Add Roomer Calculated with airSlate SignNow

Add roomer calculated

recently I've been getting a lot of calls and emails asking me the question dr. berkey where are the census 2010 Gini coefficients and this is basically everyone I've had news media and and researchers all over the place asking me this question and let me tell you why people are asking this question for the census 2000 I and a student for a project we calculated Gini coefficients which are a measure of income inequality for every state and for every county in the United States and these Gini coefficients I'll provide a link to this on my homepage WWB Archie academy com this is a web page that I made about seven years ago and it's been used by a lot of different people since that time and the reason I created it is it used to be pretty hard to find Gini coefficients for counties and so this is a web page I have where I describe what Gini coefficients are held are calculated and provide some excel spreadsheets for these now as I said I've been getting a lot of email recently since the data for the census 2010 is coming out everybody wants updated Gini coefficients so this video is to answer the question where are the census 2010 Gini coefficients and I hate to tell you the answer but the answer is that they will never be coming now not for me but not from anyone else either and I want to explain why this is in the census 2010 the Census Bureau changed the way they do things let me show you how I want you to understand what's going on and why there will not be any Gini coefficients coming from the census 2010 let's go back to the census 2000 back in the census 2000 and also in 1990 and I believe in earlier census one out of every six households and it depends on the the state in the county but but somewhere between ten fifteen twenty percent depending on the county a little bit more than twenty percent in some counties got what is called the census long form and this is the census long form here let me shrink it down just a little bit the census long form asked how many people live in the house it also asked their names it asked for their race and their ethnicity and their marital status the long form asked information about education asked information about where the person used to live five years ago so this was useful for migration studies we know where you live now we know where you live five years ago so we can study things like migration now the important variables when you're talking about income or income and equality and things like that is you know these are the things you need to know right here question 31 on the long form it asks you about income from wages salary commissions self-employment just all kinds of interesting data on public assistance etc and what this allows you to do was to calculate average incomes or measures of income inequality all the way down to individual counties individual cities all the way down to individual census block groups or census tracts but now what the Census Bureau decided to do this beginning in about two thousand one or two thousand three they started planning that for the census 2010 they were not going to use the long form anymore and since this 2010 everyone got the short form and here on the Census Bureau's web page it says explore the form one of the shortest forms in history ten questions in 10 minutes and everyone in the 2010 census got this short form where basically they only asked you how many people live in your house and do you own your house what's your telephone number provide the name for everyone living there are they male or female when were they born are they Hispanic or not and what is their race and does this person sometimes live somewhere else that is all we know from the 2010 census so if you're going to do anything studying education income what kind of occupations people have any of these questions that we used to use the long form data for now what you have to use is something called the American Community Survey so the American Community Survey now is something that the Census Bureau does every two years and instead of trying to get one in six households answering a long form like we did for the census now every two years the American Community Survey gets anywhere between one and three percent of households depending on the state and depending on the county I read a report recently that said in Florida in a particular year they only got about 1.2 percent of Floridian households but then there was another state like North Dakota where they got somewhere about 3% so there's there's a lot of very variability and I'm guessing that the American Community Survey does try to get larger samples in larger percentage samples in lower populated states and lower populated counties so that is why there will be no Gini coefficients calculated by me using census data there's no income data now what you're thinking now is well can we use this American Community Survey data to calculate Jeannie's Gini coefficients the answer is sort of sort of you can although you have to be careful now why do you have to be careful well because now if you wanted to use American Community Survey data to calculate Gini coefficients for the United States or a state a whole state that is fine because you have enough households to do that but if you want to do it for counties in the united states that becomes problematic now why is that well take a lower populated county for example in North Carolina we have some counties with less than 10,000 people and let's say that those 10,000 people live in five thousand households because households are what is important if you're talking about Gini coefficients because people usually use household income now suppose the American Community Survey samples 1% of these five thousand households then how many households are we going to have where am I going to have 50 households surveyed and so if you want to calculate a measure of income and equality for a County based on only 50 households and it gets even worse for for many other counties there's one county in North Carolina for example called Terrell County that only has about 5,000 people or maybe 2,000 households you start to get into a lot of problems and I've looked at this personally where I've looked at the census 2000 Gini coefficients which should be very accurate and compared them to some of these estimates people get by county for the American Community Survey and for almost every low population County here I'm thinking about 30,000 people or less the error or the the difference in the ACS estimates and what the census gave are just enormous in many cases so instead of using one year since this ACS has done every two years what the ACS does and they calculate the Gini coefficients already instead of calculating the Gini coefficients for all counties every two years they only calculate them for high population counties every two years and then every four years they'll calculate them for smaller counties and every six years they'll group the data for three different surveys they will publish them for all counties at least in my experience in North Carolina you can get Gini coefficients for all counties but you have to look at the six-year grouped data let me show you how to get that since I'm not going to be providing these anymore you can go to this website called social explorer com and I'm sure there are other ways that you can get to this data and if you go to social explorer calm down on the over here on that you could click on this link called reports I just did this for the first time a few minutes ago that's why I'm a little shaky on this along the left-hand side they have different survey data that you can get to 2009 survey on religion etc so this isn't just census data it also has some other sources of data here you can see the group data this is grouping the 05 07 and 09 data this is just grouping the 07 09 data if you want Gini coefficients for all the counties click the 05 209 data here and let me walk you through how you could get this data for the counties click this geographic type drop-down box and click County and then you can select a state I'll select north carolina since that's where I'm from now you can click here all counties in the United States if you wanted to I'm just going to click all counties in North Carolina and click this little Add button and then click Next and here are a lot of variables that you can choose in order to get to the gini coefficient you have a couple of options here probably the quickest one is to click search by keyword and let's just type in GI ni search and it pops up it's variable number be 1908 three gini coefficient add it and then click show result and this lists the first three counties alphabetically in North Carolina probably the easiest thing to do for most people will be to click this Excel tab and you can click which version of Excel you would like to download these Gini coefficients under and we can open this up so let me bring it over here so it'll bring it up in a nice window and you can look at the Gini coefficients well it's also nice is it provides you with a standard error estimate here for each county for example here Allegheny County in North Carolina has a standard error of plus or minus point 0 3 now if you wanted a ninety-five percent confidence interval estimate you would probably want to approximately double that plus or minus point 06 so even with six years of data you can see that there's still quite a bit of error point 48 plus or minus point 06 Anson County plus or minus point oh eight and for each county here it lists different standard errors that you can make margins of error for and so if you're interested in getting Gini coefficients by County from now on you're going to be getting them from the American Community Survey and it's very important because the samples are much smaller than the sample sizes that were used in the long form for the old synthesis it's very important that you pay attention to these standard errors so if you have any questions about Gini coefficients or how these data are calculated you can feel free to contact me on WWE academy com