Copy of Public Comments Received

." .
,

"

....,

"Sue Sautermeister "

.:sue_sautermeister (gyahoo.c
om~
08/06/2006 12:03 PM

To Lotero(geae.gov
ee
bee
Subject response to voter questionnaire

Ms La tero

Bryan Whitener sent out the draft of the survey for
the 2006 election to the members of the EAC advisory

board
my comments

1 . easy to read
2. don't spli t sections or questions between two
pages (it occurs a few times only)
3. page 14, #43, in our situation the numbers of poll
workers varies depending upon the number of registered
voters in a precinct, but this question doesn't allow
for a response where the numbers of poll workers vary
among the precincts
4. page 14, #46, do you want multiple responses here?
we have several who are involved
5. page 16, #53, not enough room to respond if you
want everyone who had input to list name & info since
not just one person will necessarily be completing the

form
6. lastly, i hope this goes out early enough so that
those who will be completing the form know what will
be asked of them so they can compile the info as they
go along

Sue

"Melanie Abplanalp"

.' ,.

.:melaniea (gwashco .

"

To lotero(geae.gov
state .ut. u

s~
08/07/200609:59 AM

,Qt~,'"

ee

bee
Subject washington county survey request

Laiza,
Thanks for letting us get a heads up on the new stats that we should

be tracking.

I noticed that as I was doing this, most of the questions were fairly
relevant. I know that
some states are a little ahead of us, but I think the information can be

useful in the future.

I did have some concerns about the time it would take, unless we could
get Vista to track
this info for us. I had to gather the info from many sources, and it
took longer than I expected.

Here are some specifics,
We do not have information to track yet by Mail, or in Clerk's office,
DMV, or other registrations with dates.
We cannot identify dupilcates, address changes or rej ected registrations.
I thought tracking the deleted list was interesting, but I can see why
they would like to have that info.
Tracking the amounts of confirmation cards would be a great help to me,
it would be good to see how many
go out, and how many come back! Track expenses better.
Election totals seemed to be over kill, but if we could track the info
in 1 or 2 places I would be willing to do
the in depth detail that they would like.
If you have questions call me!
Melanie Abplanalp

435-652-5891

melanieagwashco. state. ut. us

..
a,.,'"

"Jim Harmening"

.:jim(gbitsmail.com~
08/07/2006 10:09 PM
Please respond to

jim(gbitsmail.eom

To lotero(geae.gov
ee

bee
Subject Public Comments on Election Administration Proposed

Survey

Dear Ms. Otero,

I hope this helps, it is a great Survey and I hope the Election Jurisdictions actually collect the data you are
asking for!

Thanks
Jim Harmening
7805 Palm Drive
Orland Park, IL 60462

708-288-3314

~

public comments EAC Survey.doc

Election Administration Voting Survey Draft
QlO. I would suggest splitting up the Tax Assessor and the Tax Collector. In Ilinois and
many other jurisdictions throughout the country, they are two different offices.
Q 1 1. In Ilinois there are so many townships, cities and vilages, that you may get a list

several thousand. Likewise there are many public service agency locations.
I am not sure what the goal is for Ql 1 or what information is to be collected? Name,
address, city, state, zipcode, person in charge, phone number? It doesn't appear that the
question is detailed enough about what information is to be collected and that it tries to
collect too much information (at least for a large state with many local offices.)
Q13 - there are kind of

redundant answers e-mail/electronic

andVPNare

similar.

Is

Disk I tape the same? How about CD? Some people may use a USB drive? Either have
a category of
Electronic with sub sections under it
o Electronic

o Tape

o Diskette

o E-mail

OUSB
OFTP

o Website Form

OWebsite File Upload

OVPN

Under Other you may want to put a line for them to write it in.

Great survey and good luck with the responses!

Jim Harmening

¡

a..

To lotero(geae.gov

"Silrum, Jim"
.:jsilrum (gnd .gov~

ee "Jaeger, AI A." .:ajaeger(gnd.gov;:, "Oliver, Lee Ann M."

