 So welcome. My name is Linda. I am the chair of Godort and the senior data librarian at the Cornell Institute for Social and Economic Research. And I'm happy to invite you are welcome you here today to be help on an accidental government information library and webinar series. We are this was traditionally run by the North Carolina Library Association's government resources section. And now it is a Godort or the American Library Association's government documents around table will be taking over the series so for the next semester we'll be doing the webinar short jointly with GRS. And then in the next in the summer next year, Godort's going to take over. So we're very excited about this. It's always been a national level level webinar series. So we're hoping to tap more into the people across the country across the world who are interested in government information. If you have ideas for topics or if you would like to present on a topic that's close to your heart. Please get in touch with me. I'm going to chat in a minute. And we can talk about topics or presentations that you have in mind. We also put in the chat in a minute, our YouTube channel where you can go and see past presentations on a wide variety of topics. A lot of data because I'm a data librarian so I tend to have a lot of data friends like Bob Ray. So we but we've done a wide variety of topics like legal issues. More traditional government documents or government information topics. Lots of things. So thank you for coming. I'm happy to have you all here. You will be muted and video after an application we have 80 attendees which is fantastic I'm so excited about this. So that's a lot which means that this will be. We will need you to chat your questions in so if you want to use the Q&A, I believe only we will see those questions. But in the chat, everyone will see this question so if you are not a stickler if you want to use one of the other that's fine with us I will keep track of them as Bob Ray's presenting and make sure that your questions get answered. So feel free to chat with us either for Q&A or the chat. Yes and Sam just Samantha just noted that if you want to send a chat do make sure that you send it to all panelists and attendees otherwise you might choose just me. If you didn't have a tech problem. Okay, let's just chat. We'll just chat today that's fine. So, if you do a check question you can choose either myself or Samantha from the list, and we can help you with whatever is going on. If there are technical or you can also email me and I'll put that in the box again. Worst case scenario, we will be recording the session and have that available as soon as possible on our YouTube channel. We will have our webinar in February. So February 18. Julia as though we'll be doing reprise version of her API is a data.gov, which she did four years ago so it's, it's stated. But it's, it was the most watched webinar on our entire YouTube channels had like, almost 2000 I think over 1000 at least views so we thought that would be a great one to bring back. And then I were from March on we need some topics so if you have topics let us know we were always looking for people to put that we will have us, we just scheduled civic switchboard for September so I'm very excited to have them come and talk about the data. So, I'll put all that information in the chat and you should follow us if you're a YouTube user and stay tuned for more data and government information. So today's webinar is the current population survey. On the current population survey which is the labor survey for the United States and we're going to do a little switcheroo and while Bob Ray is getting settled. I'm going to talk about him so Bob Ray board of on who I think many of us know in the data community and the government information and the business community. So, he's from Princeton University is the economics and finance librarian in 1993, and has led data and statistical services since 2004. He has leadership roles in various data and library organizations, including a la and I assist the co teachers providing social science data services strategies for design and operation as part of the ICP ICPSR summer program, and serves on the ICPSR Council and the public advisory committee. He has been honored with with the William H Flanagan award for distinguished service as an ICPSR official representative, and the Thompson gale award for excellence in business librarianship, and you're very excited to have you here today. Thank you so much Bob Ray. Thank you Linda and thanks to all of you and it's nice to see so many of my friends and colleagues out there. I'm going to start sharing in a second. But I'm going to say a few things off the bat. So I know it is really unfashionable to do a lecture and not be interactive. I am not being interactive. There is no time for that. Next thing I'm going to say is, how can one survey take up an hour. Well, the University of Minnesota does a three full day workshop just to scratch the surface surface of this one survey. So there's a lot of detail. This was normally said but wasn't said so yes, the presentation and the note and the PowerPoints will be shared. There's also a lot of notes that I'm going to share because there is no time to include even all the things in my abbreviated version that I wanted to do. I've done three run through sessions and I had to cut 75% in order to be able to fit it in and have time for q&a. And then the last piece is that please do include your questions and so forth. Linda will let me know or Samantha if something has gone terribly wrong where you're not seeing things, but otherwise I'm going to save questions for the end unless it's something that just can't wait. Just because my experience