Everyone Counts – A guide to Point-in-Time Counts in Canada – 2nd Edition

From Employment and Social Development Canada

Request other formats online or call 1 800 O-Canada (1-800-622-6232). If you use a teletypewriter (TTY), call 1-800-926-9105. Large print, braille, audio cassette, audio CD, e-text diskette, e-text CD and DAISY are available on demand.

About this guide

This guide was developed by Employment and Social Development Canada. It is intended to provide guidelines to communities that wish to align with Homelessness Partnering Strategy Coordinated Point-in-Time (PiT) Count. It includes the standards for participation, the Core Screening and Survey questions that are used across participating communities, as well as guidelines for implementing a PiT count.

The guidelines in this document were developed through a review of PiT count practices used across Canada and internationally, and with input from communities, including the HPS PiT Count Working Group. It builds on the existing work on PiT counts by the Canadian Observatory on Homelessness (COH) as well as the Department of Housing and Urban Development (HUD) in the United States.

In this 2nd edition of the Guide, the standards, Core Screening and Survey Questions and guidelines have all been updated based on feedback from communities that participated in the 2016 Coordinated PiT Count, as well as the National PiT Count Working Group.

The Guide is divided into five sections:

Acknowledgements

Employment and Social Development Canada (ESDC) gratefully acknowledges the contribution of communities and experts in Canada and internationally that provided guidance on the development of the HPS PiT Count approach.

In particular, the HPS PiT Count Working Group was instrumental in sharing their experience with PiT counts, as well as their understanding of community needs and of people who experience homelessness. Current and former members of the working group include:

  • Dr. Alina Turner, Turner Research and Strategy
  • Bob Hughes/Ken Salter, ASK Wellness, Kamloops, BC
  • Bruce Pearce, End Homelessness St. John's, NL
  • Jennifer Hales/Celine Mauboules, City of Vancouver, BC
  • Christian Méthot, Centre de recherche de L'Hôpital Douglas, Montréal, QC
  • Christina Maes Nino, Social Planning Council of Winnipeg, MB
  • Cindy Sue Montana McCormack, Social Planning and Research Council of Hamilton, ON
  • Courtenay Defriend, Courtenay Defriend Consulting, Nanaimo, BC
  • Darren Cooney/Della Knoke,, Province of Ontario Homelessness Secretariat
  • Erica Richmond, United Way Peterborough, ON
  • Gwenda Drover, Newfoundland and Labrador Statistics Agency
  • James McGregor, Mouvement pour mettre fin à l'itinérance à Montréal, QC
  • Jane Henderson, Choices for Youth, St. John's, NL
  • Jesse Donaldson, Canadian Observatory on Homelessness (COH)
  • Jesse Thistle, Canadian Observatory on Homelessness (COH)
  • Laural Raine/Joseph Stalteri/Joey Reeder, City of Toronto, ON
  • Marcel Swain/Linda Lavalee, Lu'ma Native Housing Society (Vancouver, BC)
  • Marie Morrison, Regional Municipality of Waterloo, ON
  • Matt Thompson, M. Thomson Consulting
  • Rachel Campbell/Dr. Nick Falvo/Meaghan Bell, Calgary Homeless Foundation, AB
  • Mia Burgess, Central Okanagan Foundation, Kelowna, BC
  • Michael Mackenzie, Saint John Human Development Council, Saint John, NB
  • Naomi Leadbeater/Mark Anderson/Michelle Lemoine, Brandon, MB
  • Pierre-Luc Lortie, Ville de Montréal, QC
  • Sarah Brown, Youth Leadership Council, St. John's NL
  • Dr. Stephen Gaetz, Canadian Observatory on Homelessness (COH)
  • Susan McGee/Giri Puligandla/Robbie Brydon, Homeward Trust Edmonton, AB
  • Tim Richter, Canadian Alliance to End Homelessness (CAEH)
  • Wally Czech, CAEH and City of Lethbridge, AB

Many of the working group also reached out to their community for feedback, including people with lived experience of homelessness. Their input is greatly appreciated. In addition, ESDC would like to thank other experts that were consulted on the methodology, including Dr. Alex Abramovich (Centre for Addiction and Mental Health) and Kevin Barlow (Metro Vancouver Aboriginal Executive Council). ESDC would also like to thank William Snow, Marcy Thompson and their colleagues at HUD for sharing their experiences coordinating PiT Counts and representatives from Continuums of Care in Arlington County, VA; New York City, NY; Hennepin County, MN; Anchorage, AK; Quincy, MA; and from the Balance of State for Georgia in the United States who spoke to us about their experience implementing PiT counts.

Overview of the HPS PiT count approach

The first Point-in-Time (PiT) count of homelessness coordinated in communities across provinces and territories in Canada took place in 2016. Communities are invited to participate in a second PiT count coordinated by the Homelessness Partnering Strategy (HPS) in March and April 2017. This guide explains the HPS PiT Count approach, including information on the methodology, the survey, and guidelines for communities to participate in the count. Participating communities should review this guide and resources available on the Homelessness Partnering Strategy. For a list of other resources, please see: Where to find more information?

Why do a coordinated PiT count in Canada?

Several communities in Canada have been conducting periodic PiT counts for more than a decade. These communities have taken different approaches to conducting their counts. Differences include the time of year the count is conducted, the time of day the count happens, the people that are included as "homeless", and the questions that are asked in the survey. There are also a lot of similarities across counts. For example, all counts include people sleeping in shelters and on the streets. All counts ask for basic demographic information (albeit using different wording in their questions).

In 2016, 32 communities across Canada implemented a coordinated PiT count using common methods and common questions. This count represented a step towards a national picture of homelessness. The best estimates of homelessness in Canada currently rely on the results of the Highlights of the National Shelter Study-2005–2014 (Segaert, 2016), which was based on emergency shelter use statistics from communities in most provinces and territories. One limitation of this method is that it does not include homelessness outside of the shelter system. In the State of Homelessness in Canada 2013 report, the Canadian Observatory on Homelessness and the Canadian Alliance to End Homelessness sought to address this using the results from eight PiT counts to estimate the size of the unsheltered population.

As more communities participate in 2018, a broader national picture can be created. This can be a benchmark against which progress in reducing homelessness can be measured. With the increased implementation of Housing First programs and of community plans to reduce or end homelessness, there is a need for this national picture. It is for this reason that communities across Canada have been invited to participate in the 2018 Coordinated PiT Count.

What is the purpose of a PiT count?

A PiT Count of homelessness has two primary purposes:

  • An enumeration, or count, of people experiencing absolute homelessness: It is intended to identify how many people in a community experience homelessness in shelters and on the streets at a given time. Conducted over subsequent years, PiT counts can be used by the community to track progress in reducing homelessness.
  • A survey of the homeless population: Through an accompanying survey, the PiT count gives the community information on the demographics and service needs of their homeless population. This information can be used to target community resources to where they are most needed.

A PiT Count is not intended to:

  • Be a measure of everyone who experiences homelessness in a community over time. By focusing on a single day, the count will not include some people who cycle in and out of homelessness. What it will do is provide an estimate of how many people are homeless on a given night.
  • Be a count of hidden homelessness (e.g., people who are "couch-surfing"). The focus of the count is instead on those who are absolutely homeless (e.g., sleeping in shelters or on the street) on the day of the count. Some communities nevertheless conduct the survey with the hidden homeless population in order to provide some information on their service needs.

What is the HPS PiT count approach?

The Core PiT Count defines what is common across all individual community PiT counts. It includes the Core Populations to be included in the count and the Core Questions to be included in each community's survey.