08/09/2006 05:28 PM

-:Ioliver(gnd.gov;:, "Odenbaeh, Danette"
-:dodenbae(g ndaeo. erg;:
bee
Subject 2006 Election Administration and Voting Survey

Dear Laiza,

as I was preparing a note to send to North Dakota's county election offcials to
encourage them to review the survey that is under review prior to the General Election in
November, I noticed something in the Absentee Ballots section that made me think i should
send a comment upon which i hope you too will ponder before the survey is finalized. i don't
think we talked about this when we were together in April for the preliminary survey meeting,
Today

but if we did i apologize for not paying attention.
Question 36 asks for the number of absentee ballots (excluding FWABs) requested and the

number not counted.
1. FW AB is spelled out in question 38, but since this is the first place it is used, this is the
place it should be spelled out.
2. i understand the need for collecting both of these items of information, but the two
numbers are quite disjointed from each other unless another number is asked for
between these two.
3. The number would be the number of ballots returned.
i would suggest the following order and wording of the questions if I may be so bold:
36. Total number of statewide and by county/local jurisdiction of advanced ballots
transmitted to military and overseas citizens for the November 7, 2006, Federal General
Election: (keep the three subcategories with the total

line)

37. Total number of statewide and by county/local jurisdiction of absentee ballots
requested and returned (do not include FWAB) for the November 7,2006, Federal
General Election: (keep the four subcategories with the totals line - one column for
requested and one column for returned)
38. This one could stay just as it is
39. If you added a total
line to this question, you would know the number of absentee ballots
not counted or rejected that were received from domestic civilians
40. If you added a total
line to this question, you would know the number of absentee ballots
not counted or rejected that were received from military and overseas citizens
The survey asks for the number of advanced ballots transmitted, but it doesn't ask how many of
those were returned; rejected or not counted and why unless we are supposed to assume that
these ballots should be tabulated with all the other absentee ballots based on the definition of "

Absentee voting" at the beginning of this section. Perhaps this could be explained with the
questions relating to the return of absentee ballots or these numbers could be separated as you
have separated them for the two questions differentiating their delivery to the voter.

Please feel free to call me if you have any questions for me. Maybe i have been as clear as
mud to you or maybe I have missed the obvious that would make the current order clearer to
me.

Thanks for letting me comment,
P.S. I know you have talked in the past that the 2006 survey would be one in which the
numbers would be entered by each state through a web application. Do you know whether this
will be a manual data entry method by those of us in the state election offces? If it is, I am
concerned that there will be a high potential for "fat finger" errors to the submitted data. Or, can
you describe the survey tool that is being planned for use?

Thanks,

Jím Sí(rum
Deputy Secretary of State
State of North Dakota
600 East Boulevard Avenue

Bismarck ND 58505-0500
701-328-3660 Voice
701-328-2992 Fax

www.nd.Qov/sos
isilrum (§ nd.qov

.'.'.
"

To lotero(geae.gov

"Pat Hollam "
.:phollam (gco.okaloosa .fl.us~

08/25/2006 05:44 PM

ee

bee

Qt-"

Subject DRAFT 2006 Election Administration and Voting Survey

Laiza,
I found the Draft 2006 Survey to be comprehensive and well done, and just have one subject to be
addressed, if deemed warranted.
After Question #40, before ending the section on UOCA V A absentee ballots, would it be possible to add a
question, or questions, concerning the absentee requests that were honored based on the UOCAVA
voters' 2004 requests (per HAVA's mandated two-general-election-cycle requests)? It might be a
multi-part question, to include these issues:

. How many absentee ballots were issued in 2006 to UOCAVA voters who requested a ballot in
2004, but from whom the election official had no contact since the 2004 request?

. Was any additional contact made to those voters prior to the 2006 elections?

. How many of those 2006 absentee ballots (from the 2004 request) came back "Return as
Undeliverable" from the Postal Service?
. How many came back voted?

. How many never came back?

I am planning on including this subject in my presentation on Sept 21 in St. Louis, which is before the
close of the Public Comment period, so I will be able to further explain the value of this information.

Pat Hollarn, Supervisor of Elections

1804 Lewis Turner Blvd, Ste 404
Fort Walton Beach FL 32547-1285
Tel: 850-651-7272 Fax: 850-651-7275

Email: phollarn(gco.okaloosa.fl.us
Web site: www.govote-okaloosa.com

To lotero(geae.gov

"Jo-Anne Chasnow"
-:jae1 OOO(gjuno.eom~

ee

09/28/2006 12:50 PM

bee
Subject Comments on Draft Voting Survey

U.S. Election Assistance Commission
1225 New York Avenue, NW., Suite 1100
Washington, DC 20005.
A TTN: Ms. Laiza N. Otero

(Iotero (§ eac. QOV

Dear Ms. Otero,

Attached please find comments submitted by Project Vote responding to the "Draft
2006 Election Administration & Voting Survey" of the U.S. Election Assistance

Commission.
Please confirm receipt, and please be certain to notify me if you have any problems

opening up the attachment.
Many thanks.

