 The first one is Katie. She manages the IPIMS USA Project and directs dissemination and outreach at the Minnesota Population Center. She's also an economist researching family circumstances and labor market decisions, times filled with partners and well-being over the life course. All questions that can be analyzed with MPC data. And then Catherine Fitch is the associate director of the Minnesota Population Center and has been working with IPIMS and related census data infrastructure for more than 20 years. As a historical demographer, she is interested in family formation and measuring change over time. And IPIMS makes that easy to do. So we definitely, it's exciting to have two researchers who actually use the data that they're talking about today. And I will hand it over to you. This is Kathy Fitch and I'm here with my colleague. Hi, I'm Katie Genetic. Thank you for so many of you to join us. We will start off with some overview material. We're at the Minnesota Population Center at the University of Minnesota. Just a little background about what we do here at MPC. First, we're a university-wide interdisciplinary research center. So kind of what you might imagine as your standard interdisciplinary center. We're trying to bring together folks on campus who study various population issues and we post a seminar series and other programs to connect people on campus. And then we are always training graduate students. We have a population studies training program and minor. But mostly what you might, if you know, have heard of us, but how most people know the Minnesota Population Center is because of our data tools. We have been building data access systems and creating harmonized data for more than 20 years. So I'll go from here at the beginning. Close. And we are the home of IPHMS and several other data projects. Now just a note on that that we say IPHMS here. But out there in the world, you might have heard someone say IPHMS and Katie and I are totally okay with that. IPHMS, IPHMS, as long as people use the data and cite us appropriately, we don't care what you say. This is the web portal to all of our data sites. You can reach it at iphomes.org. And Katie is going to be walking through and giving you some background on these projects after I give you the broad overview. A lot of the data we provide are census and survey data. And we'll admit you can get them from other places. So kind of the question of like why would you get data here that maybe you could get elsewhere. We also have data you can't get elsewhere. But things like the US Census, there's lots of places to get that information. It's publicly, you know, freely available online. The main thing we do is we standardize data across time and place. We want to make it easy for people to analyze more than one year, more than one place for our international projects and do it easily. In addition to that, we add some fields in the data that are helpful to researchers who are identifying a lot of our data, our census and surveys of households. So we figure out relationships between people in the households in useful formats. We have some other geographic fields that we help that are useful to researchers anyway. So it's not just what you get from the Census Bureau. There's some extras there. We also provide the data in multiple formats. And Katie's going to dig into this a little bit how to access our data. There's a couple different ways to do it. We have excellent documentation that highlights integration over time. So what you should know if you're going to, and we love to see this, you know, an analysis from 1900 to the present or some big broad sweep of time. Sometimes concepts have changed a little. So we're out there to help people understand how that might have, might affect your analysis. And finally, we have excellent user support teams that are there to answer questions by email or on our forum. So to the start off of what we're aiming to do, just to give you some background since some of you, you know, who's out there using our data. Well, we're just at the economics meetings earlier in January. And those folks are our biggest group of users, almost a third of our users identified as being economists either by training. Demography and sociology is also a big group, GIS and geography. The history numbers surprise me even though I'm a historian. And you can also see we have a good group of people in policy in the government using us. We have journalists. So you can see our data being used in a lot of cool online widgets. And then finally, you know, non-academics and private industry. But so we've got a pretty broad group that's sort of academic discipline here to give you another sense of how we think about it. We've got, you know, independent scholars like graduate students, undergraduates, faculty members, people doing local policy research, journalists, government employees. So a few ideas of how people are using our data. As I have put in my bio and Linda mentioned, you know, I study family transitions over time. So we, and that's what the data are made to be easy to do over time and across places. Our international project has, we're disseminating, but now more than 90 countries. Yes, 92. 