Frequently Asked Questions Regarding the BYC and SPP General Questions: Q: Who can I contact if I have questions about using the BYC or SPP, or if I have not been added to the distribution list or Dropbox folder? A: Contact Focus Strategies through kelly@focusstrategies.net. Q: What is HDX? A: HDX stands for "Homeless Data Exchange", the web based portal for reporting data to HUD, including your HIC, PIT, and AHAR. Q: Will we be expected to provide identifying information on clients? A: The only piece of Protected Identifying Information (PII) that the BYC requires is date of birth. However, this information will only be used to generate the aggregated dataset for the SPP by HMIS staff in your community. The BYC does not require that client level data be shared outside of the community. Q: What is the deadline for inputting data into the BYC? A: DHCD strongly encourages CoCs and local planning groups to take advantage of the BYC and SPP. DHCD wants to ensure that emergency crisis response systems across the commonwealth are right sized and allocating resources as efficiently and effectively as possible. THE BYC and SPP will assist with this. Questions Regarding HMIS Data: Q: How many years of data are we looking at? A: The analysis will include data from any client with an open record during a two year timeframe. For some programs with longer stays (mostly PSH), you may be pulling data for stays that extend beyond two years, but, in general, this means that you will need to pull two years of data. The data set should include any record that was open at least one day during the analysis year and the year prior to that. The analysis year should be the most recently completed calendar year. For example, if we are starting this process during 2014, then the analysis year will be 2013, and the data input into the BYC will be from calendar years 2012 and 2013. Q: How perfect do the data need to be? A: The data should already be "cleaned" and checked for any potential quality issues before being entered into the BYC. More detailed instructions for how to do this and what to look for as well as a Data Extraction Checklist are discussed in the User's Guide. Besides adjustments made to ensure that the dataset conforms to the requirements of the BYC, no other adjustments should be made for the sake of trying to improve program performance measures. If you believe there are projects for which 1 P a g e
the data available does not accurately represent your program, you can discuss these specific issue(s) with Focus Strategies staff on a case-by-case basis. Q: Can we use data we have gathered through September for use in the BYC? A: Unfortunately, no. The BYC is formulated to calculate performance measures based on calendar-year data. If data through September (as opposed to December) is used, then program measures such as utilization rates will appear artificially low. Q: Will we need to create a custom HMIS report to pull the data elements for tab 4? A: You will need to export raw (not aggregated) data from the HMIS system. There are detailed instructions in the User's Guide about which variables to export. Q: Our community records stays in the night-by-night method (as described in the 2014 HMIS Standards) to count the number of nights attended. This shows several one-night stays for an individual as opposed to one several night stay, for example. In order to get a meaningful representation of length of stay (LOS), should I code the exit dates to equal the date of entry plus the number of nights stayed before putting that information into the BYC? A: Many emergency shelters record data in the way that you have described, but you are correct in thinking that the entry and exit date(s) should be modified in order to appropriately capture length of stay for the purpose of the BYC and SPP. The User s Guide for the BYC has detailed instructions on how to address this issue on page 36. Q: I know it is important to complete Appendix C steps in order. Can I skip steps that don t apply? A: Yes, you can skip steps that do not apply. Q: Our HMIS leaves relationship to HH blank for singles. Should I replace these values with self? A: You should replace the blank values with self as long as you are sure that the cells are blank because they are singles, and not because the data is missing for that record. If you do have missing data as well, you should code the missing data as data not collected first, and then replace the single records as self. Q: I noticed instances where there are overlapping but not identical entry/exit records for households. (I.e. one record is 1/1/2012-1/2/2012 and one is 1/1/2012-1/5/2012.) Or, for example, I have a household with a record for 12/5/13-12/5/13 and 12/5/13-12-6/13/14 for the same shelter. Do I need to collapse these records? A: You are correct in identifying that there is a problem with the overlapping records that should ideally be resolved. Unfortunately, without knowing for sure which of the records is accurate, the tool does not have a basis for subjectively determining which of the records to keep. If possible, we recommend contacting the project directly to confirm which (if any) of the records should be kept. If you have the information about what the record(s) should say, you can correct the record(s) manually to reflect what actually happened at the project. If you do so, you should make sure to also change the record(s) in HMIS, and also note how many and which observations were altered, just in case that information ends up being needed in the future. If this issue is pervasive and/or not able to be fixed, it may be worth 2 P a g e
