Wednesday, September 29, 2010

Census and Statistics

The past two days have had headlines in the local papers discussing data from the Census Bureau. Therefore, I thought I would talk a little bit about the two topics that have been all over the headlines.


Yesterday brought us "Census finds record gap between rich and poor" from the Daily Camera and "Recession rips at US marriages, expands income gap" from the Denver Post. Both of these articles are based on data from "Income, Poverty and Health Insurance in the United States: 2009." This is an annual report released by the Census Bureau that you can actually look at back to 1985 online. The interesting fact about this article is that both articles discuss the growth in the income gap between rich and poor, which if you read further in the Census report you will discover is "not statistically significant" (Press Release), but since it was an increase that makes for much better headlines.

Want to see how the US compares to other countries in the income gap? The variable used to discuss the income gap is the Gini-coefficient of inequality. If you want to learn what about this variable and some other variables check out the Measuring Inequality section of the Poverty pages at the World Bank. For the data itself, go to World Development Indicators, a database from the World Bank.

Homeownership/Income/Education for Colorado

Today we got a series of articles from the Denver Post: "In tough economic times, Coloradans go back to school, census stats show," "Household income drops in Colorado over past 3 years, decade," and "Rate of Colorado homeownership drops 2% in 2 years." All three of these articles used 1-year estimates of the American Community Survey for 2009, which was released yesterday. This data is available from the "American Community Survey Data Sets" section of American Factfinder. Again, these articles leave out the margin of error rates for these estimates, which come about due to the fact that this is a survey, not the census form we all filled out earlier in the year. If you want data that goes in more depth, with lower margin or error rates, check out the three-year estimates (which go down to communities larger then 20,000), the one for 2007-2009 will be released in January. Even more exciting is the first 5-year estimates which will be released in December 2010, which will have data down to the census tract level. Learn all about this on the American Community Survey web site.

Not overwhelmed by the data offerings? Then why not check out one of the library's guides to statistics.

No comments:

Post a Comment