State investigating dozens of disqualified ballots in Groton
Groton — The State Elections Enforcement Commission is investigating why 77 Election Day registration ballots for the November 2018 election in Groton were rejected and not counted, state officials say.
When the Secretary of the State's office was reviewing statewide data from the election, the staff noticed an anomaly in the number of Election Day registration ballots that were cast but not counted in Groton, spokesman Gabe Rosenberg said.
According to the Secretary of the State's website, statewide 19,121 Election Day registration ballots were issued, while 18,960 were counted, a difference of 161 ballots. Of those 161 ballots, 77 were in Groton.
The Secretary of the State's office called the town registrars of voters and they indicated that one of the poll workers assigned to Groton Election Day registration had told about 77 voters not to sign the ballot envelopes, Rosenberg said.
But signing the ballot envelopes is mandatory, so when it came time to count the ballots, those without signatures were necessarily rejected, he said. The issue was first reported in The Connecticut Post.
The Secretary of the State's office referred the matter, along with other Election Day issues in 17 other municipalities, to the State Elections Enforcement Commission, which voted at its Feb. 20 meeting to investigate the issues, according to spokesman Joshua Foley.
Groton Town Manager John Burt said the town is cooperating with the investigation. He said he isn't able to comment further until the investigation is completed.
The Registrars of Voters could not immediately be reached for comment.
State Rep. Christine Conley, D-Groton, said that while all races in Groton had a margin of victory of more than 77 votes, there have been elections in the past where the races were much tighter.
The 2018 state races for the 18th Senate District and the 40th and 41st House Districts, which include Groton, were each won by more than a thousand votes. Groton residents voted down the charter revision proposal 6,813 to 5,348, according to Election Day results.
But in 2006, the Democratic primary for the 41st House District was a tie and ended up being decided by a coin toss — making Elissa Wright the winner.
"I strongly believe that every person who casts their vote should have their ballot counted, and there have been elections in Groton and across the state where 77 ballots would have changed the outcome of a race," Conley said. "I support retraining of all people involved in Election Day registration in Groton so that this doesn't happen again."
Issues in other towns under investigation include redistricting errors and failures to properly staff Election Day Registration or regulate a polling location, according to a memorandum from the Secretary of the State's office to the Commission. Locally, in Preston, a voter "was told there was no accessible ballot marking system at the Election Day Registration site; that she was required to go to the polling place to actually cast a ballot," according to the memorandum.
Foley said that after any investigation into a matter, the commission will take a vote, resulting sometimes in a dismissal of the issue or a finding of a violation. "There's no set timeline for when it must be resolved by," he said. "Each matter is unique and depends on the facts."
Foley said once the referral is resolved, more information will be available, but the commission's policy is to not comment on an open investigation.
Under current law, every two years, moderators are required to take online training developed by the Secretary of the State and a two- to four-hour, in-person class with a moderator trainer, said Rosenberg, and both cover Election Day registration. Poll workers are trained on what they will be doing before every election, he said, and the Secretary of the State's office also recommends that poll workers go through the applicable parts of the online training. Registrars of voters complete a program to become certified, which also includes Election Day registration training.
Comment threads are monitored for 48 hours after publication and then closed.