Disclosure:

• The data used in this analysis is public information and therefore has not been altered.
• This data is provided by each state to the government…
• As most charts are done using Tableau, they are interactive and best viewed full screen.
• You can hover over data points, columns, use the filters, whatever you wish, you’re not going to break them.

HOMELESSNESS

This project was a lengthy detailed analysis of a large amount of data. The main purpose of the analysis was to have a better understanding of the areas we needed to focus on prior to developing a product to aid the homeless population. We didn’t aim to solve homelessness but to try to understand where we should focus our mission.

I was the lead on this charitable project, and wanted to make sure our resources were used efficiently.  The ultimate goal was to develop a product that provided the most help for the funds available.

• I didn’t have a specific mission or pre-determined guidelines, so I came into it with an open mind and no bias or direction.
• As you’ll see in the analysis, I looked at almost every combination trying to find a correlation.
• I omitted many charts for the sake of keeping this page reasonably short.
• And BOY did I learn new things I never thought would be true.
• I have to admit that I had a feeling that CA would have a large number of homeless people for several reasons:
  • I lived in CA and I interviewed thousands of homeless persons.
  • The weather is great so if I wanted to be homeless, I’d make my way there so I don’t have to deal with the inclement weather.
  • I thought all great climate states would have a large number of homeless persons for that same reason.
  • CA is also the most populous state, so for that reason just being in the middle pack (% wise) would automatically cause you to have a large number of homeless persons.
  • CA economy is always scraping to survive and seems to be broke all the time.

Things to know:

• I will abbreviate Homelessness with H.
• I will use standard deviation on a regular basis as you’ll see.
• I will show just a few charts of the many I analyzed for this project.

% Homeless

First off, let’s see which state has the highest % of its population counted as homeless by the government. This chart guided me through the entire analysis, here is why:

• You’ll see in the chart below, % numbers begin to level off after the first 9 states as you count from left to right.
• My cutoff was 0.2% because as I stated earlier I wanted to focus on the most cost effective product I can produce.
• To help the most homeless persons with the funds I had with the one product I wanted to develop.
• Try to ignore the name of the states for now, we’ll focus on the names after you’ve seen a few more charts.
• Try to ignore the sub-chart for now we’ll get back to it later.
• Best viewed in full screen mode.
• My first surprise: Hover over the first 4 states with the highest %. Are you surprised at which states have the highest % of their population considered H.

Again, I will watch the NINE STATES colored in red more closely for now. I will also color these NINE states red throughout the analysis to remind me if anything stands out.

Concentration of H

So now that we know which states have the highest % of their population considered H. Let’s see how much these states account for the entire H population in all of the states.

This is a very important chart for our purpose. I will first show a table showing the same individual values and follow it up with a chart showing the cumulative values as we go from largest to lowest.

• The previous chart showed us how some states have an extremely high % of their population as H. What I need to know is how to help the largest number of H persons.
• The nine states from above are colored in red for reference.
• Here you see by the size of the boxes and count in each box which states have the most H persons.
• In order: CA, NY, FL, TX, PA, OR, GA, IL, CO

• So look at the chart below, notice the 9 states colored in red are not the top 9 from the % chart in the previous section, because obviously MA and DC are one of the smallest states.
• CA is leading the way which what I thought would happen coming into the project because of it being the most populous.
• Once again I stopped at 9 states because as you see the difference between state 8 (OR) and state 9 (GA) is 2.35% and decreasing as we move further out into the other states.
• Each bar chart has a number over it, that represents the cumulative % of the entire US Homeless population as we move from left to right.
• So CA alone makes up 21.54% of the entire US Homeless population which I knew due to the great weather, plus being more densily populated.
• NY is second? That was a surprise to me with 15.74% of the US Homeless population. FL is not surprising once again due to the weather (I assume it is the weather) but let’s not have a bias right?
• Out of the 9 red states only 5 appear here.
• The 9 states in this chart make up 62.55% of the entire US Homeless population.

I will draw these 9 states as SQUARES for the remainder of the analysis so we can track them.

If a state is in the top 9 in % it will be in RED CIRCLES, and if in top 9 in Quantity/Number of H it will be in SQUARES. If a state is in both it will be a RED SQUARE.

What does that mean: well even though those other 4 states that didn’t appear here need the most work in regard to the homeless population, their count is not significant for me to focus on. But the fact that two states in particular have percentages that are more than DOUBLE all other states regarding their homeless population, is absurd. Those states need to focus on solving their issues because they deviate from the norm. We’ll get back to those states later.

H vs Employment

Of course, the first thing one would go after is employment, assuming that if the unemployment rate is high for a state one would assume that should cause H rates to be high!?

• On the vertical axis is the % of the state population that’s homeless, horizontal is the Employment rate.
• Note the nine states in red. I don’t see anything unusual as they fit right into the 1 Standard Deviation on either side of the mean.
• Notice the ones that fall outside the 1 SD (Standard Deviation) are well below the mean of the % Homelessness.
• One outlier you’ll find way down to the lower left is the state of PR?
• Puerto Rico, the state with the lowest employment rate has one of the lowest % of homeless persons! I’m not about to say what’s on my mind but let’s remember that for now and look for PR in the next few charts.
• So do you see any correlation between Employment Rates VS Homelessness?

H vs Incarcerated

Next, I’ll focus on this relationship see if anything stands out.