Core population

The Core HPS PiT Count approach includes people who are experiencing sheltered and unsheltered homelessness.

Unsheltered homelessness includes people who are sleeping in places unfit for human habitation, including the following locations: streets, alleys, parks and other public locations, transit stations, abandoned buildings, vehicles, ravines and other outdoor locations where people experiencing homelessness are known to sleep.

Sheltered homelessness includes people sleeping in the following locations: emergency shelters (general and specific to men, women, youth, etc.), extreme weather shelters, Violence Against Women (VAW) shelters, and transitional shelters. It may include people who receive hotel/motel vouchers in lieu of shelter beds. It does not include people who have security of tenure, who are in Housing First programs or in social or subsidized housing.

Core screening and survey questions

The Core Questions of the survey (Section 4) include screening questions that determine whether the respondent is included in the homeless count, and standard survey questions. Screening questions are intended to correctly identify people as experiencing homelessness. They need to be open enough that they include people who are experiencing absolute homelessness during the count, but restrictive enough to exclude people who are not truly experiencing homelessness (e.g., visitors to the city).

Survey questions are intended to provide more information about the population and their needs (e.g., age, gender, Aboriginal status, veteran status, income sources, reasons for housing loss, etc.). In addition to the Core Questions, you may wish to add questions that respond to your local information needs. The Canadian Observatory on Homelessness has compiled a list of other questions you can consider. Please see Section 4 for more information.

Standards for participation in the coordinated count

Recognizing the need to balance methodological rigour and community flexibility, the HPS PiT count approach includes Core Standards for the methodology that will be consistent across communities, while allowing flexibility for the approach to be tailored to each community's local context. The approach also includes recommended standards, based on effective practices used by communities in Canada. Communities are encouraged to adopt these standards.

Core standards

Core Standard 1: The enumeration is defined as the number of people experiencing homelessness on a single night between March 1 and April 30, 2018.

Core Standard 2: The methodology and survey used by the community includes the Core Screening and Survey Questions, and the Core Populations described in this guide.

Core Standard 3: The local implementation of the PiT count is based on consultations with the local HPS Community Advisory Board and Aboriginal Community Advisory Board, where applicable. The local methodology is approved by the community's HPS Community Entity and Aboriginal Community Entity, where applicable. The local methodology must be submitted to the HPS prior to the count for review.

Core Standard 4: The results of the Core Count are reported to the HPS via the software provided. This information will contribute to the understanding of homelessness across Canada, and will not be used to publicly report on findings from individual communities.

Core Standard 5: Sheltered counts are based on the number of individuals staying in emergency shelters, extreme weather shelters, Violence Against Women (VAW) shelters, and transitional shelters on one night of the year. Where applicable, it may also include families or individuals who received hotel/motel vouchers in lieu of shelter beds. Note that surveys done in shelters are not used to determine the enumeration, but rather to provide information on the sheltered population.

Core Standard 6: Sheltered counts are based on data collected by shelter data systems, where available, such as the Homeless Individuals and Families Information System (HIFIS). Large communities may use a sampling approach for their shelter survey (e.g., a representative sample of the shelters or shelter residents) provided that they explain their sampling strategy.

Core Standard 7: Unsheltered counts are based on a street survey conducted within a 24-hour period. This survey can cover the entire community, known locations within a community, a sample of neighbourhoods, or it can use a mixed approach (see Section 4 in this guide). The survey can include streets, alleys, parks and other public locations, transit stations, abandoned buildings, vehicles, ravines, and other outdoor locations where people experiencing homelessness are known to sleep.

Core Standard 8: The Community Entity (CE) is responsible for the quality of data collected. Efforts should be made to ensure that each person is counted only once (i.e., to limit double-counting) through the survey itself and through de-duplication of the data (see Section 5). The CE is also responsible for ensuring that staff and volunteers receive the proper training, including the count standards, survey procedures, data management and privacy, and personal safety.

Recommended Standards

Recommended Standard 1: The local community PiT count approach (e.g., additional survey questions or populations, areas to be surveyed, etc.) should be based on consultations with all sectors involved in homelessness, including the HPS Community Advisory Board (CAB) and Aboriginal CAB, the municipality, shelters and other homelessness service providers, Aboriginal service providers, local police and emergency services, and transit authorities, among others. Cooperation with these sectors will contribute to the successful implementation of the count.

Recommended Standard 2: If possible, the local PiT count report can include data from local public systems—including correctional facilities, juvenile detention centres, hospitals, detox centres—for people who have no fixed address and who are imminently going to be released, but have no discharge plan that includes housing. This number is not included in the Core Enumeration.

Recommended Standard 3: The PiT count coordinator should work closely with the Community Coordinator for the Homeless Individuals and Families Information System (HIFIS), as well as any local shelter data coordinator, where applicable.

Recommended Standard 4: The results of the count should be shared with the community, with particular attention to count partners, those who work in the sector and those who experience homelessness. The results should also be shared with the respective province or territory. Where the municipality has a requirement to enumerate homelessness, the CE should work closely with the municipality to ensure that the count meets their requirements, and should ensure that the municipality has access to the data necessary to report out.

Recommended Standard 5: While not part of the count, it may be useful to track contextual data that helps to explain changes in the population between years. This can include information such as rental vacancy rates, the number of people in core housing need, and the number of people in permanent supportive housing (Housing First) or long-term transitional housing.

Coordinated PiT count and registry week

Communities participating in the 20,000 Homes Campaign led by the Canadian Alliance on Homelessness may choose to combine their PiT count with a Registry Week. A Registry Week shares a lot of the features of a PiT count in that volunteers conduct surveys with people experiencing homelessness; however the purpose of a Registry Week is to begin to assess the needs of people experiencing homelessness and develop a by-name list to prioritize people for housing interventions. The differences between the approaches include that a Registry Week alone does not provide a point-in-time estimate of the population size, and that it can be carried out over two to three days (in a week).

For the 2016 PiT Count, several communities implemented a combined PiT count and Registry Week. In order to maintain the integrity of the PiT count data, several considerations are necessary:

  • A PiT count is always in reference to a single day. The first day of the combined process must be a full PiT count, meaning that the sheltered count is based on occupancy on that night and a full unsheltered count is conducted throughout the known locations within a community.
  • On subsequent days of the Registry Week, you may continue to screen in people to the enumeration, but it will always be necessary for the screening to refer to the PiT count day.
  • The Core Questions must come first in the survey.

While using a combined process means that the PiT count will no longer be an anonymous survey, the benefits of having a by-name list of people needing housing in your community has been shown by communities that have implemented this approach (e.g., Hamilton). Note that this approach should be combined with an integrated community data system (e.g., HIFIS 4) in order to keep the list up to date.

PiT Count implementation timeline

A PiT count is not built in one day. Months of work are typically required before and after the count. A detailed timeline is presented in Section 2. This work can be divided into four phases:

1. Planning (4+ months before the count). This time is used to identify a coordinator and engage the local community. A local PiT Count Committee can be formed to support the coordinator and take on various responsibilities (e.g., data analysis, volunteer recruitment, communication plan). The coordinator then works with these partners to begin developing the local survey questions and identifying a search strategy. Develop a communications plan for engaging the media, and the broader community for support and recruitment.

2. Preparation (1 - 4 months before the count). The areas to be surveyed and any additional survey questions should be finalized in this phase. A plan is developed for the day of the count (e.g., a headquarter is identified, a list of resources needed is prepared). Volunteer recruitment begins.

