eac data is a subset of the ff data. and so i think that's an important element. on the groove criticism about our nonresponse bias, that's was one of the first things i ininstituted and we had a nonresponse bias analysis. and what it determined was in fact o atistical waiting methodology countered the nonresponse bias that would otherwise be inherent. when we talked about moving around a lot, this is why weve towards more electronic access to ballots is so important. in my time in the military, i've been recalled active duty four times since 9/11. and even in those individual recalls, i've had two or three addresses. but i've always had robert.h..carey@navy.mil. i've had that. that's never going to change. go one step further. post the ballots online. you post the ballots online, anyone can access them at any time. not just when they get to their e-mail address, but whenever they want. hence our grant program to help the states post these ballots online. as far as the insulation voting assistance offices, we actually ha

we pay a lot of attention to the eac data. the eac's nvra report is very, very helpful in enabling us to figure out where we should indei indiai indiai indeed focus its examination. as you know, secretary 7 requires that voter registration be made availability among other places at state offices providing public assistance or disability services. one of the lawsuits involves rhode island and the other involves louisiana. in the louisiana case, i want to discuss the results because the results have indeed been remarkable. more voters were registered in the first month after the settlement than in the previous two-year reporting period. in the two-year reporting period before the lawsuit, 457 voter registration forms were submitted by the four affected rhode island social service agreement. in the second month, 1346 forms were received. in two months, a total of 874. this is a total of 4171 new registered voters in the months after the settlement. as opposed to the two-year increase in the previous two years. i want to commend a

so from our perspective it's not that the eac data is in contradiction to the fvap data. it's that the eac data is a subset of the fvap data. so i think that's an important element. on the gao criticism of the 2000 a survey of our non-response bias, that was when i first things i instituted for the 2010 survey. what it determined was that are statistical weighting methodology captures the non-response bias that would otherwise be inherent. when we talk of moving around a lot, this is why we truly believe that moving toward more electronic assess to ballot is so important. my time in the military, i have been recalled to active duty four times since 9/11, even those individual recalls either two or three addresses. but i've always had robert h. about kerry. go one step further, post the ballots online. if you post about online, anyone can access them at into. not just when he not just when he gets his e-mail address but whatever they want to hands our grant program in order to try to help states post these ballots online. as part of the -- we have stated that services need

and unlike the data that the eac, the election assistance commission puts together every year where they actually call state officials and get the numbers from them of how many military overseas voters requested absentee ballots, how many did you actually send out, how many did you get in, this is just a survey of people. and what did gao say about the report? they said fvap does not conduct a nonresponse bias analysis that omb says is a necessary step in determining whether survey findings are biased. not conducting such an analysis limits data reliability. now i don't want to get into a statistical debate here, because statistics is boring. but the point is anybody who does surveys knows that there is bias in who actually sends them back and how accurate the information is when it comes back. the other problem with the samples were that it didn't accurately reflect the overall military population. for example, 25% of the surveys that came back were completed by senior officers and their spouses, even though there are only about 12% of the overall military population. >> but in making t