is often with myself I ask questions that are then answered later. So what I'm now going to do is I am going to start sharing and we're going to attempt to get through as much as possible of the current population survey. Alrighty. So what we're going to talk about today a little bit is a lot of definitions what it is what it's not who's included a little bit of the history but again much of the history is going to be in notes that you will see later. Various questions that are asked the supplements because I also want to talk about why one would use this for things other than labor. We're going to get the actual data from we're going to do a demo from my favorite source of it I pumps and if we have time we'll talk about other places to look for labor data, and why one would use those. So for the first one, I am going to actually do something that's really bad which is reading, but these are legal definitions and this is really important. Alright, so unemployment. So, a question that we often get is, is anybody without a job unemployed. No, and I think we all know that that is not the legal definition so who is the not who is unemployed. So unemployment. It's those that are not employed during the reference week we'll talk about that later, but available for work, excluding temporary illness, and that made a specific effort to find employment during the four week period, ending with the reference So basically it's asking for the last month, or waiting to be recalled to a job from which laid off, they need not have been looking for work to be classified as unemployed with that category, or waiting to start a new job, but actively looking for that job, essentially within the reference. And if you do not fit any of those categories or employed, then you are not in the labor force. Easy, those are employed, those that are unemployed. So that part's pretty straightforward. Who is not looking some examples would be I'm retired, I'm in school, I have family responsibilities that keep me from working outside the home. Less than 10% of the people in the labor that are not in the labor force reported actually wanting a job. Now it's possible that you fit into a couple of categories overlaps or handle there's a preference so any type of labor counts is in the labor force and that is the precedent. The next question is, Well, why don't we just use the people that registered with the unemployment office to count unemployment. Well, we don't do that. And again, part of it's because of those definitions. But again, it's only those people who have applied for benefits. Every state has its own unemployment office so it is administered at the state level so that's important to note. The unemployment insurance is it's for people who are looking for work. And there are statistics that are collected as part of this program. Now until recently, we didn't pay a lot of attention to the job, the weekly job claims we tended to only pay attention to the once a month when the current population survey would release its data, and that was changed. So, one of the things that comes out is the initial claims and then the continuing claims. So I am going to now show you this very briefly. This is not our real point is what this looks like so the website they just changed it a few days ago. It is updated. This is not the Trump administration but it was literally changed right at the end of the Trump administration, and they did get much snazzier. But this is the one that we're hearing on Fridays in terms of how many people are out there and yes you can get numbers from it, but it has become, as I said, much snazzier in terms of what's out there. That's what it looks. All right, unemployment insurance. As mentioned, it is not used to measure total employment. Over the past decade only about a third of the total unemployed actually receive benefits. There are a lot of people not excluded. I have these different categories on there but probably the one that is the most troublesome is those that are self employed. So, if you are suddenly not able to work you do not officially fit this category, but there are various other groups that are also not included. Another one is disqualified workers. They are not officially due to their own actions so basically you are fired. You are not officially eligible for unemployment insurance if it was something due to your own making as opposed to a furlough or a layoff. This is CPS itself. It is voluntary. This is not the census. This is not the American Community Survey where you are required to fill it out. That said, the response rate is pretty incredible around 90%. The Bureau of the Census actually conducts it on behalf of the Bureau of Labor Statistics, which does the analysis of the Census Bureau collects the BLS analyzes. We get on it the things that we would expect labor force employment unemployment, a lot of demographics so those types of information. Even though county is identified, you should never use it for county. There is often only one person from that county, no one from that county. Personally, I would just have them remove it and make it restricted, but that is not the case. Now, you, it does provide reliable estimates at the state level, and for our 12 largest metropolitan statistical areas. DC is also included territories are not included so we don't have the Virgin Islands or Puerto Rico or other places. We get it out again, our two big ones CPS versus the unemployment insurance unemployment insurance as I said it is a widely used indicator so no it's not the unemployment rate, but it is a good measure because when it comes out weekly so it does allow was particularly to sort of see or there are more people falling for claims. The CPS is the official, but also it