Revisited: What could a Secretary of the State Do?

Four years ago, we posted a list of actions that a Secretary of the State could do on his or her own to improve the election process.  Lets revisit that list as the 2014 campaign begins in earnest.

**************

Much of what we hear from candidates is proposed cheerleading and leadership from the bully pulpit: the bills they would propose to the legislature, the leadership they would provide to improve the business climate, encourage voting, and voting integrity in Connecticut.  Don’t get me wrong these are useful and legitimate roles for the Secretary of the State.  However, there are items that the Secretary of the State can do on their own initiative to improve business registration, public access to information, and the election process.

When I talk to candidates for Secretary of the State, one of my main themes is what they could do on their own authority as Secretary of the State to improve election integrity and public confidence.

Here is my initial list of items from 2010:

  • Provide detailed, accurate, downloadable, election information and notices on the Secretary of the State’s web site, while increasing the integrity of the post-election audit. Create jobs, efficiency and election integrity in Connecticut by changing the way memory cards are programmed and tested
  • Improve documentation and training for election officials,  in substance and format
  • Provide written directives and responses to inquiries from election officials

We recognize that everything costs money, however, most of these items are relatively modest items that appear to be within the authority of the Secretary of the State.

Further Details

Provide detailed, accurate, downloadable, election information and notices on the Secretary of the State’s web site

In a PEW study the Connecticut site ranked 48th out of 50 states.  We could debate if we should be higher in the rankings, or instead work to emulate and surpass the top ranked states.

The process of accumulating voting results in Connecticut is an error-prone three step process of addition and transcription, from polling place, to town hall, to the Secretary of the State’s Office, and to the web.  Citizens have identified errors large and moderate – errors of a magnitude  which could change election results, the initiation of recanvasses, or ballot access. See <here> <here>

Without reliable, publicly posted results, post-election audits cannot be accomplished which inspire confidence and provide integrity.  A trusted audit requires selecting districts for audit against previously posted results.  Since we audit against optical scanner tapes, and the tape results are not posted, then we fail to meet that requirement.

What can be done?

  • Post copies of the original documents: All district and central count absented ballot Moderator’s’ Reports and copies of scanner tapes should be faxed to the Secretary of the State’s Office and posted on the SOTS web site. (We know this is easily possible since the SOTS web site has recently included images of all local ballots, and is capable of the quick addition of press releases)
  • Post detailed and summary data: The SOTS could use temporary employees or outsourcing to input and double check the input of all that data, then post it to the web site in human and downloadable formats.
  • >Side benefit: A free public audit: As a byproduct the public, candidates, and parties could check and audit the data at no cost to the state.  To do that today would involve visiting town halls across the state and performing all the calculations done today by hand – efficient auditing of selected districts is not possible because detailed data is not currently posted.
  • Consider using the Overseas Vote Foundation facility for empowering military and overseas voters, now in use by seven states.  This is an example of what could be done to empower all voters. (Added 4/4/2010)
  • Improve post-election audit integrity: Stop accepting reports showing wild discrepancies as extremely accurate and make all audit investigations public and transparent. (added 7/31/2010) <Extremely Accurate> <Audit Reports>

Since 2010: We have had a little progress and  some attempts in this area:

  • Starting in recent election, the actual election results data faxed from towns is posted to the Secretary’s web site, so it can be checked by citizens and a bit more transparency is thus provided.  Yet the results are not in detail, covering each election district and each voting method.
  • The Secretary initiated a well intentioned project to have all data from all districts input over the web and immediately reported. Unfortunately, the system was a failure and for at least two years, the Secretary’s Office remained tone-deaf to the complaints of election officials that the system and its requirement that each polling place moderator have an ID and input their own data was effectively unworkable.  In 2013 there was some indication that a more effective system of collecting data by spreadsheet, which could be input to officials charged with that function would be made available some time in the future.  Many towns already compile detailed data in spreadsheets, using rested officials assigned to data entry, to take and double check data from moderators who are often elderly and who all have just completed a demanding seventeen-hour-plus day.
  • Nothing significant has been accomplished in post-election audit integrity. In 2013 the Citizen Audit did a study of the audit drawing and found huge discrepancies on the list. We have long been a critic of the official audit reports created by Uconn, the only change is that Uconn has apparently stopped producing the reports since the November 2011 election. We do acknowledge an improvement this year in the official audit procedures, with several corrections and small but useful improvements suggested by the Citizen Audit and accepted by the Secretary’s Office.
  • There is on the SOTS Election web page information specifically for Military and Overseas Voters <view>, yet nothing like what is possible <see MN>.
  • Yet we applaud the SOTS and Registrars because Connecticut now does a good job of implementing the Military and Overseas Voters Empowerment Act (MOVE), resulting in Military voters achieving ballot return rates comparable with all absentee voters.