3. Implementation (less than 1 month before the count). This is the time for last-minute preparation for the count. Any supplies needed are obtained, and volunteers are trained and assigned to teams.

Day of the Count

4. Post-Count (1+ days after the count). The surveys should be entered as soon as possible and compiled with other data (where available). Data analysis can then proceed (if you are using the HPS PiT count software, certain reports can be generated automatically). The results should be communicated to the volunteers, the general community, and the homeless population themselves. The local PiT Count Committee should meet again to review the experience with the count (you may wish to survey volunteers and staff for their perspective) and to begin to plan for the next count.

The implementation timeline

Developing a PiT Count plan and timeline

The amount of time, effort and resources a community needs to implement its count will depend on its size, geography, and the methods chosen, yet there are common steps to be taken.

Planning (4+ months before the count)

Form a Local PiT Count Committee: Forming a local committee to support the implementation of the PiT count will serve to build community support for the count. The count will also benefit from the expertise and experience of the committee members. Membership should be drawn from the local HPS Community Entity (CE) and Community Advisory Board (CAB), as well as the Aboriginal CE and CAB, where applicable. Membership should include the municipality, police and other emergency responders who interact with the homeless population, as well as other stakeholders closely involved in supporting people experiencing homelessness. The inclusion of the local HIFIS Community Coordinator would help to facilitate access to shelter data. Membership should also include someone with lived experience of homelessness. A key element for the success of the count will be to encourage the homeless population to take part. Subcommittees may be formed to consider specific aspects of the count (e.g., recruiting volunteers, developing the local survey and methodology and the communication strategy).

Identify a PiT count coordinator: Depending on the size of the community and complexity of the count, this may be a dedicated resource. Ideally, this person will be closely connected with the local homelessness-services sector and have a background coordinating community projects. Many communities identify a municipal employee to act as the coordinator.

Reach out to the community: Develop a communication strategy for the count to engage the media and the broader community. Communicate the intention to participate in the count early. Building community awareness of the count will be key to recruiting volunteers, both for the day of the count and to support the planning and preparation phases. Local businesses may be interested in supporting the count through funding or donations (e.g., items for survey respondents, food for volunteers, etc.).

Preparation (1 - 4 months before the count)

Identify search method: Based on the size, geography and resources of the community, different approaches can be taken to defining the search areas in the community. This is particularly important for the unsheltered component of the PiT count. The approaches are described in Section 3.

Finalize survey: Communities may add to the Core Questions in the unsheltered and sheltered surveys. The local PiT Count Committee, a sub-committee or the coordinator should consult local homelessness stakeholders (e.g., homelessness service providers, community organizations that provide support, veterans supports groups) to determine what questions could be added. It will be important to balance the community's need for information, with the need to keep the survey short (less than 15 minutes, if possible). Each question should have a specific purpose related to planning or resource allocation.

Recruit volunteers: The count cannot succeed without volunteers. Early communication will be important. In addition to those who already work in the field, local leaders and community groups may be interested in participating (e.g., local politicians, student groups, religious communities, Business Improvement Associations, unions, etc.). Volunteers will likely be needed for survey teams, but may also be useful for soliciting sponsorships, coordination on the day of the count, and data entry after the count.

Prepare a plan for the day of the count: Identify a central headquarters for coordinating the count. Larger communities may want area hubs, neighbourhood satellite locations where volunteers meet before and after their shifts at nearby shelters or routes. Determine what your survey teams will look like. Identify what resources you will need (e.g., gear for survey teams, incentives for participation). Develop a plan for organizing and inputting incoming surveys.

Implementation (Less than 1 month before the count)

Train volunteers: Each volunteer must receive training on how to approach people, how to ask the screening and survey questions, how to fill the survey form, and how to request assistance, if required. Volunteers should receive training close to the day of the count to reduce the likelihood that the volunteers who show up for the training will be absent for the count, or that volunteers will show up on the day of the count, only. See Section 3 for more information on volunteer training.

Final preparation: Purchase supplies and prepare kits for survey teams. Prepare search maps for the survey teams (city planning departments may be able to help to develop these). Deliver forms and supplies to shelters and area hubs. Those identified through the unsheltered portion of the count may be in need of emergency services or shelter. An outreach team (or teams) should be formed to quickly address these needs on the day of the count.

Day of the count

The day of the count will look different depending on the timing of the sheltered and unsheltered portions of the count, and whether you are including other survey locations such as service locations and magnet events.

For resources relating to the day of the count, including an overview of the count from the perspective of the volunteer, considerations for the set-up of your headquarters, and considerations for surveying in shelters, please see Community Workspace on Homelessness.

Post-Count (1+ days after the count)

Enter count data and survey responses: A volunteer team can enter the survey data. The HPS-provided software may be used to enter the data and produce some basic reports. Ensure that the responses are kept confidential in order to respect the privacy of the respondents. Data entry should be completed as close to the count as possible to facilitate data cleaning and allow for follow-up with surveyors on unclear survey responses.

Analyze results and submit to HPS: The HIFIS PiT Module is provided to offer a secure method to upload the results to the HPS. The local PiT Count Committee can be consulted to determine what analyses should be done. While the PiT Module can produce basic reports, analyses will require spreadsheet software or statistical packages (for more advanced analyses). For first counts, PiT count reports from other communities may suggest analyses to try.

Communicate with the community: To maintain community engagement on the count and support for future counts, a communication strategy is essential. Be prepared for significant media attention concerning the count. A specific strategy to communicate with the homeless population, themselves, is strongly encouraged. This may increase their willingness to engage in future PiT counts. If possible, communicate the preliminary results quickly.

Prepare for the next count: The local PiT Count Committee should consider the lessons learned from the count and what could be improved for the next count (e.g., questions, logistics, search areas). A survey or focus group of volunteers and staff from the count is encouraged to get a range of perspectives.

Developing your local approach

The Core HPS PiT Count Methodology is intended to standardize the basic elements of the count across communities. This approach leaves room for you to adapt the approach to your local needs. This work will be led by your PiT Count Coordinator with support from a PiT Count Committee. This section describes key elements of the count that are necessary for PiT Count implementation.

Involving community partners

A successful count is dependent on the involvement of the community, preferably in the form of a local PiT Count Committee. In most communities, the PiT count coordinator will not be able to do all of the planning and preparation on their own. The Committee would help the coordinator by taking on certain responsibilities, and would help to build support within the community for the count. Subcommittees can be formed to focus on certain aspects of the count, such as volunteer recruitment, local survey questions, and seeking support from the community.

It is essential to have the homeless service organizations in the community on board, particularly those that provide frontline services, such as outreach and shelter, because their expertise and familiarity with the homeless population will be necessary for the count. These organizations should include those that focus on particular vulnerable populations, including those that provide services to youth, Aboriginal people, and veterans (e.g., the Legion), where available. You can also consider approaching local social planning councils who may have a research mandate that can provide some expertise to the count. Local charitable organizations can also provide support, particularly with the recruitment of volunteers.

Local municipalities in the community may be able to provide administrative and logistical support for the count. They may have maps available and can help define the search areas. They can also serve to connect the coordinator with municipal services and emergency responders, such as police, social workers, and paramedics, many of whom are likely in close contact with the homeless population. In addition, local libraries, municipal parks, and the transit authority may be contacted to help identify locations, such as bus shelters or transit stations that may serve as sleeping locations.

Local community leaders can help to promote the count with the broader community. Local politicians (federal, provincial, municipal), celebrities, and other community leaders can raise the profile of the count, which can serve to support volunteer recruitment and build awareness of homelessness issues more broadly. Getting community leaders on board may also build support for addressing homelessness issues in your region that are identified by the count.