reflects ins and outs so it allows us to look what's happening from month to month. Can we compare, obviously not completely because their definitions that are different. They look at different periods. So people do often use them in conjunction one as an indicator. All right, back to the CPS. Who is in it civilians. So that means not people who are members of the armed forces, non institutionalized population. So what are the institutionalized population. Those could be people in prisons or jails, people who are in residential care facilities such as nursing homes. Those are not included in it. Currently, it is people who are 16 plus, it used to be 14 plus but for quite a while now it's been 16 plus. Interestingly, like the census. It's pretty much anyone who is residing in the US that fits those categories that I just mentioned. So including people who are not citizens. People who are foreign born. And the next question just out of anticipation is, does it identify someone's legal status it does not. So we do know that the person. If they're foreign born if they have US citizenship but we do not know whether they are here legal or illegally. And of course, many people. You know, the or here, legally around 60,000 households or included in it. Now, this is where it gets complex. And this is my incredibly abbreviated version. As I said the Minnesota workshop spends literally about an hour on this, what I've tried to condense into a slide. So, how do they actually figure out who's in this, what's changed dramatically and change is going to be what we're going to talk about for the next 10 minutes or so. And this is why when you're looking at comparability you have to be careful. Currently, around 60,000 at its height in 1981 around 72,000. At its low point when it first started 25,000. So it has changed pretty dramatically. How do they figure out who's in it. They use something called primary sampling units. That is a county or a group of contiguous counties defined within state boundaries so it does not cut state boundaries. Originally, in the sample there were 68. Now, there are 824. And just by comparison in the United States there's a thousand nine hundred 87. So not all of these are included. How do they then figure out which then the primary sampling units they put those into strata within a state. One primary sampling unit is selected from each of those stratum. The probability of getting selected is proportional to their population so that does mean larger places are more likely to show up in the sampling. If you can start looking at these, they're 852 that are chosen. They then go from there 506 or in a strata by themselves, meaning that they are self representing and they tend to be large populated areas. They have 46 left. They then have to figure out what to do with those. So they start combining ones based on combination of various characteristics. They look at housing units that have at least three people in them. They look at industries. They look at wages for different industries. They go through all this. They come up with a sample and then for the ones that were not in the large ones because it's a group of them that have been chosen. That strata itself is not self representing because again it's representing the characteristics of all these different counties that have combined. So this is one of the reasons you don't really want to use it for counties. This is shows you how the numbers have changed over time, how many sample areas, how many places were eligible, etc. All right, what is a reference week. This one is a little bit easier. It's basically just the specific week of the month that determines when they're going to actually look at employment unemployment status. Usually it's a seven day calendar week Sunday through Saturday actually it's always a seven day cattle in the week that includes the 12th of the month. But of course there have to be exceptions because otherwise our lives would be too easy. Those exceptions are most likely to occur in November and December because we have a lot of holidays. So there's a whole formula they change it around and so forth. And it is characterized as the survey week or the interview week. It's basically the week after. So that one's pretty easy. All right, this is my readers digest abbreviated version of the history. So prior to the current population survey, there were not a lot of efforts. Then the depression comes along. We know we need to do things various groups start trying. We have some very indirect counts and some of the decennial censuses. Then the 1930s comes along. We're in the Great Depression. We've been in the Great Depression. We're trying to figure out what to do the federal government is not really doing anything so various research groups state governments municipal governments they begin experimenting on some cases with direct support surveys and in some cases a sample of the population. The federal government starts realizing that something now has to be done there's no standardization. If Boston's doing one in New York City is doing one they're not comparable because the measures are not the same. They talk a lot about who will they even count who was willing and able to work what is the gain for worker. And as I said just because there's not really time, you're going to get a ton of notes later that has more detail. We have what is called the enumerative check census. So this has to do with 1937. They start looking at this the work projects administration the WPA is now really on board. First they're looking at local areas later they begin a national basis 1940 comes along they begin a monthly sample survey of unemployment called a monthly report of report of unemployment. So many of you of course you go into your catalogs and you look for these precious government documents so this is what it was originally called. There's a lot of problems still about you know who counts who doesn't count etc. 1942 the work projects administration transfers this process to the Census Bureau they change the name to the monthly report of the labor force. 