92 countries of data, you know, it's coded similarly. So you can analyze all sorts of phenomenon. People use it to study public policies. So you could have, you know, like a state level, a state policy that goes into effect. And people can either look at the, look at just that state, look at that state compared to other places. There's lots of ways to study public policy. People also grab our data to add contextual information to other survey data and to studies. And finally, we have a growing collection of boundary files for mapping. So even more GIS analysis going on. So I have a quick background about where, where did we get if-ons? The I stands for integrated, consistent codes, labels, and documentation. The public use, it's anonymized, downloadable data. And it's microdata, which means it's at the individual level. We've got a couple slides on talking about what it means that this is microdata. And the series, it means the data are pulled over time and place. So we have this concept to start off with because the majority of our projects are microdata. So that means it's at the individual level. We capture individual survey responses for people in households. And we have all of those responses in one data set. So instead of your fact finder, what is poverty in my census tract, you can look at building your own tables and running multivariate analyses from individual level data. I think we have a slide here that shows what it looks like. Here we go. So just looking at the record here, imagine each line as a person and we've, you know, this indicates here the codes for the relationship to the householder, the marital status variable, literacy and occupation. And obviously there's lots of other information in there as well. But microdata are a little, so there, it's, it can be more difficult to use than something, than summary files where you're looking at information about places. But it gives you so much more flexibility. If you've ever been looking at a table and you've thought, oh, they made that break at just the wrong, I don't need that age group. I need a different age group. Or you don't like how the categories have been constructed. The microdata allows people to get at every bit of possible detail there is and, you know, construct your own occupation measures, for example. So the one trade-off that was on the last, I'll flip back here for a second, that we've delved into. So the one problem to release this kind of data, there's a fair amount of geographic suppression so that individuals can't be identified. So that means you're looking at places of always 100,000. Some counties have 100,000, so there's some counties you can identify with that. But sometimes it's groups of counties and places. So that's a big, yes. I see your question here on tract. Is that the last one? Yes. So tract-level data from us, we'll talk a little bit about NHGIS. We do offer that. But for the most part, the microdata, you have to be able to deal with bigger places than tracts. And that's the modern microdata. We have a lot of historical microdata where we can see where people live, sometimes even down to the address or sort of like tracts. I mean, there aren't always tracts historically as well. We don't have to have all of it yet. But to be clear, there's a break in the U.S. data. Data that's been released by the Census Bureau, which is data from 1950 forward, has this geographic and other suppression. They don't release other very unusual codes that would make it easy to identify people. So that's the overview on microdata. Now Katie's going to go over, for the most part, I'll say this. When we started, the only way to access our data was to create a data extract, download it, and analyze it in a statistical software package. We now have ways to do online analysis with the SDA program. So those of you who are ICPSR users might already be familiar with the tool. And we're going to talk a little bit about that. So that makes the data a little bit more accessible for folks who don't want to download and analyze themselves. I turn it over and Katie will start off with our biggest project and our oldest project. Hello. So I'm going to walk through all of our microdata projects and let you know what's in the data, basically what the data covers. We figured since a lot of you are coming from a data librarian background that it would be really useful to know what you can do, what is in this data. So when people come to you, you can show them, send them to a project that might be best fitting for them. Following the microdata projects, we'll talk a little bit more about aggregate data, and then hopefully we'll have time to actually go analyze some of this data online. IPIMS USA is our sort of longest-standing data project. It's been around for more than 20 years. And we have U.S. decennial census microdata samples going back from 1850 all the way through 2000. So every 10 years, we have a sample of the population that's representative that you can do analysis on. There's one exception to that. Actually, 1850, 1890, sorry, is missing because the records were burned in a fire. As Kathy mentioned from 1940 and before, we had complete information on people because that data is public. So if any of you have gone to ancestry.com and searched your family members, we've actually partnered with them to create full count census microdata for research. And we started that project and we have complete count data for 1880 and 1940 available online. We've also created some linked sample data from 1850 through 1930 using that 1880 full count file. We have samples from Puerto Rico. So Puerto Rico has a slightly different census than the rest of the U.S. and we make microdata samples for Puerto Rico available as well. And finally, we make the American Community Survey data available. So the American Community Survey replaced the long form of the decennial census starting in 2000. So now everyone, like a few years ago, got that teeny little census and they were like, oh my gosh, there's nothing on here. That's because it's possible that you've been surveyed with the long form but via the American Community Survey. So this goes out to about 3% of the population each year. And the American Community Survey data comes out yearly. The microdata is a 1% sample each year and it's our most used data set at the whole entire center. There's a question which I will answer with my next slide. Because our upcoming USA data projects include all of the complete count data from 1860 through 1930. So we're continuing, you're correct, we're continuing to work with Ancestry. They're helping us digitize the data and then we get it ready for analysis for people. So this year we'll be releasing 1850 full count hopefully within the next month or two and hopefully 