Census 2000

Background

The Bureau of the Census carried out its decennial Census in 2000. They collected information about income, race, age, ethnicity, location and (infamously) number of toilets, among many other data. For this project you will look at some of this data.

Data

You will collect the data yourself from the Census website, at http://factfinder.census.gov. At the top is a box labeled "START with Basic Facts", which allows you to select tables showing the distribution of population nationally, in a state, or in a county, by age, sex, race, type of housing, or other variables. Some tables, such as income, are not yet available in detail; however, estimates based on monthly surveys throughout 2000 are available as part of the Census 2000 Supplementary Survey, which has a link just below the Basic Facts box. Browse through a few of the tables.

You can also find tables of information for states on a county-by-county basis. To find these tables, click on the Data Sets button on the left-hand side of the Census Factfinder mainpage. This will take you to http://factfinder.census.gov/servlet/DatasetMainPageServlet?_lang=en. In the Decennial Census column, select Census 2000 Summary File 1 (SF 1) 100-Percent Data, and then choose the option Geographic Comparison Tables. From here you can select whichever state you desire, and then the format State--County. Select Show Table, and you will then be given options as to what information you would like. Not all tables are available for all states - the information is still being compiled and released. So you should make sure you pick two states for which the information you want is available.

Problems

1. Choose a categorical variable (such as race or type of housing), and find the data on this variable for two different states. Display the data for each state using a bar graph (so you will create two bar graphs). Discuss the differences in the graphs. What does this tell you about the two areas you selected?

2. Choose a quantitative variable and look at how this variable is distributed among the counties of two different states (e.g. find the tables for the median ages of the counties in California and Florida). Display each set of data using a histogram (one for each state). Try several different bin sizes before you choose one. How is the histogram affected by the choice of bin size? Discuss the shape, center, and spread of the graphs: are they skewed? are there outliers? where is the center? how spread out is the data? Discuss the differences in the graphs. What does this tell you about the two areas you selected?

3. Take one of the data sets from each of parts 1 and 2, and compare it to the corresponding data (for the same area) from the 1990 Census (for which you will also produce a bar graph or histogram, etc.). How have things changed over the last 10 years?

In each problem, be sure to include the original tables from the Census website.

Graded out of 30 points