1948 comes along they change the name yet again to what it is currently called the current population survey. And this is an important name change because it's actually starting to say this is more than just labor, there's a lot of other stuff that can happen. In 1959, the responsibility for analyzing and publishing is transferred to the BLS, but the Census Bureau does continue collecting the data. Now we're into the 40s and 50s. Big things that are happening is a lot of redesigns 1953 this is a really important one. It's what's called the 484 rotation. So what that actually means is if we get a group, we bring them in, we interview them for four consecutive months. They leave the sample for eight months, and then they return to the sample for four months. So what that means is if you were in January through April, you're out May through December, and you're back in January through April. So why that's important is that if we want to follow people for at least a short period, we can do so. And later when we talk about the supplements, the supplements are often, but not always in the same month, and you may be able to get someone in a supplement more than once computers are coming along there's punch cards there's all these exciting developments that are happening so they're able to do more work. So that's what they changed the reference week. And there's a few times this is going to be done over time. We're into the 50s now seasonally adjusted national unemployment rates are introduced so what that means is when you're looking at summary statistics, you have the raw numbers and non seasonally you also now have ones that smooth out. So, for example, Christmas time. There are much more part time workers that are hard so it evens it out so that you're actually looking at things the employment definition once again it's changed. This actually has to do with layoffs and volunteers. Well we've added some new states that they say well why don't we add Alaskan Hawaii and they do. So 1966 comes alone and more changes and one of the big ones is really just for the employment situation which is the big publication that comes out of it, they start aligning it with some of the other data that we have on employment to provide a richer picture. 1960s. Big thing is there's a commission that is appointed to come up with a lot of things you'll see it's going to actually take decades before the stuff is actually implemented. They keep adding clarifying questions etc. What counts. There were changes. Now they sort of start looking at oh there's strikes occurring there's bad weather and so forth. How do we classify these people so the big point of all these changes is that if you're looking over a long periods, who is counted is going to be different. So that is really important now I had mentioned earlier. Currently it's 16 plus so it has been for a very long time, but originally it was 14. Now, if you go to summary statistics from that point forward, they actually adjusted the older data, so that it would be comparable, and it would be 16 plus but of course if I look at a table from 1960, it's going to be 14 plus. They add more questions, and then a big concept comes along called discouraged worker what is the discouraged worker. It's someone who indicates that they want a job. But they're not currently looking because they believe there are no jobs available or none for which they would qualify more probing questions added to make things clearer. 70s. The really the big thing is occupational changes. So, there's the standard occupational classification codes, and these are often changed along the times of new censuses. And so what they're basically doing is trying to reflect what jobs are today so that was really the big thing. There is something called the leviton commission, and they issued their final report. Remember that these were the people who had done things much earlier. We're into the 80s now, much more industry changes occupational classification changes. 90s comes along more redesigns it's not till 1994 that this becomes a fully computerized survey instrument that means it's a little bit easier. More criteria added for discouraged workers. More questions more qualified clarifying more things about volunteering. All right, now we're into the 2000s. Well remember, way back in 1987. The government had decided to go away from the standard industrial classification to the nakes. Now, Census Bureau is on board with this. They're adding those new ones they're adding the standard occupational classifications. So throughout this period. And, and even before quite frankly, every time there's a census, they readjust, because we have a better idea of what is happening with the population. This has been happening throughout the Census Bureau does a good job of documenting there's, as I said, all the questionnaires, the documentation I'm not going to click on this you can look at this later and we'll actually look at some of the questions later. All right, CPS. There are two primary ones. So there is what is called the basic monthly. This is the one that comes out every single month. It is the less exciting one, but it has some very basic demographics, very basic questions about or you employed unemployed, and the reference period as I previously said was the week before. There is one that is very heavily quoted. It is called ASAC, the annual social and economic supplement. It