1930 by the end of the year and we'll slowly go back in time. It's not that slowly over the next five years. We also have in collaboration with Ancestry a 1790 through 1840 full count data set. It's a little bit different because the census back then was household data. So it was sort of like a tally you can imagine of like the number of white females living in your house and the number of white males. And so it's a little bit different than the individual level micro data with the rest of the censuses but we also will be making that data available. Actually even maybe as exciting as these other projects are the work we've been doing with the Census Bureau on creating the 1965% sample. So we recovered some data with them and we currently have a 1% sample for 1960 but now we're going to have a 5% sample. And this sample is pretty special too because this is the only micro data sample where you can get down to geographies of 2,000 people. So that's pretty exciting. That'll be going live within the next few months. This brings me to Ipon CPS. So this is the current population survey somewhat related to the census data. It was administered starting in 1940 and it's the primary source of information of employment in the United States. So when you hear every month of jobs numbers like if you're listening to NPR and they talk about the jobs numbers reports, this is where they're primarily coming from. It's a household survey that was designed for this purpose to measure unemployment and almost 60,000 households are being interviewed monthly. In addition to asking about employment and unemployment and basic demographics, each month almost they ask additional questions for other purposes. So this is at the bottom there when I'm saying additional supplements are given throughout the year. They ask lots of questions throughout the year to people and we are making all of this data available. So currently we have basic monthly surveys. That might even be wrong now. We maybe just released back to 76. But for sure in 1989 through 2014, we have all the ASEC data, the annual social and economic supplement data. So this was previously known as the March demographic sample. It's the most used data by social scientists from the CPS. And it goes back to 1962. It's actually slightly larger than the other monthly surveys because they asked it in a few months. And it asks a lot of questions related to work and demographics. In addition to the ASEC, we have other supplements we've been making available through the IPHOMs. There's voter registration, fertility. The computer internet usage and computer usage one is pretty interesting. Food security is really popular. The tobacco use supplement is also used a lot by health scholars. So in this data, similar to the census data, that you can get down to state level identifiers, some metro areas, and some counties. Again, it has this sort of threshold of 100,000. The next project is IPHOMs International. So this is the world's largest collection of publicly available individual level census data. So sort of like the U.S., how we have samples in different years of the census. We have census samples available from 82 countries around the world. And a lot of these countries have data going back to 1960 through the present. We're actively working on new countries. I think we have agreements from over 100 countries to process their microdata. And we also archive it. So a big part of what we do is help countries clean and archive their historical data. So we are continuing to work with other countries. And we also have agreements with these countries to process their most recent amount of data. So a lot of our data production across these 82 countries is getting the most recent census or the most recent data available from each country online for researchers to use the data. I'm going to have an incomplete list here, I think. So this list, I think it's actually missing our new ones. And you can see, though, there's tons of countries. You can go online and see them all. This is a fascinating data source that not enough people are using. So you should send people in this direction anyone that's really interested in doing international research. So here's the map version. So the blue countries are the countries that we're currently disseminating their data. The green ones we're working on and the gray ones. In some of those cases, we're still talking with them to try and work out contracts to archive and disseminate their data. And in other countries, they actually don't have census microdata. Somewhat related to Ipsum's International is our North Atlantic Population Project. So this is sort of the historical version of Ipsum's International going into the deep history, not 1960s. But there are actually 32 historical censuses from sort of in the northern Atlantic part of the globe. So Canada, Great Britain, Iceland, Norway, Sweden, and the U.S. We have data dating all the way back to the 1700s and all the way up through 1950s. In this currently in NAP, there's complete count data for 14 censuses. And when it's complete, it will contain 130 million individual records. What we're really trying to do with this project is link people across time and countries over this period. So within countries and then across country migration using these complete count data sets and samples. This leads me sort of away from our census data to our survey data. So we have a number of U.S. survey projects and international survey projects as well. The first of which I'm going to talk about here is the Integrated Health Interview Series, which is the National Health Interview Survey. This is a principal source of information on health in the U.S. population. It's been going on for a really long time, back since 1963 through the present. Households are surveyed each year. So this is not like the NHANES data, especially down in the south where you have all this data, where people go sit in a van and weigh them and take their blood pressure. This