If the community has a shelter data coordinator, such as a HIFIS Data Coordinator or Community Coordinator, this person or organization should be involved in the local PiT Count Committee to facilitate access to shelter data.

Local universities and colleges may be able to provide support with the technical aspects of the count, including defining the search areas (particularly if a sampling approach is used) and analyzing the data. Students may be interested in volunteering on survey teams, and for post-count data entry and cleaning.

Local businesses or business associations (e.g., business improvement associations) may be interested in supporting the count through funding, supplies for volunteers (e.g., clip boards, paper, food) or for the homeless population (e.g., socks, blankets, restaurant or grocery store gift cards), and for providing volunteers.

You should also consider including people with lived experience of homelessness on the Committee. They can help to validate any local questions added to the survey, help to identify locations for survey teams, and solicit others to volunteer on the day of the count.

Engaging volunteers

Volunteer recruitment should begin in the months before the count (see Section 2). Volunteers may be found from the networks of those on the local PiT Count Committee, as well as the general population. Most communities will include a solicitation of volunteers when the count is announced.

Many communities use an online forum or service for volunteers to sign up (e.g., VolunteerSpot, Survey Monkey), often for a particular role or shift. Volunteers may want to be on a survey team together, but you should ensure that each team has at least one person with professional knowledge of the homeless population.

It is important to maintain communication with volunteers leading up to the count. Communication can involve reminders of what to bring to the count and training materials. Invariably some who sign up to volunteer will back out before the day of the count. Reminding them to inform you if they are unable to participate will mean you will have fewer last-minute adjustments to your survey teams.

After the count, it is important to acknowledge the contribution of volunteers and to communicate the outcomes of the count. Volunteers that see the difference that they have made will be willing to volunteer again on your future counts.

Some smaller communities have conducted counts using only outreach and shelter staff. This approach is feasible only where sufficient staff resources are available, and where the homeless population is relatively small and stable. The advantage of this approach is that the staff is already used to working with the homeless population, and may know the majority by name. One disadvantage of this approach is that it reduces the community engagement potential of the count. Volunteers may have never met or spoken with someone experiencing homelessness before. Volunteering is an opportunity to reduce barriers within communities and for the volunteers to gain an understanding of homelessness.

Forming survey teams and training

Survey teams are assigned to sheltered and unsheltered locations during the count. Each team should consist of two to three people, ideally including at least one person who has experience working with the homeless population. This could be an outreach worker, shelter staff, public health worker, or social worker, among others. People from local community partners or the general public may be solicited to volunteer on the survey teams. In areas with significant minority language populations, teams should have one member who is fluent in that language, if at all possible.

If police, or other volunteers whose professions require uniforms, want to volunteer to be on survey teams, you may wish to ask them to come in plain-clothes on the day of the count. A uniform may be intimidating for some among the homeless population. The same applies to clubs or other social associations that volunteer together. As volunteers, they are acting as representatives of the count.

Training should be done close to the date of the count so that the information is fresh in the mind of the volunteers. If it is done too far in advance, you may lose a significant number of volunteers because they have unexpected commitments. If possible, provide options for training close to the day of the count to account for differences in work schedules.

Training should include simulations to familiarize volunteers with the forms and the procedures. Training materials are included in the PiT Count Implementation Toolkit developed by the COH with the support of the HPS, and are available on the Community Workspace on Homelessness.

The following topics should be covered by the training:

  • The purpose of the count: Why it is happening and what are the expected benefits for the community and for the respondents.
  • The population: Who experiences homelessness and why. There is an opportunity to educate volunteers on the population and the causes of homelessness. This could include specific information on homelessness experienced by Indigenous people and youth.
  • The approach: How to approach someone to participate in the survey. An explanation of the honoraria/incentives. How to ask the survey questions in a respectful manner.
  • The survey: How to screen for homelessness. What each question is asking for. The importance of collecting reliable data.
  • Logistics and safety: What materials will be provided to survey teams. What to wear and bring on the day of the count. Who to contact in the case of an emergency.

Because of the amount of material that you will want to cover during the count, you may wish to provide some background material prior to the training session.

Defining the scope of the count

The sheltered count

The sheltered component of the PiT count should include the following locations:

  • Emergency shelters: These include general shelters, as well as those for specific populations, such as women, men, youth and Indigenous people. They should also include severe weather shelters and informal shelters (e.g., those set up by charities or religious organizations).
  • Transitional shelters (short-term supportive housing): these programs are meant as a step to permanent housing, where clients can remain for longer terms (less than a year). Transitional housing with stays guaranteed for longer than a year should not normally be included.
  • Violence Against Women (VAW) shelters: these shelters help women seeking safety from violence and abuse. Most shelters will also allow accompanying children.
  • Hotel vouchers: Where applicable, the shelter count can also include families and individuals provided with hotel/motel vouchers in lieu of a shelter bed.

Where possible, data available for these facilities can help to determine the Sheltered Count. HIFIS 3 Software is available for free, with free training from the HPS. Where it is not possible to get this information from a data system, sample forms are available to request occupancy information from a shelter (PiT Count Implementation Toolkit). At a minimum, the information should include the number of people who spent the night, although you may also want to request the gender and age of each client on that night.

In the largest communities, if insufficient resources are available to survey all clients at all shelters, you may consider a sampling approach. This may entail taking a representative sample of shelter facilities or a random sample of clients. This approach would require expertise in statistical sampling.

The unsheltered count

The unsheltered component of the PiT count should include any outdoor or public locations that are not meant for human habitation. These can include streets, alleys, parks, ravines, under bridges, bus shelters, abandoned buildings and parked cars. It is important that the scope be as thorough as possible in order not to exclude any part of the homeless population.

In geographically small or densely populated areas, it may be possible to take a Full Coverage approach, where these locations are surveyed throughout the entire community. This is the ideal, but is also resource-intensive as it will require a large number of volunteers and coordinators. For most communities, an alternative approach is needed.

The most common approach is a Known-Locations search, where teams canvass areas known to be frequented by people experiencing homelessness (e.g., parks, ravines, camp sites). This approach is particularly useful for rural areas and sparsely populated urban and suburban areas. The locations can be identified by service providers and emergency responders. Since rural areas can be large and sparsely populated, some communities also conduct a pre-count search to identify potential camp sites.

In very large communities that have areas with a lower population density, communities may wish to consider a Sampling approach for these areas, whereby a random sample of areas is searched. These areas can be census tracts. The count from the sampled areas would be extrapolated to the areas not sampled. Use of this approach requires expertise in statistical sampling.

In communities with a mix of urban, suburban and rural areas, a mix of approaches is appropriate. For example, Toronto does a full coverage survey of the downtown core and surveys a statistical sample of suburban areas. Community geography, size and resources will affect whether Full Coverage, Sampling, Known-Locations or a combination of all three approaches is most appropriate.

Many communities include people who are observed to be clearly homeless in their enumeration. Observations may be used for individuals who do not participate in the survey, but who show clear signs of being homeless on the night of the count (e.g., if they are sleeping in an unsheltered location and they have their belongings with them). Note that the first option should be to ask the screening questions, if at all possible. The screening questions provide more reliable data and give an opportunity to the respondent to participate in the survey.

Some communities supplement their count with an estimate of the number of "missed" people. One method is to use "decoys" on the day of the count. These are volunteers who are situated outside in search areas of the city during the count. For each decoy that does not encounter a search team, a number of missed people is estimated and added to the number counted. This approach should be taken with caution. It is very difficult to confirm that the estimate of missed people reliably corresponds to actual people experiencing homelessness on the day of the count.