is done in March. It is not the same as the March basic. There is still a March basic. But what this one does is ask you for the entire year. So we're getting a larger period to get a fuller picture still have the demographics, but there's all these additional questions about your income for the year your taxes subsidies you may have received health insurance migration in terms of did you move from one place to another. Remember the 484 rotation so that you could track someone from one January to the following January, let's say, because they were in both. So I have a link here, no time to go into this this is probably about four hours trying to show people how to link files but there is a whole process that goes into this, and just know that it is possible. All right. So, in addition to the basic questions that are constantly being asked, they're all of these supplements. So now what I would like to do. So the Census Bureau just to mention, they actually do have some various guides to showing you what's there, but they don't go back as far as I would like. And so what I have done is to do two of them one is by month, because something's changed is to show so that if within a month. If that month and year correspond that means you have the people that answered all the different supplements. I also have one by name or topic. So every time they ask questions about civic engagement or computer use. Now it changes sometimes by the month. But this is just, I use it to try to figure out what was every time they asked for specific things. So there are a few that are not released like there was one in the swine flu, and it was actually not released. Alright, back to the PowerPoint. Lots of supplements that are done. This is just a few of the examples just to mention, but what I want to say is that not all of these are ones that are paid for or asked by the BLS. It's by someone else. So just some of the examples housing vacancy and voting and registration. Those are sponsored by the Census Bureau. Okay, fine. They're already collecting it. Child support is by the Office of Child Support Enforcement, which is part of Health and Human Services, public participation in the arts is done by the National Endowment for the Arts, unbanked underbanked by the FDIC food security part is the USDA. The ones that are done by multiple agencies. So fertility is both census and BLS volunteers by the BLS and the Corporation of National and Community Service. When we go to the data, it's really important to remember not just to grab the supplement, but you also need the basic monthly because otherwise you're actually not going to have the demographics. The main publication that comes out is called the employment situation. It is typically release the first Friday of the month at 8am Eastern Standard Time. Now, why is that important. Well, I get a lot of people who are looking for they want to know every time it's released, and they want to look at let's say the stock price is right around it, or how interest rates or exchange rates or something else moved. So that's a fairly frequent question in academia. Now, the BLS has done a good job of also providing the summary statistics. So let me just quickly show you some of those, what they look like. So we're going to get rid of these. So the employment situation is their primary summary. This is really interesting to read each month. There's a lot of analysis. So yeah, we can get the raw numbers, but they're also doing analysis as to what's happening within specific industries. Right now, we know nothing is going terribly well. But they're talking about healthcare what's happening they're talking about the leisure industry. And then of course they do provide numbers and they provide a lot of comparative ones this is all free, obviously available off of the BLS website. They also do a really useful one by topic. So I could come in here and now there's of course lots of things on COVID how it's impacting the workforce. And what they're doing is they're just extracting from the survey itself, those numbers that represent that specific thing. They provide lots of tables. So all of you have seen these before, but they break it out by age and race and sex and occupation and so forth. So again, if someone needs just a quick summary of what's happening now or recent. These are all quite useful. Now, I honestly, I'm almost never asked now. I'm always asked over time. And so they do allow you to do that. There is. You can do a screen where it basically just sort of shows you very quickly, people that had Union affiliations, or how many people were married and not working. I often will go to the text files. This allows you to grab many at once. And so that's just another way of getting at the same information. One question that I sometimes get is the machine readable for all the different sources that we go to. There's, it starts in 1962. And so, where is the data from before that. And then even the basic monthlies, not all of those are available and actually Linda and I actually had last year someone who was looking for an older one and it turned out Cornell actually had it but none of the other sources did. So we're looking for the summary statistics at least. Most of them have been digitized they're part of a publication called current population reports current population reports do not cover just labor. But they were regularly included in those and I just want to show you one of the early ones was it's me it's kind of fascinating. So what you do is you go in and you just keep changing your dates to get older ones, or actually newer ones in this case. But for those of you that like me have been around for a while, you quickly realize obviously this was typed on a manual