So that's all there to buy default, it selects the default sample from each year. Notice with the ACS data, that's the one year data. And the five year data is always going to be more accurate because it has a larger sample size, five year versus one year. So I always just deselect the default and go with the five year data. For this example, I'm just going to use one year, the 2018. So you can return to this interface to select your variables, and you have four ways to select it you can do household versus person, if you happen to know kind of what category, the particular variables in. You can go a to Z. So if you wanted to get grandparents you would go to G, or you can search for a particular variable. I actually am going to go to G. One grandchildren living in household. And then the other one that we used was veteran status. I know that's going to be in person so I'm going to go here to veteran status. Now, notice that this setup is quite different from MDAT. You can't choose just one variable to get period of service you essentially have to choose all individual categories. They're not entirely intuitive, and they overlap. So veteran served 2001 or later. Veteran served 1990 to 2001. And then there's also veteran served during Vietnam era versus Korean conflict. So, I like that bit about overlapping, but you still have a harder task here to understand these. You can click on a title of a variable and get lots of explanation. So here's the questionnaire text. You'd have to choose, you know which year. Oh, not available. Yes. It'll tell you the universe. It'll tell you comparability across time. It'll give you a base description of what actually is included there. In this case, I'm going to go ahead and add this to the cart. And I'm going to go ahead and view my cart in the interest of time. So here you can see what you've selected. As I mentioned, here's the household weight. And down here is the person weight. These are all things that Ipams includes as a matter of course, but then down at the bottom. You see veteran served 1990 to 2001 and own grandchildren living in household. So those are the ones that I personally selected. So to create my data extract. There's quite a few options here, more advanced options than what are available in MDAT. So I encourage you to explore those on your own, but I will put in description here because when you get to the extract listing all the extracts that you've ever submitted, you can't clear those out. You can build up over time and it's helpful to remember with the description what it was that you were doing with a particular extract. Then when I click submit extract, it will prompt me to log in password. Don't save. And now I go back to submit the actual extract. This is my data page. If I went to my data from the header, you would always get to this page listing all of your previous queries. And this one that says download dot DAT is the query that I submitted earlier this morning that only took a half hour to be processed. And you can see that that one I had named vet grams and now we're doing grand vet. You can, although you can't delete old queries you can hide them if you want to. So I'm going to stop there I know we have very little time left but I'm happy to answer questions and I can stick around for a while if you'd like to stay and ask questions. Thank you very much Michelle. Definitely tools I use a lot, especially if we've been using it quite a bit. See if you could put, if there are any questions feel free to use the chat. There might have been a couple. I see Jeremy's note that Excel now allows about a million rows. So maybe you could open some of those in there. You could open it but do you want to. I have a question all together. Yeah, I guess one of the things that we were wondering is that there are as winners. If you had resource recommendations beyond the things that are in items for you mentioned some resource recommendations on your slides but places that people could go to learn more. Yeah, so I will go ahead and share my presentation again. Let me just point out what's here. So these are resources from the Census Bureau that are helpful for the material that's in MDAT at this point. The step by step guide is just a PDF file of the different steps that you go through in MDAT. What you can talk through is super helpful and gives you important information when you choose a continuous variable like income that could go anywhere from zero to infinity. You know, not quite infinity income but anyway, you must define custom ranges in order to use a variable like that. And they use income as one of their examples there. So that's a really helpful thing to look at. And then the handbook is also a new resource since the spring I think they put it out in April. So much more detailed discussion of the accuracy of the data in the handbook. Then as I said, technical documentation. As I was preparing for the webinar there were a whole bunch of circumstances in which I found a variable that I thought I wanted to use and then I was like, I really don't understand the categories for that variable. And I don't think I'm going to take the time to explore those variables in that detail. So just be aware that you will need to refer to the technical documentation, either for the microdata, the decennial and ACS pumps and the current population survey separately. I'm pretty sure this pumps specifically links to the ACS pumps and not the decennial poems. Then there are special resources for geography, both the tire web mapping app and the static reference maps. For IPMS, there are these two resources for IPMS broadly, whether you're dealing with IPMS USA or CPS, you'll need to know how to create an account and once you create an account for one IPMS tool, it applies to all of their tools so you won't have to repeat that. And then, although the blog explanation about waiting is super simplified, it does link to other resources that may be helpful. IPMS USA specific helpful resources and then CPS helpful resources separately. Great. Thank you so much. My cat is sitting on my keyboard. The wonders of virtual life. All right, well thank you so much.