Building on the core populations: public systems

Some communities include people who are in "public systems" (e.g., hospitals, detox centres, corrections facilities) who will be discharged into homelessness in the short term. Your community may wish to count these people for local planning purposes. This population was not included in the core approach since it would require every participating community to secure access to data from each institution. These institutions are governed by various jurisdictions at municipal, provincial, and federal levels. The burden of requiring this of communities was considered too onerous.

Screening in people within systems can be difficult. Without being able to conduct surveys, it can be difficult to determine whether an individual should be included in the count. Some communities have been able to conduct surveys with people in correctional institutions, for example, although this is rare. Data systems are used by other communities to identify people with no fixed address (or who give a shelter as their address). This solution risks including people who are in systems for long periods of time who have no need for a permanent address. Further guidance and recommendations for including systems in a PiT count will be forthcoming.

Building on the core populations: hidden homelessness

People that are staying with others because they are unable to secure a place of their own are often referred to as experiencing "hidden homelessness". A PiT count is unable to enumerate hidden homelessness because people experiencing hidden homelessness are unlikely to be in the count locations and there is no means to determine how many individuals were not encountered.

Nevertheless, you may consider conducting the survey with those experiencing hidden homelessness to better understand their needs. To properly screen in this population, you will need follow-up screening questions. See Section 4 of this guide for recommended follow-up questions and considerations.

If your community is interested in estimating the size of the hidden homeless population, an alternative approach is recommended (e.g., a period-prevalence study, a phone survey, etc.).

Populations outside of the scope of the PiT count enumeration
  1. People who are staying in housing that they rent or own, including those who have access to a permanent housing that they can safely return to. For example, a visitor to the city staying with a friend may have access to permanent housing in another city. Likewise, a youth staying with someone else for a night may still have permanent housing with their parents.
  2. People who are in permanent supportive housing or Housing First programs. The PiT count, in part, is meant to measure success in reducing homelessness through these Housing First and supportive housing programs, so including them among the numbers would hide this progress. Nevertheless, you may want to use the PiT count as an opportunity to speak about the success you have had in housing people through these programs. This information can help to contextualize changes in the population over time.
  3. People that are experiencing overcrowding. This is an important contextual piece of information and highlights a need for housing within a community. Because it is not possible to assess this in the context of a PiT count, this population is considered outside of the scope of the count.

Observed homelessness

Not everyone encountered will be interested in answering screening questions. Others may be unable to respond to the questions (e.g., unable to wake, language barriers). Some of these individuals may still be experiencing homelessness. To enumerate those that are apparently homeless but who are not participating in the survey, your community can consider tracking observed homelessness.

The criteria for including someone as observed homeless should be clearly identified. For example, individuals may be considered homeless if they bedded down in an unsheltered location and they have many belongings with them (e.g., backpacks, garbage bags, shopping cart, sleeping bag, bedrolls, etc.).

Observations should only be used as a last resort, since the data they provide are less reliable than those based on self-report. In addition, asking the screening questions provides the opportunity to the respondent to be screened into the survey, receive the honorarium/token, and receive information about services. If observations are used, the proportion of your count that is based on observations homelessness should be clearly indicated in your reporting.

Data management

As you are planning your count, you will need to prepare for data entry, storage, and analysis. Each of these will influence the questions you add to your local survey, the recruitment of community partners and volunteers, and the resources needed. In particular, it is important to be able to ensure the privacy of the survey respondents. Survey forms will need to be locked in a secure location and the raw data should be shared only as necessary, and without compromising the confidentiality of the respondents.

Support for this element of the count may be provided by the municipality, where sufficient resources are available, or by a third party (e.g., university or college). The organization should have rules and systems in place for meeting requirements for data storage, and the capacity to conduct the analyses.

See Section 5 for more information.

The surveys

The script and consent

Survey teams should be provided with a script to follow for each encounter. The script is designed to be as brief as possible, while being clear about the purpose of the survey and how the results will be used. In order to meet basic ethical standards, it is important that the respondent understands how the information provided will be used. The script also informs the person that the information they provide will be kept confidential and not reported at an individual level.

The core questions

The Core Screening and Survey Questions are below, and can be found in sample survey forms on the Community Workspace on Homelessness. The sheltered and unsheltered survey forms are almost identical with the key difference being the screening criteria. For the unsheltered count, only individuals who will be unsheltered on the night of the count should be screened into the survey.  In contrast, for the sheltered count, individuals staying at the shelter may be screened in, as well as those who are present at the shelter without a bed, and are planning to sleep in an unsheltered location. Communities who add to the Core Populations (e.g., by including hidden homelessness or people in public systems) may adjust the screening filters accordingly.

The Core Questions include items intended to identify whether the respondent is homeless for the purpose of the PIT Count, to provide some demographic information about the homeless population and to highlight potential service needs. They can also be used to identify potential supports for the respondent. For example, veterans may be eligible for supports through the Veterans Affairs Canada, or may be connected with veteran support groups. See the list below for the screening and survey questions with an explanation for each.

Core screening questions

Screening question A

Have you answered this survey with a person with this [identification e.g., button]?

  • Yes
  • No

All search teams should have an identifying item or piece of clothing. Over the course of the count, the same individual may be approached by two survey teams. This question is intended to reduce double-counting.

Screening question B

Are you willing to participate in the survey?

  • Yes
  • No

This question follows a brief explanation of the purpose of the survey.

Screening question C

Where are you staying tonight?

  1. Decline to answer
  2. Own apartment/house
  3. Someone else's place
  4. Motel/hotel
  5. Hospital, jail, prison, remand centre
  6. Emergency shelter, domestic violence shelter
  7. Transitional housing
  8. Public space (e.g., sidewalks, squares, parks, forests, bus shelter)
  9. Vehicle (car, van, rv, truck)
  10. Makeshift shelter, tent or shack
  11. Abandoned/vacant building
  12. Other unsheltered location (specify) [insert location]
  13. Respondent doesn't know [likely homeless]

This question includes a list of sheltered and unsheltered locations. It will be the primary tool for determining whether a respondent is considered homeless. The screening procedure will depend on whether you are surveying only the Core Populations, or including hidden homelessness. For the Core Populations, responses f to m screen-in to the survey. In the case of a daytime unsheltered count, the question would be, "Where did you stay last night?"

Core survey questions

Survey question 1

What family members are with you?

  • None
  • Partner
  • Child(ren)/Dependent(s) [indicate age and gender]
  • Other adult [insert answer]
  • Decline to answer

This question is asked first so that family data is captured together. The survey number for each adult family member who is surveyed would be included on the forms for each other adult family member. The age and gender of each dependent child that is with the respondent during the count is noted on the form. Families have distinct service needs. This will allow you to identify the number of homeless families in your community and track progress in reducing family homelessness.

Script: For the next questions, "homelessness" means any time when you have been without a secure place to live, including sleeping in shelters, on the streets, or living temporarily with others.

Survey question 2

How old were you the first time you became homeless?

  • Age (in years): [insert age]
  • Don't know
  • Decline to answer

The purpose of this question is to better understand the respondent's history of homelessness. People who experienced homelessness as youth are at greater risk of experiencing later homelessness.

Survey questions 3 and 4

In total, how much time have you been homeless over the past year?

  • How long [insert number] days / weeks / months
  • Don't know
  • Decline to answer

In total, how many different times have you experienced homelessness over the past year?