typewriter, and how it looks. And, but they did kind of the same basic things. So they would give analysis they were telling you what was happening, etc. And there are lots of these and they all tend to be topical based. Alright, where do we get the raw data. So summary statistics are great I need to know trends I just wanted to have any people were married and so forth but often I need to know detailed relationships. Well the Census Bureau has recent ones. They're one at a time. ICPSR and the National Bureau of Economic Research have the vast majority, but again it's one at a time so if I only want to know a point in time it's really nice, because I actually I get all of the data for that survey at once. But if I want to do over time. I always use iPoms. So we're all familiar or many of us are familiar with their census products they also do the current population survey. And we're going to go into that. And I'm going to show you some of the data that is actually out there. So, if you have not used items for other products. You do have to register anyone can actually browse you can actually look at all the documentation which is incredibly rich, but when you want to actually retrieve the data, you actually do have to go in. So, we're not going to retrieve the data now as we're not going to have time, but just to sort of show you the process. So we come in. Select samples. What this is actually telling me is at what point in time and which ones am I interested in. So, they divided up the most popular or the ASACs those very detailed ones that are done annually in March that cover a one year period. I don't want those. So if I click on basic monthly, it's mislabeled, it's basic monthly plus the supplements. So, the first tab or the basic monthly they've checked off the last year or so assuming you're interested in that. Perhaps I'm not. I want everything I could do that. I'm going to select everything just to show you. I'm going to go into the supplements. Now, they do not list all of the supplements that they show the major own going ones. I'm going to look at I'm going to combine something with labor because I know we're here for labor but at the moment I'm really interested in something else. Why would I use the CPS for something that is not just labor. So perhaps, one of the things I might care about is food security. This is a big issue right now. We know that many people are going without. So, what I could do is choose those that I'm interested in and let's say I wanted to do over time. So, great. This is going to give me the food security questions, but I also need the corresponding basic demographics. So luckily they checked them all for me. So that makes my life much easier. I say submit sample at the household level. It's going to have things like which state are you in, you know, so the basic demographics. But the food security supplement is going to be those questions that dealt specifically with food and you can see there's a lot. It's showing you for each one was that question asked or not. Now unfortunately I forgot to uncheck the asex. So what I could do is say you know what this is kind of getting annoying, and I don't really want the asex checked off. So, I could say uncheck the asex. So it's not an, let me do a select, unselect all right. So now I will have only those that I'm interested in. And we're going to start seeing questions like did you receive enough food. How much are you spending or are you getting food stamps. Why are you not able to get enough food that you just don't have time that you don't have the money it's enough to give your diet. Even though it's a point in time they're asking you actually over quite a few months, did you receive food stamps so this could be a really useful what do you do you add the ones that you care about. And then at the person level. And if you notice there's actually a supplement. This means they may not ask these as regularly, but it's a supplement to a supplement just kind of interesting. And then there's just a few additional. This case just the weights for us actually it's not. So this is to use the data. Now under the person. This is where we're going to add demographics, like age, sex, race, marital status, perhaps. And then all of the work ones which is from the basic monthly. I can show up here. And then I could get someone's occupation, and they were labor for status and whatever other items I might want. Alright, pretending I've gotten all the things that I am interested in I go to view cart. I can show me all the things that I selected, plus a few I didn't select like it included the survey year because otherwise I'm not going to know what I really asked for some IDs for each person. And what you want to look at is do your variables cut across. And then you want to look at it over time. I'm not going to actually do the data extract but I would just want to show you one little piece for it. Under select cases. This allows you if there's things that you're not interested in so maybe for example I only wanted people who were married, and the spouse was present. And this would allow me to make a smaller data set or let's say I only wanted one specific race or something like that. At this point I would submit the data, and I would get the data. This one I very quickly show you one other piece of this. That is when I go into browse and select data under the supplements again. So most of these you can kind of figure out obviously what they're going to include. But some of the other ones like another one that I've been using quite a bit lately is the unbanked and the under bank so just to that one's perhaps not quite as obvious. Some of the questions that they might ask this is the newer one done as I said earlier for the