Insights from Pit Data: Mobilizing Results and Understanding Homelessness Trends

Mobilizing Results: Working with PiT Data Aaron Segaert November 2015 Module 7 Mobilizing Results

Now that we have the data, what do we do? Clean and prepare the data Get set up for analysis There is always a possibility of data entry errors or other mistakes. As you work with the data, keep an eye out for contradictory or nonsensical results If something looks very wrong, delete the incorrect bit of information (i.e. treat it as missing information). You can still count the person and use the rest of their information

Results: Simple and to the point PiT count results are intuitive, simple and easy for people to understand: On any given night, about 200 people are homeless No complicated statistics or methods are required, mostly counts and percentages: 5.7% have Aboriginal ancestry 27% are female 14 of the 26 people surveyed on the street have not used a shelter in the past year

Small numbers Aside from Canada s largest cities, the total number of people counted will be in the hundreds or even dozens, so some categories will have very small numbers: Key Demographics N % Male Female Veteran Aboriginal Recent Immigrant/refugee Child Youth Adult Senior 139 52 5 11 5 5 37 143 7 72.4% 27.1% 2.6% 3.1% 2.6% 2.6% 19.3% 74.5% 3.6%

Use charts and graphs 25% 22.9% 19.8% 19.8% 18.8% 20% 15% 13.0% 10% 3.6% 5% 2.1% 0% 0-14 15-24 25-34 35-44 45-54 55-64 65+ Age Groups

Pregnancy 1.0% Brain injury 1.6% Learning disability 2.6% Fetal alcohol spectrum 4.2% Physical disability 8.3% Chronic/acute medical 12.5% Mental illness 20.3% Addiction or substance abuse 28.1% % of respondents

Show detailed info in tables Times homeless in past three years One 135 70.3% Two 28 14.6% Three or more 29 15.1% Fifteen percent are episodically homeless

Play to the strengths of PiT An opportunity to put shelter statistics in context: How many people are not being counted in shelter statistics? There is scant real-world data on this question 20% of respondents were sleeping in locations other than a shelter, and 53% reported not using a shelter in the past year Overall, 92.7% of those enumerated were either in a shelter on the night of the count or used a shelter at some time during the past year

Play to the strengths of PiT What are the characteristics of people who do not use shelters? Are some groups less likely to use shelters? Used a shelter in past year? Yes 91.3% 96.2% 60.0% 81.8% 91.9% 92.3% 100.0% 92.7% Demographics No Male Female Veteran Aboriginal Youth Adult Senior Total Sample 8.7% 3.8% 40.0% 18.2% 8.1% 7.7% 0.0% 7.3%

Compare and contrast Use existing community data from Statistics Canada and other sources to compare and contrast with the general population: Seniors are under-represented relative to the general population, with only 3.6% of those surveyed aged 65 or over 2.6% reported having military service, about the same as the percentage of Veterans in the wider community

Compare and contrast Look at differences between groups, e.g., those surveyed on the street vs. in shelter: Location % Male Street 83.8% Shelter 73.7% The proportion of males is higher on the street than in shelters

Dont expect big surprises It s okay if there are no earth-shattering results! The results should be interpreted with caution if they show something unexpected. After all, this is a single point in time If your community is thought to have a high rate of Aboriginal homelessness, or lots of people migrating to the community and becoming homeless, etc., the results will probably show this But you never know maybe the results will force people to reconsider their preconceptions about the homeless population