We would be remiss if we did not applaud, once again, the report on the Secretary’s Election Performance Task Force, while its promise has yet to be matched with an equal measure of progress.

Another area of irritation and loss of integrity in the post-election audits is that while the Secretary of the State’s Office is able to post ballots for every municipality, they are not provided with a reliable list of polling districts in those municipalities.  By requiring the faxing of district Moderator’s Reports and posting them, this uncertainty would be eliminated.

As we said above: In 2013 the Citizen Audit did a study of the audit drawing and found huge discrepancies on the list.

Another problem is public notification of audit dates, times and locations.  By procedure (unfortunately not by law), registrars are required to inform the Secretary of the States’s Office three days in advance of the local audits.  If these dates were posted by the SOTS Office within 24 hours of receipt, then the public, candidates, and parties would have much better access to actually observing audits.

Beyond audits, having centralized voting district locations and assisting voters in finding their polling place would help the public and partially relieve that burden from towns.

Create jobs, efficiency and election integrity in Connecticut by changing the way memory cards are programmed and tested

Currently, before each election, memory cards are programmed in Massachusetts by our distributor, LHS Associates.   The cards are shipped to local election officials for pre-election testing.  There are two problems:

  • We have no effective supervision over the process.  One of the risks is insider fraud or intimidation of those who program and ship memory cards.
  • The process is not perfect, and the memory cards are physically unreliable.  Extra effort is required when bad cards are discovered by the Registrars and new cards need to be ordered and shipped.  See <UConn Report.>

What can be done?

  • Perform the programming in Connecticut: In other states (outside of New England) large counties program their own cards and often perform programming at a fee for small counties.  We have paid for two machines which we can use to program the cards. They could be used by state employees or outsource the programming within Connecticut.
  • Independently test the cards nearby the programming: UConn has developed, at taxpayer expense, a program to eliminate many causes of error in the cards and easily detect bad cards with “junk” data.  Currently this program is underutilitzed in no non-random testing of cards selected by local officials after pre-election testing and after the election.  We can exploit this program to 100% pre-test the cards, enhancing integrity and reducing wasted effort by local officials when they discover “junk” cards.
  • Side Benefit:  Jobs: It might be small, yet every job moved back to Connecticut would be a benefit to the state.  Like large counties in other states, we might provide the service to other states in New England – we could compete with LHS for business and with the added advantage of the UConn testing program.  Perhaps we could expand beyond New England to service other states.

We also note a large cadre of very part-time election officials, many of whom served as lever mechanics and later as vendor trained optical scan technicians.  We expect that several of these same individuals could quickly train to meet the seasonal demand for programming and testing, and appreciate the opportunity for work and public service.  It would not take many.

Once again, slow if any progress. Over the last couple of years memory cards have continued to degrade, while Uconn continues to report that less and less cards are sent in from towns for testing. Nobody seems to be taking responsibility to see that the cards are sent as requested by the Secretary (or is that a fiction that the Uconn reports continue to maintain).

In the meantime, some new cards of a new design are tested in each election, yet there seems to have been no systematic effort of the State to hold the vendor accountable for selling such unreliable technology, and quite a price.

Improve documentation and training for election officials,  in substance and format

The current Secretary of the State and her staff have worked to improve documentation and training.  This work should continue and be taken to a new level.

What can be done?