is self-reported survey data where they're asked many, many health questions, and it varies a little bit from year to year. There's about 45,000 household samples per year. In total in this database, there's more than 13,000 variables. They ask loads of questions. They vary over time. Because this data is really, really sensitive, the geographic identification in this data is only census region and the public data. And I'm talking like four regions of the country. You can't get state-level data unless you go to the restricted data. And you can bring our integrated data into the restricted data because the data we have, we've integrated it over time. So you can easily compare health information from 1963 through the present. See the information on there, but it's various health, many health questions related to individual-level health. So one of our other survey projects is the American Time Use data. This is sort of related to the current population survey. And it's funded by DLS and fielded by the Census Bureau. If you've been in the current population survey, which is actually a household panel, then you may be asked to be in the time use survey. This is a 24-hour time diary where respondents are called on the phone and they asked what they were doing, who they were with, and where they were over the past 24 hours down to the minute. It's a pretty extensive survey. It takes a lot of time and energy to run, but it provides us with very rich data on people's activities. And we also have information on the respondent and members of the household via the current population survey. So there's a lot of demographic information in there as well. Our international survey project is called the Integrated Demographic and Health Survey. So the demographic and health surveys have actually been around for quite some time. This is a partnership with Measure DHS and ICF International. They've been collecting this data for many years, which is really focused on women and children, as well as other partners as well, from over 90 port and middle income countries. There's over 260 surveys total in the DHS, and we are in the process of integrating this data and making it available online to users. It covers a wide range of topics. Again, really focused sort of on women and children, specifically family planning, health, domestic violence, a lot of focus on HIV and AIDS over the past few years, and healthcare in general. The soon-to-be-release project is Iplum Tire Ed. This data should be released within the next two months. It contains data from CSTAP, which is the Scientists and Engineers Statistical Data System. So this is three surveys of U.S. residents with science and engineering degrees and occupations. This data is particularly hot now with a lot of interest in STEM fields and where people are going with them and gender equity in these fields. And this survey is an NSF survey, so a National Science Foundation, and it's given every two to three years. It started back in 1993, and we have demographic information on people, education and work history, current employer characteristics, and their earnings. So this is going to be used a lot by education scholars, and we hope that by integrating this over time, it will make it easier to use for researchers. So those are all the microdata projects, and again, we sort of breeze through them all. All of the information on these microdata projects is available online. If you go to iflums.org, which we will go to here together, you can get to every single one of our data projects. We've been disseminating data online since the 1990s. We were one of the first sort of data producers to put our data online, and everything I've talked about here can be found online. So we really focus on documentation, and we hope that everything is documented well enough that you can go online, or any student can go by themselves and get this data. So this brings me to my summary data project. So we just have two of these, and this is what summary data looks like. So this is an example. I pulled this table from FactFinder 1, actually. So it's a little bit dated here, but this is the total population of the U.S. in 2000, broken down by sex and age. So FactFinder is sort of where we think of to get the typical aggregate data, so census tables that have been produced, and I have this here for the country, but you can go down to your census block or your tract and pull out different information that the census has created tables for. We also have summary data from the census via our NHGIF project. However, unlike FactFinder, which only has the last year or two available of data, we have the summary tables from the census, going back to 1790 through 2010. We have every sort of table that they've made available online, and we've also created GIS boundary files for the geographies when we have been able to throughout time. This is a big wealth of data that more and more people are using. We also have ACS data, and we also make the blocks. I think block groups are blocks available, which you can't get from FactFinder for ACS data. And I've heard about other people doing kind of wild things to try and go get that data, and it's really easy to go get it from NHGIF. One of the crowning achievements of NHGIF, too, is that we've created these time series tables. And as many of you probably know, when you try and use aggregate data over time, it can be really difficult because they change how they group either geographies or they group sort of income chunks or age groups. And what we've done is really tried to reconcile these changes over time to create time series tables that we can compare so you can easily look from 1970 to the present over time at some of the published tables. So I encourage you, if you use this data or if you know people that are trying to use aggregate data over time to head to NHGIF. So, and I keep calling this a summary data project, but one of our newest projects is actually a summary and micro data project