Python Tutorial: Census Subject Tables

Key Takeaways

welcome to analyzing u.s. census data in Python my name is Lee Hatcher dorium I am a geographer teaching graduate courses in geographic information systems and geospatial analysis in this course you will learn how to access and analyze United States census data most governments need to know the number and characteristics of their population in the u.s. a census is taken every 10 years by the Census Bureau the Bureau of the Department of Commerce often we just refer to the census when we mean the decennial census of population and housing but the Census Bureau produces many other data products in this course you will also learn about the annual American Community Survey but once you learn the census API you can use it to explore other products including the current population survey a monthly survey used to calculate official unemployment figures the economic survey a survey of businesses conducted every five years or the annual survey of state and local government finances which can be used to study taxes and service provision by sub-national governments this course assumes knowledge of core Python object types as well as programming concepts such as package imports control flow and list comprehensions it also assumes an understanding of pandas dataframes which we will work with a lot you will create some simple visualizations with Seabourn and geo pandas but no previous experience is assumed the decennial census counts as near as possible all persons and housing in the United States it covers demographic topics such as age and sex and housing topics such as home ownership and persons per room people living in group quarters are counted separately vacant housing units are also counted the American Community Survey is an annual survey of approximately 1.5 percent of housing units that covers a large number of economic and social topics which we will explore throughout this course the data is released in subject tables devoted to specific topics we will familiarize ourselves with the subject tables by working with table p5 Hispanic or Latino origin by race the column identifier x' begin p 0 0 5 4 the subject table followed by a column index of 1 to 17 column 1 is the total population it is broken down into two categories not Hispanic or Latino in column two Hispanic or Latino in column ten these columns are broken down further into seven racial groupings indented columns add up to they're out dented parent for example columns three to nine add up to column two we've created a pandas dataframe named States with data from table p5 you can examine the first few rows with States dot head each row is a state in the US the variable codes have been replaced with descriptive column names and the state name appears as the row index we will use Seaborn imported here with the alias SNS for data visualization SNS dot set is used to set the visualization style we will invoke SNS dot set with no parameters to set Seaborn's default style and use it exclusively throughout this course SNS bar plot naturally creates a bar plot x equals total puts total population on the x-axis how does the function know what total means total is understood as a column name in the source specified by the data parameter states y equals states index labels the vertical y-axis using the data frames index the axis labels are a bit crowded but we're not going to spend much time on plot customization in this course check out these other data camp courses if you want to go further enough preamble