excluding the program from analysis altogether. Feel free to contact us with more specific information about the programs and/or projects if you are unsure of which route to take. Q: Should I recode blank to data not collected for the HMIS data? A: The BYC will recode all values listed as BLANK to Data not collected in tab 5, so as long as the entries say BLANK (as opposed to simply blank cells), then this is not an issue. If, however, you have blank cells, you will need to refer to Appendix C in the BYC User s Guide for directions on how to recode all of the blank cells to say BLANK. Questions Regarding the HIC: Q: Is Rapid Rehousing in the HIC a point-in-time count, or is it measured in beds per year? A: All of the capacity counts in the HIC are a point-in-time count of capacity. The BYC calculates an annual capacity based on the utilization rate and average length of stay in the project. Focus Strategies has suggestions for calculating the point-in-time capacity for rapid re-housing and other projects that are not tied to particular units or beds. Q: How is the fact that Rapid Rehousing estimates are measured at a point-in-time taken into account for the BYC? A: Capacities for all project types are measured at a point-in-time in the HIC. However, for projects that don't have fixed "beds e.g., rapid re-housing, scattered site leasing projects), an alternative approach is to calculate the average capacity over the most recent year. To do this, add the total lengths of stay (exit date minus entry date) for all single adults served within the year and all persons in family households served within the year and divide both totals by 365. This gives you the average number of "beds" at any given point in the year. For projects serving families, this number is further divided by the average household size to calculate the unit equivalent. The HIC capacity for these projects should be adjusted prior to transferring the data into the BYC. Q: If I know I need to exclude some data from HMIS in the BYC, does that mean that I should just delete that information in the HIC tab of the BYC? A: While you could simply transfer a modified HIC (excluding any projects you know you will not want to analyze) to the BYC, this is not recommended. The HIC tab in the BYC includes a column (AC) that allows the user to exclude a project that is in the HIC from further analysis in the BYC and SPP. This approach is preferred such that, should the project need to be included at a later point, the user will not need to reinput the HIC. Q: Which projects should I consider excluding from analysis in the BYC? A: You should plan to exclude programs that do not have enough HMIS data to make analysis relevant and/or consistent. This generally means new or under construction projects that do not have two years worth of data in HMIS, but could also include projects that are changing operations (for example, a TH project that has been reallocated to PH) such that the data available represents project operations that are no longer relevant, or projects that have closed since the HIC was prepared. If there are projects in 3 P a g e
which the data is poor, insufficient, or not representative of the project today, and, therefore, cannot be included in the BYC, they can later be added as part of the "Year-One Changes" in the SPP. Q: I would like to include rapid re-housing in the BYC if possible, but there are some challenges. I am planning to use calendar years 2012 and 2013 for the BYC, but our rapid re-housing projects started in July 2012. Would it be possible to use July 2012 July 2014 RRH HMIS data for the BYC? Or would we be better off just including RRH in year 1 of the SPP? A: The BYC was developed to use two calendar years of HMIS data. If you don t yet have two years of data, you should include the RRH projects in the first year of the SPP. Q: We have family units for TH and PSH. In the 2014 HIC, one of our TH programs was at 100% unit occupancy (10); however, due to the size of the families, only 26 of the 30 beds were utilized because families do not share units. How is this taken into account for the BYC and SPP? A: The utilization rate calculated by the HIC can be overridden in the BYC if there is reason to believe that the utilization rate on the night of the sheltered PIT/HIC does not represent the average daily utilization rate of a project. In the case of this particular project, it is possible to manually override the HIC utilization rate to a more accurate utilization rate of 100%. The HIC utilization rate is used only as a (potential) indicator of data quality. The BYC will calculate a utilization rate based on HMIS data (see page 12 of the BYC Technical Guide for details on how and why this is calculated) that will be used in modeling changes in the SPP. Any discrepancies between the reported utilization rate (HIC) and calculated utilization rate (HMIS) are reviewed as one of the data quality measures in the BYC. Q: I have my HIC uploaded with an error. Do I need to go back and start the template over since I have already started entering budget information? A: Yes, unfortunately, you will need to fix the HIC and reenter the information you have put in the budget tab before proceeding. Q: Do we need to address the differences in HMIS/PIT info by changing utilization rates in the HIC? A: No, you should not specifically change your utilization rate for the purpose of matching the utilization rates from HMIS data and that from the HIC. If you see a large difference between the two rates, that could be indicative of a count night that was out of the ordinary in some way, due to the nature of the program itself, or potentially due to data quality issues. You are able to override the utilization rate as calculated in the HIC in the BYC if there is reason to believe that it is not representative of the average daily utilization rate for the project. Questions Regarding the Budget Tab: Q: Since ESG and state funding have been combined into one grant this year, how do we account for that when entering budget information? (Note: This question is specific to CoCs in Virginia.) A: One purpose of gathering the sources of budget data is to determine which funds may be considered for reallocation in the homeless system of care. To the extent that the state funding is tied to ESG 4 P a g e
funding this year, the state funding is tied to the same programmatic requirements as ESG funds. Therefore, the state funding should be added to and represented as ESG funds in the BYC. Q: Why might some of my projects not show up in the drop-down list on the budget supplementary spreadsheet? A: This may happen if two (or more) of your programs have the same name. Make sure that all of the programs from the HIC have unique names. If this does not solve the problem, then contact us directly so that we may help you further. Q: Are we able to change the headings of the budget categories on the supplementary budget spreadsheet? A: The information compiled on the budget spreadsheet is meant to go into the BYC. Changing the budget spreadsheet by creating extra columns or changing the names of one or more of the columns might distort the link between the information in that spreadsheet and the information that is meant to go into the BYC as well as create the potential for inconsistencies with the way data are presented between communities. For these reasons, we do not recommend making any changes in the budgeting document. There is an area to put comments in the budget spreadsheet if you feel that some information would be missed otherwise. Q: For the budget info for the BYC, should we be asking providers for budget info from the same time period? (e.g., calendar year 2013, a fiscal year) or should we just ask them for their most recent annual budget (which could be for different time periods but would likely be more accurate than if they have to break it out)? A: Please ask them for their most recently completed budget year. So, for some providers, this may be a calendar year, some might be on a Federal fiscal year, etc. Even though this may not match exactly to the data time period, this generally is not an issue. Further detail is provided in Appendix B (starting on page 31) in the User's Guide for the BYC. Questions Regarding the SPP: Q: If we want to include DV projects in the SPP (or other projects for which there is no HMIS or data or insufficient HMIS data), is there anything we should collect now to prepare for that? For example, should we gather their budget information also? Is there anything else we should get from them? A: If there are projects for which you do not have HMIS data, but you would like to add as "new" projects in Year One of the SPP (including DV projects, projects not participating in HMIS, and new projects), you should include them in the budget collection process, asking for the same level of detail that is asked of HMIS projects. The SPP will require the user to input the capacity and budget for the "new" projects, and will calculate the performance outcomes for these "new" projects based on the average performance of similar projects. If you are able to get more precise, aggregate level performance outcomes for these projects, the SPP will allow you to override these calculated averages. The performance outcomes that the SPP will consider are: % of capacity for subpopulations (Vets, Chronic Homeless, TAY) 5 P a g e
Annual average utilization rate Annual average length of stay (all project types except PSH) Annual unit availability/turnover (PSH only) Annual average exit rate to permanent housing Annual average rate of returns to homelessness Annual average rate of entries from unsheltered, ES, TH/SH, institutions, housed locations See the Technical Guide to the BYC and SPP for more detailed definitions of these performance outcome measures. Q: How are the projects listed? A: The projects are listed in order of project type, and then ordered by their appearance in the HIC (which is generally in alphabetical order by organization). If you would like the projects to be in alphabetical order (after being sorted by project type), you can sort the exported HIC alphabetically by project before inputting the HIC into the BYC. Q: Is the sheltered number in the planning tab for year two of the SPP duplicated? A: The SPP uses a calculated, annualized population count of households in ES, SH, and TH from which to model changes. These sheltered counts are counts of stays in programs, and are not an unduplicated count of people or households (as one person or household could have multiple stays over the course of a year). The unsheltered annualized population count is calculated from the most recent unsheltered Point-in-Time (PIT) count, and is annualized using the method established by the Corporation for Supportive Housing in their March 2005 publication, Estimating the Need. Details on CSH s approach can be found here: http://www.csh.org/wp-content/uploads/2013/08/estimating-the-need.pdf. Q: What are the differences and similarities of the SPP and the Performance Improvement Calculator (PIC)? A: The PIC is another Excel based tool developed by Focus Strategies that uses community specific program to allow the user to see the immediate impact of changes to the system. Unlike the PIC, the SPP uses data at the project level and includes two dynamic concepts to better demonstrate the impacts of system change: 1. The user can make changes over time system changes can be made for up to five consecutive years, and 2. As performance and system changes are made, the impact on the size and composition of the homeless population and significant subpopulations are modeled. The PIC is also typically be populated more easily, as it uses aggregate numbers for program types, and does not require manipulated and collapsed project level data. Q: How do I change a project from singles to families (or vice versa)? A: In the Investments, Capacity, and Subpopulation section of the Year One Changes tab (and subsequent Year Two Changes, etc.), the SPP allows the user to change many of the characteristics of individual projects, including the total capacity and subpopulations served. Given the complexity of the logic behind the formulas controlling these changes, the user cannot change the project type from 6 P a g e
singles to families (or vice versa). The proportionate breakdown of household types served from the HIC is maintained in the other changes the user makes. If you wish to change the household type(s) served, please remove the existing project and re-enter it as a new project, with the correct breakdown of adult only beds and family units. The new project will be assigned the average performance outcomes of all projects of the same type, although the user can change these back to the actual project performance measures. 7 P a g e