- Introduction
- About the Data Set
- 1. Diffusion of Ethinicities in the US
- 2. Unemployment Insights for the US
- 3. How is Unemployment Distributed across the counties of the Top 3 States with the Highest Unemployment Rates?
In this project I am exploring a US Census Demographic data set, and visualizing some findings using Tableau.
The data set comes from Kaggle, and includes the census data for all US counties in 2015. The data originates from the DP03 and DP05 tables of the 2015 American Community Survey 5-year estimates, while the Census Bureau updates the estimates approximately every year.
For this project, I am using the data file acs2015countydata.csv, which provides data for each US county, including DC and Puerto Rico. For the full data sets and more information, please refer to the United States Cencus Bureau website.
You may find the Worksheet here
In this visualization, we can see all the ethnicities for which there is data for in the provided data set:
- Asian
- Black
- Hispanic
- Native
- White
appearing in alphabetical order.
We can see is that the more evenly distributed ethnicities are the White & Asian populations. To further explain, by even distribution, I mean that these two ethinicities seem to have less peaks and appear more evenly in most states. In contrast, the black community, for example, seems to exist mostly in 9 states (9 peaks), while the Latino community seems to be living in mostly 2 states, Puerto Rico & Texas, which correspond to the two prominent peaks. Lastly, the Native community is somewhere in the middle, showing 6-7 peaks, but with the lowest population numbers.
I chose the bar chart as the data is categorical and this is the most appropriate chart type to use in this case. It is a simple but effective choice that reveals something that might not be so intuitive to understand otherwise.
I added two filters, to facilitate the filtering based on state & county, in case the reader wants to get the appropriate info in a more in depth level. Upon selecting a state, we can also see the actual data for all ethinicities for the given state.
N/A
You may find the Dashboard here
In this visualization, we can see some insights about the unemployment in the US. I am focusing on the top 10 states with the highest unemployment values. In a descending order, these states are:
- Texas
- Georgia
- Puerto Rico
- Kentucky
- North Carolina
- Mississipi
- Virginia
- Tennessee
- Missouri
- Michigan
The first obvious finding is that all these 10 states are in the eastern part of the US, with most of them being in the south as only Michigan lies on the north.
Not so surprisingly, we can see that Unemployment and Poverty usually go hand by hand. This is represented by the visualization on the top right. In the tooltip, I chose to also include the data for Child Poverty as an extra interesting piece of information. As we can see by the blue color differences, Child Poverty follows more or less the same pattern with Poverty, with the exception of Missouri that has higher than expected values for both Poverty and Child Poverty.
The last part of the Dashboard, on the bottom right, has a two-fold role: the first one is to give us some more extra arithmetic data about the 10 states that we examine, as an attempt to expand our understanding of the relationship between Unemployment, Income per Capita, and Population. At the same time, the visualization gives a quick sense of the size of Poverty as the bar appears as thick or thin.
I made use of the map that can be embedded in a Tableau worksheet in order to see how the top 10 states with the highest unemployment are spread geographically.
In the map, the Highlight State bar is used for an easy navigation: upon choosing a state out of the 10 that are available, the data is automatically highlighted in the visualization parts on the right. We may also click on a state in a table on the right, and get directly the data on the map and on the other table/graph, as both tables act as a filter.
For comparing Unemployment and Poverty, I actually took my inspiration from the popular population pyramid charts. I have also added the data for Child Poverty, using hues of blue as a differentiator in order to give a prompt sense of the metric: the darkest the blue color is, the highest the Child Poverty is.
In the last part of the Dashboard, I am giving mainly some arithmetic data along with the dimension of Poverty represented by the thickness of the bars: the more thick the line is the higest the poverty is, and the opposite.
Finally, I chose the combination of blue & orange for the visualizations as it is the suggested combination for people with color blindness issues.
- How to Create a Population Pyramid Chart in Tableau
- 5 tips on designing colorblind-friendly visualizations
3. How is Unemployment Distributed across the counties of the Top 3 States with the Highest Unemployment Rates?
You may find the Story here
I created this visualization because I wanted to see how evenly (or not) the unemployment is spread along the counties. I chose to examine this only on the top 3 US states with the highest unemployment rates. These were:
- Texas
- Georgia
- Puerto Rico
The charts revealed that unemployment is spread in the counties of Texas & Georgia following the same pattern, more or less: the highest unemployment rate is 1.2% for Texas & 1.3% for Georgia. At the same time, the highest value for Puerto Rico climbs is 2.4%, i.e. double than that of Texas. Furthermore, there many more large bubbles in Puerto Rico graph comparing to that of both Texas & Georgia. As I understand it, this means that unemployment is much more widespread in Puerto Rico than in Texas or Georgia.
Please note that I chose on purpose to keep all the counties in the bubble charts as I want to illustrate that in Texas and Georgia there are many bubbles of a smaller size and only a few bigger ones, while in Puerto Rico there are many bubbles of larger size. If I was to limit the bubbles to 10 or 20, this finding would no longer be obvious.
However, in order to facilitate the reader in further exploring the data on the bubble charts for the three states, I added a highlighter in all three of them. This way, it is possible to select any county and promptly get the available data for it.
All visualizations were created as bubble charts in order to facilitate the visual comparison.
The size of the bubble is directly related to the unemployment rate, where largest size represents higher unemployment.
As I did in the previous visualizations, here too I chose the colorblind-friendly palette that is available in Tableau.