Thank you for having me here today.
My name is Aaron Segaert. I led research for the homeless individuals and families information system from 2009 to 2017. HIFIS is the common term for this. It's a computer system that was built by the federal government and installed at homeless shelters across the country. It gives information about shelter stays. We do research with that data.
Although not everyone experiencing homelessness uses a shelter, HIFIS contains a very large slice of the homeless population and allows us to understand a lot about the different types of people who use shelters across Canada.
HIFIS started collecting information about homeless veterans in 2013, and it took about two to three years for the data to really accumulate.
The first research we did was in 2014. Actually, it took place in 2015, using 2014 data. That report was called the “The Extent and Nature of Veteran Homelessness”. I didn't name it that. It was named by comms instead of by me. They put the “The Extent and Nature of” on there. HIFIS can't tell us a whole lot about the “nature” of veteran homelessness. It can tell us a lot about the numbers, though.
This was the first report that had ever given us an estimate of how many homeless veterans there might be in Canada. No one really had any idea before that. What we found, using a sample of, I think, 60 shelters, was that there were about.... We estimated that there were about 2,250 veterans using shelters annually. That's about 2.7% of shelter users.
The important thing to keep in mind with shelter statistics is that we look at them over the course of a year. That doesn't mean that there are 2,200 homeless veterans right now. It means that over the course of a year, that's how many use shelters. The other thing is that any veterans who didn't use shelters wouldn't be included in that number, so there could be more.
The other interesting finding in that initial report was that veterans were more likely to be episodically homeless than other shelter users were. What this means is that they're in and out of homelessness. They keep returning, over and over again. Technically the definition is.... A homeless episode is a time using a shelter separated by at least 30 days before returning to the shelter again. If you have at least three of those in one year, you're considered episodically homeless. We found that veterans were about twice as likely as other shelter users to be episodically homeless.
The other thing we found interesting was that there were quite a few female veterans using shelters. In the general population of homeless shelter users, it's about a 70:30 male to female split. We found the same thing among veterans, but when we looked at veterans under 25, it was actually about half and half. About half of the veterans under 25 using shelters were female.
I think a lot of the idea we have about homeless veterans is that they're older men, but at this point, most of the veterans of the big wars are very old and are not using shelters anymore. These are people who have been in the military in some capacity. It's also not necessary that they had been overseas or in combat. That's why we have some people who are saying they had served in the military, and who are quite young and finding themselves homeless.
Two years later we released another study called “The National Shelter Study”, which isn't specifically about veterans, but it did mention veterans in there. We have an updated number.
For that report, we used a much larger sample and a more sophisticated methodology. We found that out of the approximately 137,000 to 156,000 Canadians who use homeless shelters each year, about 2.2% were veterans in 2014, which is the first year that we were able to produce that number. This comes out to about 2,950 veterans, in that estimate. It's a little bit higher than the other one, but these are estimates. There's some error around it. It's just a guideline. It's not an exact number.
Again we found the same type of pattern, where about 70% were male and 30% were female. We also again found that over half of the veterans under 30 were females. As far as the males go, they tended to be slightly older on average than other male shelter users.
In that study, we found no significant difference in the length of stay between shelter users with and without military service. That particular study is not really designed to look at chronic and episodic homelessness, so we didn't have any new figures for that.
I would say that the figure of 2,950 is more accurate than the initial study with the estimate of 2,250 veterans. It uses far more data, a better sampling method, and all that.
I think ESDC will be releasing 2015 and 2016 figures in the next few months. I know that at the time I left ESDC and the homelessness partnering strategy, they were working on updates to that study.
The third method that gives us a glimpse about homeless veterans is the point-in-time count. There was a Canada-wide point-in-time count in 2016. The counts in cities are where they go out on one day and just look for people who are in the shelters and on the street. They typically find a few more veterans than we do in the shelter studies. We've heard anecdotally that a lot of veterans don't like to use shelters. Typically, in the point-in-time counts, around 5% to 6% of the people counted have served in the military. This varies by community as well. I think the results showed that somewhere between 0% and 13%, depending on the community, were veterans.
Some of the reasons for the differences could be that veterans are just less likely to use shelters, so more of them are found when you consider sheltered and unsheltered homeless people. It could be due to missing data in our shelter studies. As I mentioned, within HIFIS, the software that the government provides to shelters to count homelessness, we only started rolling out the veteran question in 2013. It takes quite a long time for that to be populated, so there are still some cases where that is missing and we don't know whether someone is a veteran or not. It could also be under-reported, because it's based on self-report and some people might not disclose that they are veterans, or they might not be asked. For whatever reason, that data might be missing.
These are estimates over a one-year period that I have been talking about, the 2,250 and the 2,900 veterans. That shows that there definitely are homeless veterans, and we can probably safely say there are more than 2,000 veterans experiencing homelessness in Canada each year. It's about 2% to 3% of the shelter population. I believe veterans are around 2.4% of the Canadian population, so that means they're not overrepresented in homeless shelters.
Male veterans tend to be older. Female veterans tend to be younger. This is probably an area for further research, probably not using the HIFIS dataset but in terms of actually going out and trying to find young veterans and talking to them about their experiences of homelessness.
The other interesting thing when we compare the shelter studies that I worked on with the point-in-time counts is that we should try to find out more about the veterans who are experiencing homelessness, whether they are less likely to use shelters than others and find out what the reasons for that might be.
That's about the gist of what I can tell you. If you have more questions about statistics, I'm your man for that. If you want to know more about people who actually interact with veterans, I'm not really able to say much about that.
Thank you.