  • Number of times [insert number]
  • Don't know
  • Decline to answer

Surveyors should try to get a best estimate if at all possible. These questions provide an indication of potential chronic or episodic homelessness. Housing First interventions prioritize services for these vulnerable groups. Responses to these questions will help you to track your community's progress in reducing homelessness and can help to reach people who may be eligible for Housing First programs.

Survey question 5

Have you stayed in an emergency shelter in the past year? [Prompt: give local examples]

  • Yes
  • No
  • Don't know
  • Decline to answer

Many communities have comprehensive shelter use data, but little information on the homeless population that does not use shelters. This question can provide information on the size and needs of this population. For the surveys with individuals staying in a shelter, the surveyor can indicate yes.

Survey question 6

How long have you been in (community name)?

  • Length [insert number] days / weeks / months
  • Always been here
  • Don't know
  • Decline to answer

[If length indicated] Where did you live before you came here?

  • Indicate community & province/country [insert answer]
  • Decline to answer

Some communities experience high rates of migration among the homeless population. Recent arrivals may not know where to go to access services in the community. The inclusion of this question can provide your community an indication of the number of recent arrivals and their needs. The follow-up will provide information on inter-community migration.

Survey question 7

Do you identify as Indigenous or do you have Indigenous ancestry? This includes First Nations, Métis, Inuit, with or without Status. [If yes, please follow-up to specify.]

  • Yes
    • First Nations
    • Inuit
    • Metis
    • Non-status or have Indigenous ancestry
  • No
  • Don't know
  • Decline to answer

You may consider including "Aboriginal" or locally-used terminology here, in consultation with your community. Aim to ensure that the language you use is respectful and widely understood by potential respondents. Survey teams can provide examples that suit the local context, being as inclusive as possible. Indigenous people are over-represented among people experiencing homelessness. This question can provide more context on this population and track progress in addressing their needs.

Survey question 8

Did you come to Canada as an immigrant or refugee?

  • Yes, immigrant
  • Yes, refugee
  • No
  • Don't know
  • Decline to answer

[If yes] How long have you been in Canada?

  • Length [insert number] days / weeks / months / years
  • Or date [insert date] [day | month | year]
  • Don't know
  • Decline to answer

Most newcomers (immigrants and refugees) will have received support with resettlement in Canada. High numbers of newcomers may signal an issue that can be addressed upstream in order to prevent homelessness.

Survey question 9

Have you ever had any service in the Canadian military or the RCMP?

[prompt: Military includes the Canadian Army, Navy and Air Force]

  • Yes, Military
  • Yes, RCMP
  • No
  • Don't know
  • Decline to answer

This question was recently integrated into HIFIS, providing more information about shelter use among veterans. PiT counts provide more context for their experience of unsheltered homelessness. In addition, former members of the Canadian military and of the Royal Canadian Mounted Police (RCMP) may be eligible for support through Veterans Affairs Canada as well as veterans support groups.

Survey question 10

How old are you [OR] What year were you born? [If unsure, ask for best estimate]

  • Age (in years): [insert age]
  • Or Year: [insert year]
  • Don't know
  • Decline to answer

Respondents may answer with their age or year of birth. It provides basic information on the age of the homeless population. Youth, adults and seniors each have needs for different services and approaches.

Survey question 11

What gender do you identify with? [Show list]

  • Male/Man
  • Female/Woman
  • Two Spirit
  • Trans Male/Trans Man
  • Trans Female/Trans Woman
  • Genderqueer/Gender Non-Conforming  
  • Not listed: [insert answer]
  • Don't know
  • Decline to answer

As with age, it this question provides basic demographic information on the homeless population. Surveyors show the response options to the respondent to help ensure that they can self-identify. "Not listed" responses must be specified.

Survey question 12

How do you describe your sexual orientation, for example straight, gay, lesbian? [Show list]

  • Straight/Heterosexual
  • Gay
  • Lesbian
  • Bisexual
  • Two Spirit
  • Questioning
  • Queer
  • Not listed: [insert answer]
  • Don't know
  • Decline to answer

People identifying as LGBTQ have been found to be overrepresented among homeless youth. The purpose of this question is to better understand their needs and inform service delivery. Surveyors show the response options to the respondent to ensure they can self-identify.

Survey question 13

What happened that caused you to lose your housing most recently? [Do not read options. Select All That Apply. "Housing" does not include temporary arrangements (e.g., couch surfing) or shelter stays.]

  • Illness or medical condition
  • Addiction or substance use
  • Job loss
  • Unable to pay rent or mortgage
  • Unsafe housing conditions
  • Experienced abuse by spouse / partner
  • Experienced abuse by parent / guardian
  • Conflict with spouse / partner
  • Conflict with parent / guardian
  • Incarcerated (jail or prison)
  • Hospitalization or treatment program
  • Other reason  [insert answer]
  • Don't know
  • Decline to answer

This is an open question with categories provided. Surveyors would select the categories that best fit the response to the question. These responses can help to determine the pathways into homelessness in your community and identify potential areas to focus on interventions to prevent homelessness.

Survey question 14

Where do you get your money from? [May give examples from list. Select All That Apply]

  • Employment
  • Informal/self-employment (e.g., bottle returns, panhandling)
  • Employment insurance
  • Welfare/social assistance
  • Disability benefit
  • Seniors benefits (e.g., CPP/OAS/GIS)
  • Child and family tax benefits
  • Money from family/friends
  • Other source: [insert answer]
  • No income
  • Decline to answer

This is an open question with categories provided, including those related to employment, informal sources (e.g., bottle returns), tax benefits, and family and friends. This question can be used to identify which supports are being accessed and which are not. For example, it could indicate whether homeless seniors are accessing Old Age Security or the Guaranteed Income Supplement.

Families and children

The family question asks the respondent to report the age and gender of each dependent child that is with him or her during the count. Other family members (partners, parents, grandparents, etc.) should also be linked. Each survey form should have a unique survey number that can serve as an anonymous identifier for the respondent (see Community Workspace on Homelessness for an example). In the case of families, each member should have her or his number indicated on the forms for their family members next to their relationship (i.e., partner, parent, other). This serves to keep the information on the family together and can provide information about family size.

Building on the core questions

The Core Questions of the HPS approach provide some demographic information on the homeless population, but you may want to consider additional questions for your local count. However, be cautious when selecting questions since the length of the survey should normally be kept to 15 minutes or less (with some exceptions).

What questions should you add? The Canadian Observatory on Homelessness, in consultation with several communities, has developed a set of questions that can be added to the HPS Core Questions. These questions are not required as part of the HPS approach, but you can review them to see whether they meet your community’s information needs. They can be found here: PiT Count Implementation Toolkit.

Including hidden homelessness

While it is not possible to enumerate hidden homelessness with a PiT count, conducting surveys with people experiencing hidden homelessness can provide valuable information about the population and their service needs. The key challenge with this approach is to have screening questions that correctly screen in people who are experiencing hidden homelessness and screen out those with access to secure housing. In part, the challenge is due to the difficulty of defining hidden homelessness. A working group of community representatives and researchers was formed to provide a recommended set of questions.

The approach involves follow-up questions to people whose response to screening question C is that they are staying at someone else's place. For these responses, the two follow-up questions are recommended:

Recommended hidden homelessness screening questions

Screening C1

Can you stay there as long as you want or is this a temporary situation?

  1. Yes
  2. No
  3. Don't know
  4. Decline to answer

The purpose of this question is to screen out people who are staying longer term in secure housing that they do not consider their own (e.g., a partner's house). Those who are staying temporarily with others, or who are not sure (responses b. or c.) would be asked the next question (C2).