• HOLY COW. Talking about standing out!
• It seems that those two states like to stand out in a crowd.
• Once again just hover over the data points and you’ll have more information.
• Nothing really stands out with the other 7 red states.

H vs SNAP

SNAP is the food assistance program provided to population that falls below the income level requirements set by each state. Could there be more HOLY COW moments?

• Hover over the lone state above the 1 SD, all the way to the right?
• Well what do you know: PR
• So now we have two outliers being PR
• First one being PR having the lowest employment rate in the nation, and now PR having the highest % of people on the SNAP food assistance program, and yet it has one of the lowest Homeless %! Hummm

Unsheltered vs Sheltered

Now that we have a general idea where to find the highest number of H persons, and we also know which states have the highest % of their population being H, let’s see which sates actually take care of their H population by providing shelter for them.

• Once again the 9 states in red, starting from left the highest one is: MA (hover over the column)
• Here we see in the lower section that MA has one of the lowest % of unsheltered H persons. So that might lead us to believe that the state is aware and is providing shelter for them.
• Remember, shelter does NOT mean housing, it means just that “shelter”
• DC is second and has an even lower % of unsheltered H!
• I’m beginning to suspect something here, are you feeling what I’m feeling? Maybe we shouldn’t have any feelings since we are analyzing data right? We shouldn’t have any biases?
• It is odd that the 2 states with the highest percentages of H persons, and the highest percentages of incarcerated persons have also the lowest percentages of unsheltered H persons!
• If my goal was something else I would pursue this information further but since my project is aimed at helping the largest NUMBER of H persons let’s move on.
• So the 4 highest % of unsheltered H persons are in CA, OR, HI, NV and if you remember from a previous chart in the Concentration of H section, CA, OR are the only two states in the top 9
• Looking at the numbers from that chart we see that the H numbers in CA= 118,142 and for OR=13,238

So far, it’s obvious that if I want to help the largest number of H persons, the focus should be CA.

Section 8 vs H

If you are not familiar with Section 8, it’s a government sponsored housing program to provide housing, both public and private, to families/individuals that fall below a certain income level.

I wanted to see if the availability of that type of housing could have a relation to the number of H person in each state.

SNAP and section 8 have similar guidelines if not identical, so I will combine the two variables together.  If you just focus on each section of the chart on its own, it will not be as confusing.

Let me explain the chart:

• Bottom/Horizontal axis is how may Section 8 unit provided for each 1 SNAP recipient.
• The upper left vertical axis is the State % H population.
• The lower left vertical axis is the Unsheltered % of that state H population.
• You can click on any data point, and it will isolate it in both charts, that will make it easier to read.
• Let’s click on the upper left most data point: DC a red dot.
• It will match it with a point for DC in the lower chart.
• Hover over it and it will display that DC has 1.23% of its population as H.
• It also tells you that the state has 1 Section 8 Housing Unit for every 2.85 SNAP recipient.
• That is the lowest and best rate of all states.
• What’s glaring at me is the fact that NONE of the red colored data points are even close to being outside the outer limit of the 1 SD. FL is one of the states in the top 9 regarding the NUMBER of H population, and you see it in BLUE Square.
• That was shocking to me. The fact that the states are trying to provide housing for their H population is encouraging.
• Now let’s see which states have the least number of housing units available for their SNAP recipients?
• Using the lower chart, go all the way to the rightmost data point and click on it.
• AZ, has the lowest number of government housing 1 for every 14.34 SNAP recipient, and yet has one of the lowest % of H population, so I presume they don’t think it’s important to them. Let’s see what other states have the lowest number of housing units available for its H population.
• Next ones from the far lower right are: PR, TN, MI…..
• PR again. It is even higher than CA in the % of unsheltered H population.

H vs Income & Poverty

Right, if you don’t make money that should be the obvious reason you are homeless.

Poverty level at that time was = 15.5

Median Income level = 53,889

• Since the number of H is not relevant here, but the % is what’s more related to the Median Income and Poverty level of that state I chose to use the % H population for the primary chart (lower chart)
• I’ve combined the two 9 states in one. Look at the bottom chart, the bar chart is the % of H from each state’s total population, so you’ll see the left most bars are the 9 red states from earlier section.
• I’ve also colored in Blue, the top 9 states with the greatest NUMBER of H persons per state, as you remember from an earlier section.
• Hover or click on the first bar to the left- MA: It has a poverty level of 11.6% well below the poverty line, Median Income of 68,563 which is way above the mean, and yet has the HIGHEST % of H population? What in the world is going on here?
• Next one DC, has an even higher income level but a higher poverty level
• CA above median income, and close to poverty line.
• Would you like to guess which state that tall green bar belongs to? Of course, PR.

One in Every is H

Here is a summary that we actually ended up using for marketing and to raise awareness. You’ve seen the chart before, and now I’ve presented in a different setting.

• Feel free to use the filter in the upper right corner to narrow the chart down to a specific number of states, as you slide the handles in or out the chart will filter itself
• You can see from the chart several tiers for embarrassment.
• the first tier being the first two states: MA at 1 in every 68 of its residents is considered H, DC is next with 1 in 82. Those two states set themselves apart from all other states with more than 3 to 1 percentage.
• The next nearest being at 1 in 180 in HI, heck I wouldn’t mind being homeless in HI and that tier continues till around 397 at NV
• Then it jumps to 1 in 525 in CO and so on….