DOD, FVAP accused of cooking the military voting books

At the link, a powerful article on how FVAP, the agency responsible within DOD for military voters, is manipulating military voting data to serve its purposes. As this article revealed, DOD spent over $2.3 million for this survey – almost $1 per military voter and is now attempting to permanently wrestle away the reporting of data away from EAC to the Department of Defense. It is noteworthy that the head of FVAP, in his oral testimony to Congress (at minute 14:12 and 14:44), specifically argues that DOD should be the sole collection authority for election data and that states should only be required to report these voting statistics to DOD. However, his written testimony failed to make these same assertions so the question remains. As noted here, it is unusual that DOD would go after another executive branch agency such as the EAC, especially with so many House and Senate Democrats are trying save it. With all the pending cuts to DOD warfighters, this appears to be a costly and ill-conceived effort to gain control of the collection and reporting of military voting data.

Unlike the EAC data, which relies on actual absentee ballots sent and received by state officials, the FVAP reaches its conclusions based on survey responses from 15,037 military members and their spouses, although it sent out more than 123,376 surveys. The low response rate is the first red flag about the credibility of the FVAP’s report — a fact that the Government Accountability Office has repeatedly emphasized to the FVAP. In 2010, the GAO criticized the FVAP’s reporting methodology because “FVAP does not conduct a non-response bias analysis” that the Office of Management and Budget has said is “a necessary step in determining whether survey findings are biased. Not conducting such an analysis limits data reliability.” The report was aptly named, “DOD Can Strengthen Evaluation of Its Absentee Voting Assistance Program.” Yes, it can (but it hasn’t).

Low survey response rates are not the only problem. The sample that responded does not accurately reflect the overall military population. For example, approximately 25 percent of the surveys were completed by senior officers and their spouses, even though they represent only 12 percent of the actual military population, thus giving a biased and skewed view of the voting experiences of the military. Another significant problem appears to be the low response rates from combat troops, who have the greatest difficulties in voting. In fact, the response rate from members of the Navy, Air Force and Coast Guard is nearly 70 percent higher than the response rate from those serving in the Army and Marine Corps. Yet the Army and Marines make up over 50 percent of the armed forces of the United States.

Obviously, whether a military member is in the field or on the front lines in Afghanistan, as compared to sitting at a desk in the Pentagon, can have a significant effect on the ability to receive, vote and mail back an absentee ballot. Yet the FVAP report doesn’t appear to account for this critical difference, as well as many others.