# Notebook

## Pick a Domain of Interest

• 2/17: The domain of interest that I chose has to do with human mortality. I chose this particular domain because of my interest in aging population of the world. Due to a variety of causes, the population of the world is growing, but it is also getting older. Many nations will soon be under enormous economic pressure due to their large elderly population. Consequently, I would like to see how trends in human mortality has changed over the years and how it might provide a projection for where the populations might end up.

## Pose an Initial Question to Answer

• 2/17: Based on the graphics provided, what will the growth of the human population result in?
• 2/18: After looking over the data and what it is actually providing, I realized my question is slightly off. The HMD does not provide population statistics or growth, but rather simply death statistics (the chance that a person survives their current year of life). I am therefore altering my question to be: How likely is it that a female and male reach certain life milestones of age? These milestones include the ages of 5 (baby and toddler survived), age of 12 (pre-teen), age of 16 (mid-teen), age of 18 (late teen), age of 21 (US drinking age), age of 25 (mid-20's), age of 30, age of 40, age of 50, age of 60, age of 65, age of 70, age of 80, age of 90, age of 100. These ages may be altered as I begin to synthesize the data. These ages will also be focused over a period of time (yet to be determined).
• 2/21: After learning more of how to use Tableau and trying to format my data into a useable format, I changed my question. I decided that trying to figure out what the likelihood that a person reaches a certain age milestone wasn't providing a particularly interesting story. Additionally, it the data was becoming too jumbled and great to properly and easily handle. I decided therefore to change my question to the following: What is the difference in average life expectancy at ages 65 for the years 1933-34 and 2005-06, and how does that difference effect what Social Security will be supporting in pure numbers of people?

## Find a Database that has the Data needed

• 2/17: I chose to use the Human Mortality Database to get my data from.
• 2/18: Based on data, I may choose to pull in other sources to juxtapose mortality against.
• 2/21: Stuck with only the HMD database.

## General Progress

• 2/18: Narrowed down data. Converted it to Excel so that it could be loaded into tableau (didn't like the text files).
• 2/21: Formatted data. Learned basic operation of Tableau. Created visualization. Finished write up.

# Visualization

## Visualization Description and Write-Up

Visual depicts data about the United States. Data mined from the Human Mortality Database (mortality.org).

The purpose of this visualization was to answer a question about the amount of people that were surviving at the age of 65 and how much longer they would survive. The reason I thought this question was interesting is that it helps to quantify how many people social security was originally built to handle, and what it is now handling. I chose the years in question because social security was founded in 1935 and then the most current available years. I originally thought that I would only need to plot life expectancy, but found that number was amplified when I included a graph of surviving people (out of 100000). As can be seen in the graph, for females there is about a 7 year life expectancy increase and for males there is a 5 year increase; for females approximately 87000 are surviving to the age of 65 which is up about 27000, for males the figures are similar in that 79000 are surviving to 65 which is up from 52000. So not only are significantly more people surviving to the age of 65, but they are also expected to live until they are almost 80. Thats about 20 years of support from social security.

While my visualization helps to provide a good visual of the rate at which people are being added to the pool of potential social security people, it doesn't provide a percentage of the entire population versus those taking social security. I think that now that I have made this visualization, a visualization that I would like to see made is one that combines this one (the rate) with the numbers of people over 65 as a percentage of the entire United States population. I think that perhaps this might provide a better story by allowing for understanding about how many people are supporting a section of the population that no longer works.

The reason I did not implement that visual as another step in the iterative process, it that I spent too much time trying to figure out tableau and organizing my data into such a manner that Tableau would be happy. I think that if anything, this exercise really made me appreciate how difficult it is to get your data into a workable format that can be worked with.