  • Update and improve the value of manuals: The Moderator Manual, the Absentee Ballot Moderator, the Recanvass Manual and the Post-Election Audit Procedures were modified after the commitment to optical scanners.  They need to be updated and expanded based on experience.  They need to be rewritten and edited by professional technical writers to make them more effective as training and reference documents.  For instance,Improve and expand certification: The Secretary has begin efforts for Registrar Certification and Training to complement the current Moderator Certification and Training.  The Registrars job is much more involved than that of a Moderator, yet Registrars have no formal training and certification program (And currently there is no requiremen for Registrars to be certified Moderators or to attend training.  Moderators are supposed to be certified, yet that is not enforced – these are issues for the Legislature and an example of where the Secretary of the State could be an effective cheer leader)
    • The Absentee Ballot Moderator’s Manual still calls for multiple counting throughout the day, but only one count is necessary with optical scanners. The Post-Election Audit procedures should provide more details in several areas:  Counting incomplete bubbles, counting write-in votes, exactly what levels of differences should call for recounting and investigation, and help with accurate and efficient counting methods.
    • The Recanvass Manual should also cover details of counting incomplete bubbles, counting for voter’s intent, voter identifiable ballots, and the role of designated observers.
    • There may be value in following the examples of other jurisdictions in creating observer manuals for post-election audits and recanvasses.

Here we have seen some laudable progress and some questionable progress.

  • The Moderators Manual has been completely revised and is much more readable, has clear sections and checklists for each position and function.  Not as polished and some we have seen in other states, yet a vast improvement.
  • The Moderator Training has been revised with an online course which must be passed as a prerequisite for taking the in-person training. Unfortunately, this apparently costly outsourced training, suffers from some problems that tend to irritate officials required to take the training.  The training itself and the quizzes, apparently done by out-of-state consultants makes some glaring errors in Connecticut election law, and the quizzes score is based on incorrect answers to some questions. The training suffers from too long, too simple presentations on the basics of elections and on serving those with disabilities – one suspects that it is easy boiler-plate sold to every state. Further, the training is very choppy with short videos and many different/inconsistent navigation between videos, slides and quizzes – it all takes extra hours of time and frustration, especially for those not especially skilled in the web or loaded with extra patience.  A good idea, not well done.  We have heard that many Moderators are reluctant to take it a 2nd time (required every 2 years) and threaten to quit.  We will see. Once again, we have heard anecdotes of a tone-deaf response from Hartford.

 Provide written directives and responses to inquiries from election officials

One responsibility of the Secretary of the State’s Office is to advise election officials on proper procedures according to law, regulations, procedures, and directives.  We have heard registrars complain that the advice given depends on who one talks to, and on what day.  We have no way of determining if that is true or how prevalent the problem is.  The uncertainty and over-reliance on verbal communication should be eliminated.

We hear the same from committee political treasurers. In fact it is the subject of a proposed law before the Legislature.  The law would require the State Elections Enforcement Commission to follow-up with a written version of any verbal advice within ten days. <H.B. 5470> However, it is difficult to write law that distinguishes from a simple inquiry from one that provides a critical distinction that the caller must rely on to avoid error and avoid potential fine or jail time.

What can be done?

  • All directives should be in writing and publicly available. Like any laws, regulations, and procedures there is usually some ambiguity, unanticipated situations, new regulations can take years to be approved, and problems which must be overcome.  The Secretary of the State Office from time to time must issue directives to cover these situations.  Such directives should be in writing and posted publicly.
  • All substantial advice and rulings should be recorded in writing. The current Secretary of the State’s Office keeps track of all citizen inquires.  The Office should keep track of all official inquires, and summarize any significant election advice and rulings for future reference.
  • Economies of Scale: Perhaps the state would be best served by a common system, that tracked similar advice for the Secretary of the State’s Office and Elections Enforcement.  Perhaps we should use the same or similar laws, and the same system for all agencies that have a similar requirement.  The Environmental Protection Agency comes to mind.  What do other agencies do today?  What do other states do?  We have only one Freedom Of Information law for all agencies.  Perhaps we need a uniform law for agency directives, rulings, and advice?  There will always be a need to consider economies of scale vs. excess bureaucracy – but the third option is to search for the ways that increase economy of scale, increase democracy, while also streamlining bureaucracy.  Once again, the Secretary cannot change the law alone, but could cooperate with other agencies to produce efficiency and provide leadership in going beyond requirements of the law.

Perhaps we ask too much. We did support a bill in the Legislature that would have at least made the Secretary of the State’s directives enforceable, with the caveat that they be marked as such and centrally posted.

Perhaps soon, we will do a similar review of the progress on the report on the Elections Performance Task Force referenced above.

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailFacebooktwittergoogle_plusredditpinterestlinkedintumblrmail

Leave a Reply

You must be logged in to post a comment.