Homeless data per state – this data source offers insights into sheltered, unsheltered, and chronically homeless families, individuals, and unaccompanied youth.

My hypothesis is that if you have a lot of homeless people in your state, you also have bad survey scores, and that correlation is easy enough to see with a scatter plot.

Data.gov offers insights into other measure values that could hold statistical relevance, visit the website for more details.

Homeless data vs HCAHPS Hospital Survey Data per state, showing a negative trend and strong correlation.

Homeless data vs HCAHPS Hospital Survey Data per state, showing a negative trend and strong correlation.

Website – https://catalog.data.gov/dataset/ahar-part-1-pit-estimates-of-homelessness

The table below demonstrates Pearson Correlation value and a p-value, to determine its statistical relevance. The smaller the p-value, the more significant our data becomes.

We are using Pearson Correlation to compare the HCAHPS hospital survey data against different ways to categorize the count of homeless people per state. Sheltered Homeless Individuals and Sheltered Unaccompanied Youth are the most statistically relevant.

Measure Name Pearson
Correlation
p-value
S_HomelessIndividual -0.54017 0.000051453
S_UnaccompaniedYouth -0.52146 0.00010311
TotalHomeless2014 -0.51123 0.00014825
S_HomelessUnaccomp
YoungAdults
-0.50765 0.00016788
S_ChronicallyHomeless -0.50304 0.00019669
S_Homeless -0.49385 0.00026775
S_Chronically
HomelessIndividuals
-0.4899 0.0003049
S_HomelessVeterans -0.48562 0.00035051
HomelessIndividuals -0.47243 0.00053198