ABCDEFGHIJKLMN
1
2
3
Data Balancing
4
5
6
7
8
9
Overview
10
11
We collect real-time data to solve the real-time issue of homelessness. To make sure every known person experiencing homelessness in your community can be supported and moved into housing, you need to make sure they are accounted for in your system. To make sure that everyone is accounted for, the data you collect should balance each month. This means that from one month to the next, all people entering, leaving, and remaining in your system are accounted for.

Here is a quick example:

• January 31st: 10 people are experiencing homelessness.
• February 2nd: 3 people are housed. There are now 7 people experiencing homelessness.
• February 11th: 2 people are added to the By-Name List. There are now 9 people experiencing homelessness.
• February 28th: There are 9 people experiencing homelessness.
• From January to February, there was a difference of 1 person (3 people left the system, 2 people entered the system). Therefore, for the data to balance, there should be one less person actively experiencing homelessness in February (9 people) than there was in January (10 people).

The next tab provides information on data points, calculations, and examples.
12
13
14
Reporting
15
16
Everyone who enters, remains, and leaves your homelessness system must be accounted for so that the data balances. We know that data entry and updates are not a perfect process. There will be times when data is entered incorrectly, or an update is late. Regardless of the scenario, the data should balance from one month to the next.

If you discover that you need to update an entry to be more accurate, and the update occurred at least one month prior to the month you are currently collecting data for, then you must account for this change in your data report. For example, if you found out that Jane was housed on November 5th and today is January 2nd, you cannot just count Jane as no longer experiencing homelessness. You would need to record Jane as a Move-In for January, since that this the month in which you discovered this information.

The way in which you enter data (new entries, updates, and corrections) will depend on your report and how it extracts data. For example, there are some HIFIS reports wherein you can backdate correct data the month in which the event occurred and the report will still record the data so that it balances. Using the example above, this means that you would enter Jane’s move-in date November 5th, and the report would still include her as a January move-in.

Conversely, there are reports in Excel workbooks wherein you would have to record the date in which you discovered the information (not the date the event occurred) for the data to balance. Using the same example once more, you would enter Jane’s move-in date as January 2nd so that the report can include her as a January move-in.

When it comes to choosing or building a report to extract the information for the data points, be sure that however it is written, it will balance data from month to month. Here are some tips for doing this:

• Review the definitions and calculations of the data points;
• Record how data should be entered in your report to make it balance (i.e. are you able to backdate information or do you have to record the information the day in which you discover it); and
• Test out your report with fake data and multiple scenarios to determine if it balances before using it system wide.

For more information on data balancing, you can view the CAEH webinar on System Level Data and Data Reliability. Here is a link to the recording and slides.
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
Data Balancing Quiz
42
43
Test your data point knowledge with this quiz.
44
45
46
Definitions
47
48
Reporting Month
49
50
Definition: The month and year the data is from.
51
52
Example: Data is from March 2019, captured in April 2019 and it is now June 2019. The reporting month is March 2019.
53
54
55
Active Homeless
56
57
Definition: Includes the total number of people experiencing homelessness at the end of the reporting month.
58
59
Calculation: Any change in this number from the previous month should be accounted for in either inflow (newly identified, aged in, returned from inactive, returned from housing) or outflow (move-ins, moved to inactive) in the reporting month.
60
61
62
Move-Ins
63
64
Definition: Includes the total number of people who moved in to permanent housing during the reporting month AND remain housed at the end of the reporting month.
65
66
Calculation: In addition to meeting the definition of a move-in, a person must be included in the number of active homeless the previous month OR included in inflow for the reporting month. If a person move's into housing, but loses that housing before the end of the month, do not count them as a move-in.
67
68
Example: #1: Archie was included as active homeless in January 2019. In February 2019, he moved into a home and remained housed at the end of the month. Archie was included as a 'move-in' for the reporting month of February 2019.

#2: Archie was newly identified at the start of January 2019, moved into a home around the middle of the month, and remained housed at the end of the month. Archie was included as 'newly identified' and a 'move-in' at the end of the reporting month of January 2019.

#3: Archie was newly identified at the start of start of January 2019, moved in a home around the middle of the month, and lost his housing before the end of the month. Archie was included as 'newly identified' and 'active homeless' at the end of the reporting month of January 2019.
69
70
71
Moved to Inactive
72
73
Definition: Includes the total number of people who were actively homeless, but became inactive in the local system during the reporting month and remained inactive at the end of the reporting month. A person may become inactive for the following reasons: left community, deceased, no longer meets criteria, lost contact, institutional stay (e.g., hospitals, correctional facilities, or treatment facilities) for longer than the local inactivity policy of not more than 90 days. The community should have an inactivity policy that clarifies the reasons, as well as the length of time required, before a person is considered inactive.
74
75
Calculation: In addition to meeting the definition of inactive, a person must be included in the number of active homeless the previous month OR included in inflow for the reporting month.
76
77
Example: #1: Riverdale has an inactivity policy that states a person must be inactive for at least 90 days before they are 'moved to inactive.' Betty was added as Newly Identified in November 2019. She remained actively homeless until January 2020, when she became incarcerated with no plans for immediate release. She remained on the list as 'active homeless' until the end of April 2020, when she was still incarcerated and was therefore 'moved to inactive.' In November 2019, Betty's was included as 'newly identified' and 'active homeless.' From December 2019 - March 2020, she was included in 'active homeless.' In April 2020, she was included in 'moved to inactive.'
78
79
80
Newly Identified
81
82
Definition: Includes all people who were added to the By-Name List for the first time during the reporting month.
83
84
Calculation: This is a cumulative total of all people added to the BNL for the first time during the reporting month, regardless of their status at the end of the reporting month.
85
86
Example: #1: Jughead became homeless for the first time in April 2019, but didn't become known to the local system until he came to an emergency shelter in May 2019. Jughead was added as 'newly identified' and 'active homeless' in May 2019.

#2: Jughead was homeless in Greendale for 2 years before moving to Riverdale in June 2020. He immediately became known to the local system. Jughead was included in 'active homeless' AND 'newly identified' in June 2020 in Riverdale.
87
88
89
Aged In
90
91
Definition: This data point is only for communities tracking chronic data. If a community is not tracking chronic data, this data point can be left blank. 'Aged In' refers to individuals who are 'active homeless' on the BNL and have met the definition of 'chronic' for the first time during the reporting month.
92
93
Calculation: This is a cumulative total of all people who were 'active homeless' on the BNL and met the definition of 'chronic' for the first time during the reporting month.
94
95
Example: #1: Veronica was included in 'active homeless' in March 2019 on Riverdale's By-Name List for all person's experiencing homelessness. In April 2019, she met the definition of chronic and was included in 'aged in' on the Riverdale's By-Name List for chronic homelessness. Veronica would be included in 'active homeless' AND 'newly identified' on the chronic By-Name List for the reporting month of April 2019.
96
97
98
Returned from Inactive
99
100
Definition: Includes all people who were previously reported as inactive that have returned to active homelessness.