The Election Reform Network, a two-year-old citizens watchdog group which monitored the polls in Montgomery County last election day (Nov., 06), released a pilot study of election day operations documenting vote count discrepancies in 19 out of the 25 election districts analyzed. In addition, the citizens group found a range of problems on the ground, including: one polling place opening 30 minutes late; long-time voters being wrongly told that they were transferred out of their election district and were unable to vote; and poll workers leaving their posts, resulting in only one person available to serve voters for a one and one-half hour period. The Network also documented machines not working for up to two hours, machines taken out of service and replaced hours later, machines not starting up, causing long lines, and voters using machines which were not turned on, resulting in lost votes. Problems with provisional ballots were uncovered when poll workers either failed to offer them to potential voters or did not assist voters in finding their correct polling place.

The vote count disparity uncovered by the Network revolves around the difference between the number of voters who signed in at the polls and the actual vote count from the machines. There is documented evidence of undercounts by Sequoia machines in other states. The Network's analysis of 25 election districts found that in 19 of them the machine count did not match the number of signatures in the poll books. The greatest difference in any one district was 14 more signatures than votes. The overwhelming likelihood is that this disparity reflects a loss of 14 votes, what election officials sometimes refer to as the "fleeing voter syndrome" connected to Montgomery County's Sequoia machines.  After making their selections, some voters forget to press the "cast ballot" button, which causes their votes to go unrecorded.  Since there is no prompt that alerts the voter to the problem, the voter leaves the booth under the mistaken impression that her votes will be counted. 

"Our findings are disturbing. After 10 years with these vote machines, there are still difficulties and poll workers are still not adequately trained on the machines and on general procedures, especially provisional ballots. The scary part is that this may just be the tip of the iceberg," noted Stephen Strahs, co-founder of the Network.

The Network was able to monitor directly about 15 election districts across the county and later analyzed vote totals in 25 districts. "If what we found is a microcosm of what is out there," said Strahs, "we have some very serious problems with our elections. Remember that we had one state legislative race in Norristown that was won by about 100 votes. Every vote counts may be a cliché, but it is not a meaningless one." Click here to read the report.