FDIC. And I forgot to uncheck the food security so it's still showing up, but for this one they're asking things such as, you know, do you have a checking account do you have a why do you not have. You know, or you sending payments or you, they don't say you zoom or Western Union but they're asking questions about money transfers so this is a really useful one for a lot of things. Let me go back to our PowerPoint, you know that we're running a little short on time so just a few of the tips, never forget to include the monthly basics of including the supplements, make sure your variables cover the full period that you need. You know the 484 rule for rotations, and particularly for industries and occupations they've changed a lot of times so you do want to make sure that you're using a standardized one. All right, I'm going to very quickly go through some other sources for things so the current population survey remember great, wonderful demographics and so forth, intended for national but you can use that at the state you really cannot use it at low levels of geography. So what happens when I need other things. So the current employment statistics, this one, it's not about people, it's about the payroll. So it's the payroll survey. So this is coming from businesses itself. It is the non form sector so not agricultural workers but we do have a lot of industry detail and geographic detail. It is a monthly survey so high frequency just like CPS, but no demographics so it's really just within an industry within a geography, what's happening. So my friend Jennifer butchers on the call, and she has done many wonderful things talking about the economic census, and talking about the difference between employer based and non employer based this is employer based only. This is a comparison of the two. The next program that's done is called the occupational employment statistics. This is annual so no high frequency, no demographics so we don't know men versus women or race, but it's very detailed profiles for about 800 occupations, and it's good for states it's good for urban areas, some non metropolitan local area unemployment statistics this one is actually done monthly but they also have an annual supplement. This one actually you can go down to the county level and a lot of cities no demographics but it's the numbers over time. This is an American community survey not officially a labor force survey but they do ask some questions about employment of course, we need to go down to a very low level of geography. We know we have to use the five year samples we've had quite a few sessions on ACS so I won't go much into it. And then my last set I want to talk about just very briefly, or what are called the national longitudinal surveys. This is a series over time that allows me to track people over very long time period so why might I want to do this perhaps I want to look at what impact did once college major have over occupation and earnings over time what impact would having children at different points in time correlate with your occupation and so forth. So, back in the 60s, they looked at four groups of people, young men, young women, what's called mature women and older men. And they followed these for a long period, you can see the dates here when they stopped each, but all of them did stop. And they followed them all for long periods 1979 comes along they said well you know the people that we interviewed that were young in the 60s are not so young anymore. Let's have a whole new sample. So they did. They added those they continue to follow these people to this day. So this is really nice that we can continue getting for all of these groups. The next comes along they say you know what would be really interesting is if we started asking questions about the children of the women of the 1979ers, and they are so they continue doing so, I mean obviously they're asking do you have children, but now they're asking questions about the children. And some of these of course are now adults themselves are young adults. So they've continued doing that 1997 comes along 79ers are not so young anymore we need some more young people. So they've introduced yet a new one. And so the 7090s are continued the children are continued, and then we have a whole new cohort with the 97s. I have some of these are literally thousands of pages so if you really want to know a lot about the CPS do this but I'm going to recommend the last link, which is the icon CPS summer data workshop. It's a three day workshop that goes into much more detail than I can in the period that I have. So what I'm now going to do is stop talking. And I'm going to let my good friends Linda and Sam, tell me the zillion things that I haven't answered that I now need to answer. So I'm going to read these out and if we need to clarify anything I'll ask you to unmute I think Sam's going to let you do that, and you can clarify if I'm getting butchering question. So the first questions are in the institutional population our college students considered institutional populations. And that is a great question and they are not in this case so they would typically be not in the labor force. But that's a really good question because again institutionalized differs from survey to survey so always read that section as to who counts. Yeah, for the census I would consider them that's that's that's interesting. This is a question that you said that the workforce is within America so what does that mean for people working abroad. Do they not count for example faculty teaching abroad. How they do not strictly people who are in the United States at this time so that's another difference from you know some of the other things. And this is my question actually, do you have some