Census 2000

In this project I took county-level race data from the 2000 Census and used it to create maps showing the population density of various races across the continental United States. I made three individual maps displaying Black, Asian, and Some Other Race categories. To begin this project I started with the map I created in the ArcGIS Census2000 tutorial which included an excel table with basic census data such as county area, population per square mile, etc. To this table I joined the race data that I collected from the US Census Bureau website. To do this, however, required me to clean up the census data I collected and simplify the headings in of the excel spreadsheet. By using the same State and County FIPS (codes for identify the individual counties) in each of the census data tables, I was able to use this specific field to create the join. Thus in doing so I was able to add this new race census data and create my three maps of Black, Asian, and Some Other Race population densities.
When choosing a map projection I decided to use the Lambert conic projection for all three maps because I feel that this projection shows a relatively accurate model of the continental United States. Next when it came time to classify the race data, I chose to use the natural breaks to model the data. I feel that using natural breaks instead of a preselected range of percentages yields a more accurate representation of the data that I am displaying. I also chose to use only four categories and grayscale for simplicity.
When looking at the map of black race data it shows that in the year 2000 a higher percentage of blacks lived in counties found in the Southeastern states such as Louisiana, Mississippi, Alabama, Georgia, South Carolina, North Carolina and Virginia. Also near Chicago, Illinois there is a relatively large black population. The map of Asian race census data shows that the highest population densities of Asians are found on the West Coast, especially around the San Francisco Bay Area. One trend that I can see from looking at both the Black and Asian race census data is that the highest populations of each race in America are found nearer to their original homelands. More Asians live on the West Coast which is closer to Asia and more Blacks live in the Southeastern United States which is closer to Africa. It is interesting that in the year 2000 immigration to America--either recent or centuries ago, willingly or forced--people for the most part people have remained closest to their native homeland.
The map for Some Other Race shows highest population densities from Texas westward through the Sunbelt up to Washington. According to the US Census Bureau website, “some other race” includes all other responses not included in “White,” “Black,” “American Indian and Alaska Native,” “Asian,” and “Native Hawaiian and Other Pacific Islander” race categories. Thus this race category is very broad and can include people who are multiracial, interracial, or consider themselves to be from a particular Hispanic/Latino group--Mexican, Puerto Rican, or Cuban--or those who are Jewish. Therefore since this racial category is not very specific, it is difficult to determine and patterns or trends from the map. The most one could infer from this map is that the majority of people who consider themselves in this category are found in the Center-West part of the continental United States and that most people on the East Coast do not consider themselves in this category.

Projections 101

Map Projections 101 Report

This lab demonstrates the significance of map projections and how different projections can show you very different representations of the world. Since we are trying to make an accurate two dimensional model of the Earth, which is three dimensional, selecting the correct type of map projection requires consideration of what is the purpose of the map. Since every projection causes some distortion, some projections more so than others, making the correct choice of map projection should be determined by what type of information is hoped to be derived from the map.

In our case for this lab, we wanted to figure out the distance between Washington D.C. and Baghdad, Iraq. When each of the six different map projections were applied to our original map, it yielded very different results when determining the distance between these two cities. The conformal projections had drastically different measurements, with the Mercator projection stating 8,414.87 miles from Washington D.C. to Baghdad and the Gall Stereographic projection stating 5,951.85. The equal area Bonne and Sinusoidal projections showed 6,103.85 miles and 6,732.92 miles between the cities while the equidistant projections Plate Carree and Equidistant Conic measured the distance to be 8,410.27 miles and 6,266.63 miles. I went online and looked up the distance between Washington D.C. to Baghdad on www.timeanddate.com and it said that the shortest distance between these cities is 6213 miles. If this data is correct, it suggests that the Equidistant Conic and Bonne map projections are the most accurate.

When looking at these six different projections we can see the various degrees of distortion. The conformal projections seem to maintain more accuracy in displaying the land found closer to the equator, however towards the poles there is significant distortion--Greenland is much too large and Antarctica is humungous. Between the two conformal projections, the Gall Stereographic projection is a better map projection than the Mercator, especially for our purpose in determining the distance between Washington D.C. and Baghdad. Of the equal area projections, the Bonne projection significantly distorted some landmasses such as Australia which is shown to be much larger than it is. Also land towards the North Pole, specifically Greenland, Russia, and Canada, are proportionally much smaller to the than they are in actuality. Nonetheless, land nearer the equator seems to be more accurately preserved and thus the Bonne projection provided us with the most accurate distance between our selected cities. Finally the equidistant projections provided us some of the best and worst measured distances between Washington D.C. and Baghdad. The Plate Carree projection looks strangely warped with the southern hemisphere larger than normal while the northern hemisphere is shrunk, which probably contributes to the great exaggeration of the distance between our selected cities. The Equidistant Conic projection, like the Mercator, shows Antarctica to be too great in size, yet nonetheless provides us with a relatively accurate distance between Washington D.C. and Baghdad.

Ultimately choosing the correct map projection to use requires one to determine which projection will display the best results in answering the given question. Since map projections can distort shape, area, distance, and direction, it is important to identify the projection that will provide the least amount of distortion to properly display a map suitable to answer the given question. Thus in our case, since our question was in regards to distance, choosing a map projection that distorted distance the least was most important in answering our question. Though having a map that does not distort everything else too much is still important, preserving distance was first to shape, area, and direction.