Appendix I: Objectives, Scope, and Methodology
To prepare case studies, we calculated the expenses associated with a nonrepresentative selection of three group sites and one commercial site in Mississippi. We used cost information issued by FEMA to calculate expenses associated with trailer purchase, site design and construction, and trailer installation. To identify the specific trailer barcodes located at each case study site, we searched several databases provided by FEMA, as well as data provided by the contractors for park address or occupant name matches. Because FEMA could not provide us with a definitive number of trailers at each site, for purposes of our analysis, we assumed a best case scenario for FEMA: that the parks were operating with a trailer on each available pad. Using the list of trailer barcodes we identified, we analyzed the invoices submitted by the MDC contractor responsible for each site, and the accompanying FEMA receiving reports to determine the number and type of services performed on each trailer and paid for by FEMA. The charges cover the period of June 2006 through January or February 2007, depending upon each contractor’s available data. We also added in the following costs as provided by FEMA: group site contractor costs for each site, including a portion of their phase-in cost, and monthly security costs and monthly lease costs, if applicable. The one-time and recurring costs were combined for each park, resulting in a total cost for each park. To provide a general lifecycle cost for a FEMA trailer, we estimated these totals through March 2009, which is the date FEMA stated the travel trailer rental assistance program will end. To determine the general costs for a FEMA trailer located on a private site, we identified trailers noted as “private” in the FEMA databases, and selected the first three for each MDC contractor. We then searched the contractor invoices, covering the period of June 2006 through January 2007 and recorded and totaled the charges for each barcode. The resulting totals were projected for 1 year, and used as an estimate of the annual costs for maintaining a trailer on a private site. We also projected the costs for these trailers through March 2009.
Our estimates are likely understated because did not have access to trailer maintenance and group site maintenance payments made to the original four contractors. We also could not calculate MD phase-in costs, nor could we calculate deactivation expenses because it is not certain which of the current MD contractors will be responsible for deactivating the trailers in 2009. In addition, we do not know how much it will cost to return the group sites to their original condition, as required by the terms of the group site lease. Results from nonprobability samples (case studies) cannot be used to make inferences about a population, because in a nonprobability sample, some elements of the population have no chance or an unknown chance of being selected as part of the sample. Our
GAO-08-106 Hurricane Katrina