combined, and it's called TerraPOP. The goal of TerraPOP is to preserve, integrate, generate global scale spatial temporal data. So we have micro data in it, and we have aggregate data, but we also have government land use statistics. We have land cover data from satellite imagery, and we have climate data, so temperature, precipitation, and cloud cover. So this is from around the world. We have a lot of data from the U.S. in this too, but we, again, we're using all of this data we have from around the world to merge land use and land cover and the climate data. So this is really an environmental GIS and people project. Currently, I actually think this is completely out of date, so I apologize, but it's this and more. This is a very much in progress project. So we have micro data. We have aggregate data for all of these countries. We have GIS boundary files, I think, for more than six countries now, and we have raster data or sort of this climate and land cover data across the globe for a number of years. So this is really exciting for people who are doing sort of mapping and environmental research. So those are all of our current data projects. We're currently disseminating those. As you can see, we're working extensively on all of them as we move forward. We also are starting new projects, so we're really trying just to integrate as much data as possible to make it interoperable to do data across sort of disciplines and surveys, et cetera. So in addition to that integration, kind of one of our key components is our data extraction. As I mentioned, we've been disseminating data online for like 20 years now. And what makes it kind of special is this data extraction. So the ACF data, obviously, you can go get the micro data even just from the Census Bureau, but people come to us because it's so easy to use, primarily because of our data extraction tool. And what it does is allows users to create custom data files. So you can pick any samples you want, as many or as few, and any variables you want, as many or as few as you want. So this is huge, especially for students. And even in the new space is cheap now, this data is pretty big. Micro data can get huge. I mean, these 100% data files are crazy. They're even hard for me to run on the server here. But if you want to look at 100% of the population, but only look at children, our system allows you to do that. So you can just look at that one year, look at children, and look at just the variables you want. It makes it much more manageable. And this has allowed us to get these data in the hands of many more people than otherwise would have been able to use them. In addition to being able to sort of create this custom data file, our system creates a custom syntax file for you to read the data into a statistical package into SPSS, data, SAS, and it creates a CSB. So you can read it into R or even look at it in Excel. All of the data is labeled within it. All of the values are labeled. And there's a code book available that's custom to the data set you've created. There's also a record of this extract on your account. So when you make a data extract, it gets created on our system fairly quickly. Sometimes within a couple minutes, sometimes within a half hour. And you get an email that is ready. You then go download it onto your own computer. And it'll sit online for about three to five days. After that, the data is no longer available. But you can always just resubmit that data. I have actually gone back and resubmitted data extracts from 2003. So it has exactly what I put on it. And I can just go get those data, those variables, and samples immediately. As I mentioned, there's some pretty great features with this for usage. One of them is the case selection feature. So if you're only interested in looking at women or elderly people, you can easily create sort of a subset of the large data sets for download. We also allow people to attach variables within the household from other household members. This is a little complex to explain, but basically anyone doing family level or household level research finds this immensely useful. You can also use a custom sample size. So if this is really too big, we will pull a random sample for you and reweight the data within the extract system. Finally, all of our data sets are really similar. And we've done that for our own purposes to make it easy to process all this data, but also for users. So if you know how to create a CPS or a USA data set, then you can go over and start using IHIS really easily. And this is just a nice thing for our users, because how all of them operate is nearly identical. As Kathy mentioned, in addition to being able to look at this data within your statistical package, we can actually look at, analyze the data right online. We did not create this system for doing this. This actually was created by developers at UC Berkeley. And it basically harnesses the Internet to crunch this microdata. You can even look at multiple years of data online. And I'm going to demo it here so you can just see us go through it, because it's pretty cool. And there are help guides on the webpage as well. And we have online tutorials to train you on it. This is a really great way if you just want to look at something quickly or you want to investigate something before you actually dig in to the data. So I have two examples here. It's already 11.40, and I want to give time for questions too. So I might go to the second one, because it's a little more fun. So I did a little Googling since we are working out of North Carolina here. At least that's where Linda was from. And it turns out Greensboro lands, if you do some searching, in the top 25 bikeable cities. So I decided maybe we should use the Census microdata to see how many people actually bike commute in 2014. So let's just say you were curious about this. You saw this, like you heard it on the radio, and you were like, wait a second. Do people actually bike to work? No one in my office does. Or if