resources for recommending resources you can recommend for learning about the data linking part of that. I had a link, I had a link within the presentation but what I would actually also say is the IPOMCPS workshop has a whole. I don't know if it was an afternoon or morning but where they actually focused on that. Okay, more than just, you know saying all right here are the IDs that match, you have to do weights you have to do a lot of other things. Okay, great so that's that's awesome. Um, all right. What's in the annual March survey versus the more frequent supplements which seem to cover the same topics. So the, the basic monthly, it's strictly are you employed unemployed. The demographics are the same the big thing is that it's a year's picture. So you're getting income, but also some questions that are different so they ask questions about your taxation like do you have health insurance. Did you move. So it's a lot of additional questions it's just richer detail, but it's also I mean all of us for you know we're lucky we're librarians where we work in these stable jobs, but a lot of people change fairly often. And so you can get a bigger picture of how a person operated over a year, rather than just sleep. And this one I think you talked about this one earlier on, but just maybe reiterating it, do federal agencies suggest topics for the CPS like the FDIC data and how does the CPS decide which ones to do. So they sponsor, to be fair. So now like, so I guess the question would be maybe turning that around it's like let's say the FDIC had a question that was labor related not banking related that they really thought was critical. So no one can suggest questions and you know surveys do change over time and go toward it's wonderful about putting out these calls for, you know, there's a review of the ACS, or the whatever. So any place can, but it's not often that questions get added, but certainly if they want to sponsor a supplement, I'm sure it wouldn't just be we'll do it automatically. But if they had something they felt would, you know go along with they could, you know one of the other things is you don't want to have so many supplements that the person is sitting there for six hours, and then refuses to. I think co paths to will sometimes have presenters talk about the way that the questions have been added. Yeah, you are having a membership to co paths at your institution. That's another week. What if you know the demographic or data items, but that you want, but you have no idea which survey might contain them. This is a broader question. Data librarianship. It's a great one. How do you know where to look. So any tips you might have one. I assume when we're talking the United States so I will say I mean I know we all disparage at times, the American fact finder and it's successor, but that actually is a pretty good way to start because it you know you can go in you could say well here's the level of geography I need here is I need these basic things. So that's one way to start. The other I would say is going into a variable database. So for example, I see PSR so I see PSR may not have the latest necessarily feature the surveys but a lot of these don't change dramatically over time. So you can put in a variable, and it will tell you which surveys it shows up in the other is, you know, use your colleagues, you know, post a question to go to order ISIS or some other group that is relevant. And the other one I found that when I working with interns I would suggest was always the before it was well even now the statistical abstract abstracts. Absolutely. Yeah, that's a great one. Okay, so whether it's the statistical abstract or, you know, the digestive education statistics or these different ones, I would say like almost never find what I want there. That's not why I'm going to it. I'm going to see what the source is. Yeah, yeah, yeah, definitely looking because I will say if it's coming from the current populations are they or BLS or any of the different statistical offices. And the other one is somebody asked if we could use a video clip of your tutorial on how to use it comes in our life guys. We will be putting the recording up on YouTube so you're welcome to use that. Absolutely. And as I said, I will, you know, I'll be sending to Linda and Sam. I don't have a PowerPoint but as I said like, you'd have to. Well, you've all done this so you know like you do something with the best intentions, and then your script is three hours. So there's a lot of notes like particularly on the history section but there really are so many changes and again I don't expect anybody myself included to remember oh in 1967 they changed this, but it's really more just to be aware that if someone's going to really try to compare over them, there have been a lot of changes. Right. And so you have to be careful because otherwise you're not necessarily measuring the same thing. Well, thank you very I think those are all the questions we have this is really great. Thank you everyone for coming to the presentation. Thank you so much Bob Ray for representing. Thanks for, thanks for having me and things are one taking over this things to Linda I mean honestly just for doing this for such a long time. And for when it was with NCLA including after she had moved to New York State for a few years so this is incredible dedication. And thank you it's it's been a pleasure. I've learned a lot different from it. And I might be tapping you for some of those other surveys. Have a great day all. Yeah, everybody have a great day let me know if you have topics you're interested in or something that you would want us to try and find a presenter for. Thank you Bob Ray thank you to everyone and have a wonderful weekend to come.