Screening C2

Do you have a house or apartment that you can safely return to?

  1. Yes
  2. No
  3. Don't know
  4. Decline to answer

The purpose of this question is to screen out people that have access to safe, secure housing that they are not staying in on the night of the count. Examples include people who are visiting the city and staying with a friend, or who are house sitting. Those who answer no or don’t know (b. or c.) would then be screened into the survey.

It is important to note that there will always be exceptions and that they may still lead to some people being screened out who should be screened in (false negatives) or screened in who should be screened out (false positive), but this is true of all of the alternatives that were considered. Carefully training the volunteers on the populations you are trying to reach through the survey will help them discern who should and who should not be included.

Implementing the sheltered survey

Preparation: In most communities, shelter clients make up the majority of the PiT count. Shelters should be approached early to support their inclusion in the count. Each shelter should identify a key contact who will be present the night of the count. If you will be using shelter systems data, you will want to develop a data provision agreement with the shelter if one is not already in place.

Some shelters may be reluctant to have survey teams stationed in their facility. Depending on the size and resources of the shelter, they may be able to conduct the count with shelter staff (who should receive the same training as the volunteers). In particular, this may be true of domestic violence shelters.

Other shelters may be unwilling to share the raw data of their clients. In these cases, seek as much information as they are willing to share. At a minimum, this should include the number of clients for that night. In the largest communities, it may be possible to use data from similar shelters to estimate the characteristics of clients in shelters with missing data.

The number: To the extent possible, your count (or enumeration) of people in a shelter facility should be informed by systems data. If you are unable to use systems data, it is recommended that you request the information from the shelter. Some options for doing so can be found here: PiT Count Implementation Toolkit.

The survey approach: Surveyors (shelter staff or volunteers) may choose to conduct a full survey of the shelter, or to take a sample based on the number of people who check in on the night of the count. Because some shelters allow very late check-ins, you may consider conducting the survey the next day. This may be particularly helpful for people who are unable to consent for the survey during the night of the count. Consult each shelter to determine the optimal time to conduct the survey.

Consent: Surveyors must obtain consent from the person to proceed with the questionnaire. It is important that consent be obtained to ensure that the respondent is aware that confidentiality will be maintained and will understand how their responses will be used.

Screening: In a sheltered location, only screening questions A and B are necessary if you can confirm that you are only speaking to clients who are staying overnight in the shelter.

The questionnaire: There are 14 Core Survey Questions for the sheltered survey. Each question should be asked as it is written in order to encourage consistency in the interpretation of the question.

Honorarium: The honorarium may be seen as a means to engage the person, as a thank you for participating in the questionnaire, or both. Depending on this viewpoint, the honorarium may be given after the person has been identified as homeless, after they have completed the questionnaire, or it may be split in two, with one at each point. The honoraria may include small items (e.g., socks, toiletries, snacks) and gift cards (e.g., for a coffee shop or grocery store). These can be purchased or donations may be solicited leading up to the count.

Services: If it is not done already by the shelter staff, you may wish to use the PiT count as an opportunity to provide information on available services. Some communities offer respondents a list of available services in the community. Depending on survey answers, respondents may be eligible for various supports. For example, people who are chronically homeless may be eligible for Housing First supports, seniors may be eligible for support through Old Age Security or the Canada Pension Plan, and veterans may be eligible for supports through Veterans Affairs Canada. Shelter staff may follow-up with respondents to take the next steps to explore eligibility.

Implementing the unsheltered survey

Preparation: The locations to be surveyed will be identified by the coordinator and local PiT Count Committee where applicable. Route maps should be prepared for each survey team.

The number: The unsheltered number is determined by the number of surveys returned with unsheltered overnight locations (Screening Question C). If your community is including observed homelessness, each surveyor should be provided a tally sheet in addition to the survey forms (see Community Workspace on Homelessness for examples). The tally sheet provides a record of those who do not participate in the count. The number of people observed to be experiencing homelessness would be added to the total from the surveys.

The survey approach: Survey teams should approach everyone they encounter to ask the screening questions. In some communities, it has been noted that people are reluctant to participate in a "homelessness survey", so it is labeled a "housing needs survey" in the recommended script, while still being clear that the purpose is to better understand homelessness.

Consent: As with the sheltered count, surveyors must obtain consent from the person to proceed with the questionnaire. Consent must be obtained to ensure that the respondent is aware that confidentiality will be maintained and understands how his or her responses will be used.

Screening: All three screening questions must be asked during the unsheltered survey.

The questionnaire: There are 14 Core Survey Questions for the unsheltered survey. Each question should be asked as it is written in order to encourage consistency in the interpretation of the question, although it is possible to provide clarifications where necessary.

Honorarium: The honorarium may be seen as a means to engage the person, as a thank you for participating in the questionnaire, or both. Depending on this viewpoint, the honorarium may be given after the person has been identified as homeless, after they have completed the questionnaire, or it may be split in two, with one at each point. The honoraria may include small items (e.g., socks, toiletries, snacks) and gift cards (e.g., for a coffee shop or grocery store). These can be purchased or donations may be solicited leading up to the count.

Services: If possible, you should offer to find shelter for the person for that night. You may also wish to use the PiT count as an opportunity to provide information on available services. Some communities offer respondents a list of available services in the community. Depending on survey answers, respondents may be eligible for various supports. For example, people who are chronically homeless may be eligible for Housing First supports, seniors may be eligible for support through Old Age Security or the Canada Pension Plan, and veterans may be eligible for supports through Veterans Affairs Canada. Staff may routinely explore these options, so it is advisable to work closely with them when conducting the survey to ensure that there is no overlap in efforts.

After the PiT count

The amount of work needed after the actual PiT count should not be underestimated. Depending on the size of the community and number of completed surveys, data entry, cleaning, and analysis can take weeks after the count. You will also need a communication strategy for the findings that can be initiated after the count. The work of the local PiT Count Committee will continue after the day of the count as you review what went well, what can be improved, and what can be explored in future counts.

The next day

Collect surveys: Completed surveys should be collected from shelters and survey teams on the night of the count or the following morning, at the latest. Collecting the surveys quickly reduces the likelihood that the forms will be lost or damaged.

Acknowledge contributions: A PiT count cannot happen without the contributions of a range of project partners. These include members of the local PiT Count Committee, local shelters, volunteers, donors, and any local organizations who were consulted. A short thank you message can be shared with them immediately after the count acknowledging their investment of time and/or money to the success of the count. This encourages their ongoing engagement on the issue of homelessness, and increases the chances that they will support future counts.

Data management and analysis

The PiT count module of the HIFIS software will be provided to communities to facilitate data entry, analysis and reporting. Each survey response is entered into the software and data are stored for the Core Questions and any additional questions added by your community. The software facilitates data cleaning by highlighting potential double entries or unrealistic birthdates. Once data are entered, they can also be exported as a spreadsheet. The software can automatically generate a basic PiT count report based on the Core Questions, and can be customized to generate other reports to account for additional questions. The software can also be used to securely transmit the results to the HPS.

Support on installation and its use will be available through the HIFIS Support Desk.

Storage and privacy

PiT count responses should be anonymous. Surveys do not ask for names or other personal identifiers. Even without identifiers, however, survey responses can sometimes be used to identify specific individuals. To mitigate this risk, the data should be stored on a secure computer or server, with limited access. If possible, the file should be password-protected. Individual results or raw data should never be presented or reported. Volunteers and researchers who work with the data should be asked to sign confidentiality agreements.