you work in an office like mine, it's like half the people bike to work. Is this remotely normal? So what we're going to do here, what I'm going to walk through, I haven't looked at this at all. I actually have no idea. We're going to go check it out. I have no idea. I haven't done this. So I am doing this basically live just like you are. So this is what I would do if I wanted to see if people really are biking to work in Greensboro. And we could actually maybe compare it to somewhere else. There's a couple of questions right away. We need to ask ourselves, can we even identify respondents in Greensboro? Remember, Kathy mentioned to you this 100,000 limit. This is really tough, like partly because sometimes these areas, which maybe you guys know about PUMAS, these public youth microdata areas, they don't always align with cities or counties. So even if there are 100,000 people in your cities, you maybe can't identify them in the data. We have to figure out if we can identify Greensboro. Then we need to figure out what information there even is on commuting in 2014. Can I actually look at this using this data? And who gets asked this question? This is just important for our analysis. Often in these surveys, there are a whole set of questions that don't go to people. We don't ask 12-year-olds how much money they make. So how do we identify them in the data? So I went directly to the IPUMS USA home page. You can see this is located at usa.ifum.org. Now there are sort of two key things. This website is being updated as we speak. So however we sort of like this throwback 1990s style. So the two big things you're going to do are under this data link on the left. And you can sort of see my bleeping cursor over them. So this browse and select data is important. So I'm going to start with browsing and selecting data. I'm going to open it in a new tab though, so we can keep this open. So I've clicked browse and select data, and I get this somewhat cryptic looking select variable start here. This is a way for you. We are actually in the extract system if you wanted to make a data extract. But this is also where all of our metadata lives. There's so much information about our samples and variables here. I know we're just interested in 2014. That was my question. So I'm actually going to start with select samples because I'm just going to look at the 2014 ACS data. So this is the American Community Survey data. All of these, this is a link. So I can go to the ACS and learn more about it. But for now, we're just going to check it out. So I'm going to submit my sample selection so that now I only get ACS. So my first question was whether or not I could identify green throw. And the variables are broken down into household and person variables. And you can see the dropdown menu here about sort of where these type of categories the variables fit in. Obviously for me here, I'm interested in a geographic variable because I want to see, look at green throw. So I pulled down geography there. And there is this metropolitan area, 2013 OMB delineation. So this is what I'm going to go check out. So here I am. This is where what we call the variable description area. So you can see here I have tabs across the top which gives me a description of this variable which is kind of long. There's a lot of information. There's a codes page. There's also even a case count view. So I can see what codes are available in the 2014 ACS. I can also see how many people in the sample are in each of these areas. So I'm going to turn that on right now. There's also some other information here. There's questionnaire text. There's what samples it's available in. There's the universe. As I mentioned, some questions are not asked of everyone. Every household is asked while we know where it is because that's what it's sampled on. And any comparability issues we have found in using this variable over time. But what's really important for me here is this codes page because this is a metropolitan area and I am looking for green throw. So I'm going to go ahead and go ahead and look for green throw. So I'm going to look for it. Here we are. Green throw 24660. And you can see there's about 3,600 people in the sample. So I think this seems like a pretty decent sample size too to be checking on if this is a really biteable city. All right. So now that I have gotten the, we can identify green throw with this Met 2013. So this is a metropolitan area for green throw. I'm going to go back to my dropdown menus and look into this communing variable. So obviously I know that I can do this with the ACS. And part of the reason I'm doing this question in particular is because I think this transportation information in the ACS is fascinating. It seems a little obscure except for that the transportation policy makers use this data extensively. So urban planners, the transportation people, I've given a number of talks to them. They all use this place of work and travel time information all the time in their city planning. So you can see here there's a couple different variables they've asked. Means of transportation to work. If you car pool, your travel time, time of departure for work, and time of arrival at work. We're going to check out that Tran work variable. You can see we're back at the variable description for this. Tran work is the respondent's primary means of transportation. We can handle that. Here are the codes. So I click the codes page. So here are the codes so we can see I'm interested in biking. So that's 40. So it's a Tran work code of 40. I mentioned that we were going to check the universe. So the universe in the ACS is persons age 16 plus who worked last week. So the universe thing, that means everyone under the age of 16 who was not working last week did not get asked this question. I'm going to go back to the codes and I see an NA code is 00. So everyone in that group is getting a 00. Just something to keep in mind when we're looking at this. And finally, just for fun, we'll