Human Resources and Skills Development Canada
www.hrsdc.gc.ca
Common menu bar links
HIFIS Data Analysis Framework
Section A: Roles and Responsibilities of HIFIS Community Coordinators as Data Analysts
Provide an accurate picture of homelessness by identifying data quality issues
A key responsibility of the data analyst is to improve data quality by identifying and correcting inconsistencies in the HIFIS community database. The HIFIS Community Coordinators (CC) will provide consistent and continual training of staff involved in data collection and entry. The data analyst might rely on a data quality committee to develop a data quality plan, procedures or benchmarks for the community. The HIFIS CC and committee members can tap into HIFIS Initiative resources such as the data dictionary, training manual and HIFIS 3 data software.
Improve data quality by switching to HIFIS 3
The HIFIS CC, assisted by a data quality committee, will encourage HIFIS user compliance. The HIFIS CC will generate incentives to ensure timely data entry, help identify and correct data quality issues to facilitate the production of monthly or quarterly reports. The current focus of the HIFIS Initiative is on the processing of more HIFIS 3 data exports. By upgrading to HIFIS 3, most of the data quality issues (described in the following paragraph) will be resolved, thereby improving the quality of exports from sites and communities across Canada. HIFIS CCs need to train in the use of the HIFIS 3 Data Integrity Tools in order to introduce the system to the users in the community. Step-by-step instructions for HIFIS users to correct errors in their data banks will be included in the Data Analysis Training Toolbox.
Data quality issues inherent in HIFIS 1 and 2
Because of the critical importance of data quality, HIFIS 1 and 2 sites will have to be particularly vigilant with the following data quality issues inherent to these versions. Errors in data entry must be corrected at the starting point, as source data is the foundation of accurate and credible reporting. The data elements described constitute the essential elements of a client’s unique identifier: name, date of birth and gender. Precise data entry of this information will ensure the availability of critical information on the characteristics of service users and will help planners improve homelessness services systems.
- If one element is missing: For example, if the message “Date of birth is missing” occurs, the system cannot calculate the age. If the database cannot link the information on “Date of birth” and “Book-in,” it will treat a given client as two different individuals. Accurate client data entry avoids double counting. Double counting inflates the total count of homeless individuals.
- If one element is inaccurate: For example, if the message “Date of birth equals the book-in date” occurs, the system will produce an age of zero (0) and downwardly skew the average age. If the message “Date of birth occurs after book-in date” occurs, the system will produce a negative age calculation and downwardly skew the average age. If the message “Date of birth exceeds the limits of possibility” occurs, individuals living in shelters appear to be 150 to 350 years old, and the average age is upwardly skewed.
Because they are essential in calculating the length of stay, book-in and book-out dates have to be entered precisely.
- If the “Book-in date” is missing, it impedes the calculation of length of stay.
- If “Book-out date” is missing, it will appear that the client has never exited a shelter, and thus overstate length of stay and inflate the count of homeless individuals.
- Inaccurate “Book-in/book-out dates” skew the actual average or total number of days stayed in a shelter. For example, if the “Book-in date” is later than the exit day, the result is a negative length of stay, and if the length of stay exceeds 1000 days, it upwardly skews the average length of stay.
Analyze data adequately within the community context in a coordinated way
The Government of Canada assists communities in the implementation of HIFIS to improve the delivery of services to homeless clients and to increase understanding of their characteristics and needs at the local, regional and national levels. As full participants in the HIFIS Initiative, the HIFIS CCs will continue in their current efforts to coordinate shelters and service providers, to deliver HIFIS training, to report back to their communities on data collected and to foster sustainable partnerships. The HIFIS Community Coordinator’s analysis of data and provision of precise and consistent reports on shelter users, shelter use and patterns in shelter use will provide accurate information on the nature of homelessness, the patterns of service utilization and the effectiveness of HPS community plans.
Nature and scope of homelessness
To know more about the nature of homelessness, the HIFIS CC must analyze data on the shelter users within the broader context of the population, housing and income dynamics of a community. The HIFIS CC data analyst also needs to become familiar with community trends in homelessness and populations at risk of homelessness. Namely, the analyst needs to be mindful of methods used in the community to calculate the size of HPS target population groups: the hidden homeless population, those living on the street, individuals and families who are short-term or crisis sheltered and those who are supportive housed. In most communities, homelessness coalitions also gather and analyze data on households in core housing need (living in substandard housing). In some communities, HIFIS data on the use of food banks can also shed light on populations who are sheltered and those at risk of homelessness. Individuals or households at high risk often enter into homelessness because of life situations such as insufficient social assistance to meet a rent increase, receiving an eviction notice or being discharged from institutions without a permanent address.
Homelessness service utilization
To report on homelessness service utilization, the HIFIS CC has to analyze data on shelter use within the context of the sheltering and social housing capacity in the community. This analysis also must be done in conjunction with the community advisory boards involved in the HPS planning process and the local housing and homelessness coalitions. The HIFIS CC can use the wealth of data collected on owned/rented dwellings and shelter cost from various sources such as the 2006 Census. The community coalitions usually have access to data on social housing units, gathered from municipal planning, housing or social services department, provincial or territorial housing ministry or housing corporations and community housing corporations. Coalitions frequently consult coordinated waiting list data for social housing, supplemented with waiting-list data from housing providers in the private sector.
Homelessness service system effectiveness
In the pursuit of an effective homelessness service system, community planners, funding agencies and all levels of government arrive at key decisions based on the analysis of data in patterns of shelter use. Here is an example of how an HIFIS CC could compare data from shelters and transitional housing among communities, by placing it in the broader sheltering and social housing context of each community. In one city, length of stay in shelters and transition housing can be short, because access to better affordable housing exists, whereas stays in another place have to be extended because no other adequate and acceptable housing options exist. In analyzing and seeking explanations to the different trends in length of stay of various communities, the analyst can point to promising approaches in the development of sheltering options that exist in some places. This type of analysis of shelter use patterns will help planners and decision makers identify creative solutions to homelessness service system issues, such as low-income and affordable housing.
Work with service providers and shelters to comply with Data Sharing Protocol principles
Service providers and shelters may wish to develop their own reports; if they do, they will not receive additional support for them. They should however agree to follow the principles and procedures detailed in the Data Sharing Protocol (DSP) concerning: confidentiality of client-related information; informed consent of the client regarding personal information-sharing between organizations; and staff adherence to the procedures and structures set out in the DSP and regarding the accuracy and the security of the data collected. The tools available in HIFIS reporting and data analysis toolkits can assist them in developing their reports; however, if they encounter difficulties, the national HIFIS team will be unable to provide support beyond what is in the toolkits.