The physical survey forms should also be stored in a secure location (e.g., a locked cabinet). You may wish to keep the forms between PiT counts in case the entered data needs to be verified. Once they are no longer needed, forms should be shredded or otherwise destroyed.

If your community is collecting names through its PiT count (e.g., to create a registry of people experiencing homelessness), please ensure that the data submitted to the HPS is anonymized, such that no individual person is identifiable.

Data cleaning

Once the surveys have been brought to a central location, data entry should begin as soon as possible. You may wish to develop an agreement with a local university or college to do the data entry. In return for access to the data, a university or college may be willing to provide support with data entry, cleaning and analysis.

In order to reduce the likelihood of errors during the survey entry process, it is recommended that each survey be entered into two separate databases, and then responses compared to identify any inconsistencies. The physical survey form can then be consulted to verify the response in the event of any discrepancies.

A contact from each survey team should be available during this process to provide clarity if there is any ambiguity in the written responses. Reasons for this can include:

  • unclear writing or checkmarks;
  • multiple checks for questions requiring one answer;
  • missing responses; and
  • answers that are likely miswritten (e.g., a birthday in 1901, a 16-year-old Veteran).

After the data has been entered, it should be reviewed to identify suspect entries (e.g., a 16-year-old veteran) and to de-duplicate the data (i.e., identify people who have answered the survey twice). Regardless, you will need to manually review the data for errors. For example, when two entries have answers that are the same across many demographic variables (e.g., age, gender, veteran status, Aboriginal identify, etc.) they may be double entries. Note that when you have a large number of respondents, there is a greater chance to find two individuals with similar demographics, so the standard for de-duplication should be higher. To facilitate the process, you may want to consult a statistical expert to set a working standard.

Data analysis and interpretation

The enumeration

The Core Enumeration is the total number of people who stayed in sheltered and unsheltered locations on the night of the count. Because this number is based only on the Core Populations, it is a number that can be compared across years and can be rolled-up across communities. It is not a perfect number—some people who are homeless in unsheltered locations may decline to be surveyed—but it represents the minimum number of people who were homeless on the night of the count.

It is calculated as A + B + C, where:

  1. The number of people in shelters (including extreme weather shelters and VAW shelters)
  2. The number of people in transitional shelters
  3. The number of people identified as experiencing homelessness in unsheltered locations (i.e., who gave an unsheltered location when surveyed + those who were observed homeless, where applicable)

Survey results

Before analyzing the data, it is important to remember that a PiT count should not be considered representative of all people who experience homelessness in your community. Because it is done only over a single day, it is more likely to include people who experience long-term, chronic homelessness than those that experience only single episodes of homelessness. As mentioned before, it is a snapshot of people who are experiencing absolute homelessness on the day of the count.

The HPS-provided software can generate a number of reports based on the Core Questions. This will give you basic demographic information on the respondents. They include reports on age and gender, veterans, and Aboriginal status, among others.

If this is your community’s first count, analyses will be fairly straightforward, looking at relationships among the variables. Once you have completed two or more counts, you will be able to track changes over time. For these analyses, use of a statistical package is recommended (e.g., SPSS, SAS, Stata).

If you do not have the necessary capacity to conduct the analyses in-house, you may wish to hire a contractor or develop a partnership with a university or college.

Submitting results to the Homelessness Partnering Strategy

Communities receiving support for participating in the HPS-coordinated PiT count have been asked to submit their results to the HPS. The HPS-provided software is equipped with a function to securely report the results to the HPS once the data have been entered and cleaned.

The data will not be shared with third parties and the results from individual communities will never be reported. Instead, your community’s data will contribute to a cross-Canada picture of homelessness. It will deepen our understanding of the population affected by homelessness, and the supports that may be needed to help them find secure housing.

Communicating the results

Once the count has happened, there may be a lot of anticipation among the community, and the project partners in particular, to see the results. Depending on the size of the data set, it may take weeks to enter, clean and analyze it. In some communities, a preliminary report is made with a few preliminary details, which is later followed by a full report. Examples of community PiT count reports can be found here: Community Workspace on Homelessness.

Communicating with the general community. The first results should be released as soon as possible and can include the size of the homeless population and some demographic results that can be generated by the HPS-provided software. The community, after having invested the time and money in the count, will be anticipating the results. Be prepared for significant media attention around the count, particularly if it is your community’s first count. Communicating early helps keep them engaged and interested in finding solutions. This can be done through a media release or announcement, website update, and/or social media release. You can also consider hosting an open community meeting to talk about the results.

Communicating with people who are experiencing homelessness. Many people experiencing homelessness will have taken the time to answer the survey. While they may have received a small remuneration for their time, it is important to acknowledge their contribution and to convey how the PiT count will help them. Unfortunately, some may still be experiencing homelessness during future counts. The more they see the value of the count the more likely they will be to participate again. You may wish to work with shelters and other front-line services to determine the best communication strategy.

Planning for the next count

The local PiT Count Committee should meet again after the count to review what went well and what can be improved in future counts. You should consider conducting an anonymous survey with PiT count partners and volunteers. Feedback from volunteers, in particular, will provide information on how the count worked on the ground. How were the surveys implemented? Did survey teams use the standard wording of the questions? Did the locations or the sampling approach miss significant populations? Was the homeless population motivated to participate in the survey, or did they avoid the survey teams? The responses to these and other questions will help your PiT Count Committee to refine your PiT count implementation for future counts.

If possible, keep the Committee engaged between counts. In some communities that hold annual counts, planning for the next count begins as soon as the last count has ended. The Committee may wish to meet as milestones are hit (e.g., preliminary report released, final analyses done, post-count review completed, etc.).

Where to find more information?

Resources are available from communities that have conducted counts and from organizations that support PiT counts. Note that not all of the information in these resources will necessarily align with the HPS PiT count approach, but you may wish to consult these resources to help you with your local implementation:

The Community Workspace on Homelessness

The Workspace includes a PiT Counts page that links to a number of tools and resources that have been developed to support communities, including sample survey forms that can be adapted for use by your community:

  • Planning
  • Stakeholder engagement
  • Media engagement
  • Volunteer training
  • Volunteer management
  • Survey and Tally
  • Data management and analysis
  • Reporting

It also provides an opportunity to ask questions, share learnings, exchange resources and improve our collective understanding of Point-in-Time counts in Canada.

Community Workspace on Homelessness

Point-in-Time Count Toolkit

It includes optional templates, modules, case studies and additional guidance to support each step of PiT count implementation. It also includes specific guidance on engaging homeless youth and Indigenous communities in the planning and implementation of a count. It was developed by the Canadian Observatory on Homelessness.

PiT Count Implementation Toolkit

The Homeless Hub webinar on PiT counts

This webinar provides a brief introduction to PiT counts from researchers and communities with experience.

Homeless Hub Webinar: Using Point-in-Time Counts to Measure Progress Towards Ending Homelessness

Canadian PiT count reports

Several communities have been conducting periodic PiT counts in Canada. These reports are a useful reference for examples of local implementation of PiT counts and how the results are analyzed and reported back to the community. Many of these reports have been archived on the Community Workspace on Homelessness.

20,000 Homes Campaign

This site describes the campaign and what is involved in a Registry Week. If your community is interested participating in the campaign and considering implementing a joint PiT count and Registry Week, more resources can be found there. 20,000 Homes Campaign

Report a problem or mistake on this page
Please select all that apply:

Privacy statement

Thank you for your help!

You will not receive a reply. For enquiries, contact us.

Date modified: