To: Virginia State-Local Cooperative Program for Population Estimates members From: Michael A. Spar, Ph.D., Research Associate, Demographics & Workforce Section, Weldon Cooper Center, UVa Date: December 3, 2007 Re: REVIEW OF INPUT DATA FOR FINAL 2006 & PROVISIONAL 2007 POPULATION ESTIMATES The population estimates input data will be posted this month for your review. Data elements will be added to the web on a flow-basis, beginning with the data now on hand. Input data for your review includes: The number of building permits issued in your locality, plus an estimate of mobile homes since the last census.  A count of the number of students enrolled in public schools in grades 1-8. A count of the number of licensed drivers in your locality. A count of the number of births in your locality. A count of the combined total of personal and dependent exemptions claimed on Virginia state income tax returns. The total civilian group-quartered population residing in your locality. A list showing the names of all group-quartered facilities in our file. This list includes the names of all universities and colleges that provide student housing, local and regional jail facilities, state and federal prisons, and selected “other” facilities, such as Veteran’s Administration hospitals, youth detention centers, and the like. YOUR JOB: CHECKING THE INPUT DATA FOR ERRORS OR OMISSIONS The input data for the 2006 final and 2007 provisional estimates will be posted on our web site at: http://www3.ccps.virginia.edu/slcpe/InputData/inputdata3.html Please check these data against your local sources. If there are discrepancies between the data from your source and the information I have posted, please let me know as soon as possible. Send me an e-mail indicating the nature of the discrepancy and any supporting information about your data source. Regardless of whether you discover a discrepancy or not, send me an e-mail indicating that you have reviewed the information for your jurisdiction. If possible, I would like to hear from you by Friday, December 14, 2007. If I have heard nothing by that date I will assume that you have no problems with the data for your locality. In your review of the data, please bring to my attention only non-trivial discrepancies. I expect local records (especially those dealing with the number of building permits issued—see more about building permits below) to differ slightly from the state and federal data sources I use. Small differences in the data will not materially affect the population estimate for your jurisdiction, and it takes a considerable amount of staff time to investigate their origins. Send the results of your review to Dr. Michael A. Spar at mas6g@virginia.edu. ABOUT THE BUILDING PERMIT DATA Building permit data is the most problematic of our many data series. It has generated ten times the number of questions as any of the others. This is due to the fact that the numbers are fairly easy to check locally, and because there are often multiple local agencies that maintain these data. This creates the potential for slightly different numbers to crop up from different agencies, or even from different individuals within the same agency. These “different source” problems are compounded by the fact that we often do have incorrect permit numbers. In some cases the “bad” numbers are those reported by the jurisdiction. In other cases, the “bad” numbers come about because the jurisdiction failed to report, and we have had to make do with an estimate of their numbers. One fairly persistent problem leading to bad building permit numbers is the adoption of some new software systems. We have seen several instances where newly installed software seems incapable of distinguishing between a permit issued for a single-family unit and one issued for a 50-unit apartment building—to the software, both are counted as just one permit or unit. If this happens in a locality with lots of apartment construction, the result can be a severe under-estimate of building activity. Another problem has to do with the building reports that are sent by the locality to the Census Bureau. The Cooper Center’s main source of building permit data are the monthly and annual building permit tabulations issued by the Census Bureau’s Construction Division. The Construction Division receives monthly data reports (on form C-404 "Report of Building or Zoning Permits Issued and Local Public Construction," and hence, C404 reports, sometimes abbreviated by the Bureau as just C40 reports) from all large jurisdictions, and annual reports from smaller permit issuing places. At year end the Construction Division issues an annual report that combines information from all the monthly reports with the annual reports for the smaller places. As mentioned, this is our main source of building permit data. Each year around 30-40 localities fail to make complete reports to the Construction Division. In these cases the Cooper Center follows up by contacting the locality for the missing data. We succeed in getting the missing data in about half the cases. In the remainder of the cases—where there is an incomplete report and we have been unable to get the missing data—we use the Construction Division’s estimate of the missing data. We do not like using these missing value estimates (the Construction Division calls them “imputed data”) because they are often inaccurate—but we use imputed data in lieu of using nothing at all. What all this means to our SLCP data reviewers is this. If you discover a data discrepancy in the building permit series of fewer than 5 units, it’s not worth reporting to us. If your data series takes a sudden dip, see if new software could be to blame. If you do discover a major discrepancy, please check it out first with the Construction Division’s contact in your jurisdiction. So – first thing to do is to call the Construction Division. Their telephone number is (301) 763-5160. Find out the name of the person who is responsible for submitting the C-404 report from your locality to the Census Bureau. Then contact that person and see if you can resolve the numbers with them. If, after doing that, you still feel there is a problem with the numbers we are using, contact us and let us know about it.