look at the questionnaire text. How did this person usually get to work last week? That's exactly what was asked in the 2014 ACS. This is really useful for people when you're looking at things over time to see how the actual question changed. So I'm back at the Ipsum's USA homepage here on my other important link, which is Analyze Data Online. You can see my little cursor on the left panel under the data square is over it. And I'm going to select it and we're going to go to a whole new page. So bear with me. I'm going to click it. Oh, my gosh, I lied. I have to do one more click. So then we get to this. There's some information on analyzing data online. You can see there's an instruction, video tutorials. I'm going down to the 2014 ACS. This authentication is your registration information. To use this data, it is free. All of our data is free of charge, but you do have to register to use it. And once I log in, here is what it looks like. Also kind of a throwback to the 1990s, but it's super powerful. So I'm just going to show you this so we can see the data. If you are more interested in using this, obviously we have video tutorials, help guides, and user support that you can email, which we will touch base here on a bit. So I am going to, I know what I want to do. I still don't know what it's going to look like. So I'm going to look at this variable, Tranwork. And I am going to use some selection filters here. So I'm going to ignore everyone that's not in the universe. So I'm going to ignore all those zeros. So I'm going to do that. I'm going to look at just Greensboro, and then I'm going to apply this person weight. So as I mentioned, all of the Census data are samples that are nationally representative. However, due to sampling and over and under count, it's not, they're not all flat samples. So even though it's a one percent sample, not every person in the sample is equivalent to 100 people in the population. So the Census Bureau or the National Statistical Office generally provides weight. So it can account for under representation, over representation. Oftentimes in the Census, they over sample rural areas, because not as many people live there anymore, but they still have a lot of policy and important things that we need to be aware of. So the weight accounts for that and will make this nationally representative. And at least this case, representative to the Greensboro area. We're on the table. 0.1% of the population gets to work on a bicycle. 398 is the estimate for all of Greensboro, which is estimated to be about 350,000 people in 2014. 3.9% are working from home. The majority of people get there by auto, truck, or van. Look at those walking numbers. Walking only, that's pretty good. We should, what I'm going to do now, I am going to take off the weight once, so I'm going to go with none, so we can just see in this sample. This is kind of important when we're looking at these smaller areas, so what we're making these claims on. So two people on the sample, send motorcycles, three to actually have five bikes, 16 miles per week, because I don't walk. That's very good. So there was just a little preview of something you can do with our system. Obviously, this is just an overview, so I am going to try and unshare and go, oh, maybe I do this. I just sort of wanted to wrap up here. We'll have like five minutes for questions here, but we do have a YouTube channel with many tutorials about how to go through our extract system, how to analyze data online, how you can utilize our system to do a lot of different things. We also have online trainings, so there's training materials at this ridiculously long website, and these are, there's online modules that you can walk through the data, or you can, there's exercises you can print out to learn how to use our data within Excel, SAS, SCSS, data, and our online analysis. So encourage people to go that direction or send people this direction to utilize these. And finally, we have a user support team that answers user questions around the clock. It's personalized help. Definitely within like 24 to 48 hours, you will get an email back, and they're incredibly helpful. They will help you if you just can't figure out the website too, if you think you found an error in the data. If you find an error in the data, we actually send users logs, so we want to hear about anything you discover in the data and make it better. So with that, I'll open it up for questions, but then also if you have any questions you'd like to ask more in depth, you're welcome to email Kathy or I or email the IPHMS line. Yes, so you can, the best way to do it is just to register for one of the datasets. You can also email if you don't want to do that, but if you register for IPHMS USA, you will be on the email list and we'll send you any updates about when the data is becoming available or once it becomes available. Yes, you can use R, so we don't make right now like a program file to read our fixed-wiz files into R, in part because our data is a little hard to use R on a computer, like a personal PC. It tends to crash them. R doesn't handle long datasets very well. However, if you have someone that's using it on the server, you can easily do that and you can take our CSV file and we will help you with that or you can do this data file and do it. The other thing is there's some pretty cool new things out there. I will type it in. There's a great tutorial from MonetDB that's about taking like the large IPHMS international data and making it possible to use it on R. It might even be MonetDB. So here are the YouTube tutorials. And these are on all of our websites. If you go to the help section on our website, there's a link to our tutorials. Thanks so much. I really know how much it looks like you do to direct people to resources like ours. So yeah, feedback is always welcome. If you have a student or you send someone to us, then you feel like they don't get what they need. Please contact us. Let us know what we can do better.