Sincerely,

Jo-Anne Chasnow
Policy Director, Election Administration Program
Project Vote
201-863-3412, jac1 OOO(§ juno.com

~

www.proiectvote.org ProiectVote Comments on EAC 2006 Survey.doc

Project Vote
2101 S. Main Street
Little Rock, Arkansas 72206

501.376.6451 phone 501.376.3952 fax

September 27, 2006

U.S. Election Assistance Commission,
1225 New

York

Avenue, NW., Suite 1100

Washington, DC 20005.

ATTN: Ms. Laiza N. Otero

(Iotero~eac.qov)
Dear Ms. Otero:

We are writing today to provide comments on the "Draft 2006 Election Administration &

Voting Survey". Thank you for this opportunity.

i. On Page 1, the survey provides a "definition" of a duplicate registration.
However, a vital piece of the definition is missing: same birth date. Without
matching birth date, it is impossible to definitively identify duplicate
registrations. We encourage you to add the words "same birth date".
II. We believe additional questions should be included as a subset of question

#12 if the respondent answers "yes" in either of the two "yes" boxes. The first
question we recommend that you include is "If yes, how often?" and the
second is "When was the last training offered?" It is vitally important to know
some details about training of public employees who are offering voter

registration.
III. We believe you should insert a question following question #16. Our

recommended language is: "If the voter registration application is not

successfully matched against any existing record, is the next step to assign
the registrant with a unique identifying number? If not, what IS the next

step?"

IV. In question #18, there is a "yes" box, followed by "(If yes, please identify
which states)." Please consider adding to the words within the parenthesis the
following phrase: "and the elements that must 'match' to establish a
duplicate". We further recommend a follow-up question: "If duplicates are

found, how are those duplicates handled?"

Project Vote

Comments on Draft Election Administration & Voting Surey
September 29,2006
V. In question #19, please add the word "all" between the word "Are" and the

word "applicants", so that it reads "Are all applicants whose applications are
rejected notified of the rejection and the reason for the rejection?"

Vi. We encourage you to reinsert a question that was asked in the prior survey,
to be added after question #21: "Identify any restrictions on the acceptance
of voter registration applications such as paper weight, fax, electronic, mail
(for non-NVRA states), batch, third-party handling (changed from delivery to
handling), etc." These sorts of restrictions have placed significant burden on
many of the civic organizations engaged in voter registration across the
country. A question soliciting information from the states would be very
informative and encourage transparency.

ViI. We also encourage you to add another additional question here, related to
acceptance of the national application: "Do all election jurisdictions in your
state accept a completed National Mail-In application as a complete voter
registration application?"
VIII. It would appear that two sections of the previous survey have been entirely

eliminated. One section is "Voter Registration Drives". Voter registration
drives are a centerpiece of participatory democracy. But little by little the
ability for these drives to function is being eroded. Specifically, we encourage
including the following questions:
a. "Does your State manage the number of applications given to the outside
groups and the number and timing of the return of those forms by these nongovernmental registration organizations If so, how?"
b. "Does your State allow organizations to print additional copies of the state
voter registration applications by photocopying a blank application?"

c. "Does your State use serial numbers or other identifying codes on voter
registration applications given to non-governmental organizations or
individuals in order to be able to identify who handled completed applications
that are received by State and local officials?"

IX. The other section that has been entirely eliminated is "Public Information".
Confirmation of whether applicants have actually been placed onto the voter
rolls, especially when a state does not have a public access
portal, is critical,
especially for voter registration organizations. It is vitally important to know if
voter registration cards are available for public inspection; and with newly
created statewide databases, to know if voter registration files are available
for sale - and if so how much they cost.

Project Vote

Comments on Draft Election Administration & Voting Surey
September 29,2006
We then recommend the following questions, included in the previous version
of the survey, be included here as well:
a. "Are voter registration records public information?"

b. "Are voter registration files available for sale to the public? If so, how

much do they cost?"

X. Finally, we encourage you to add a section specifically for the states to report
the breadth of voting machine problems experienced in 2006. This will
respond to the rash of problems that jurisdictions around the country have
experienced with elections in the 2006 primaries and most likely will
experience in the 2006 general elections.
Thank you for the opportunity to share these comments.

Sincerely,

Jo-Anne Chasnow
Policy Director, Election Administration Program
Project Vote

"Michael Cragun"

To lotero(geae.gov

.:meragun(gutah.gov~

ee

09/28/2006 10:53 PM

bee
Subject DRAFT 2006 Election Administration and Voting Survey

Laiza,
The one thing I have worried about since you shared this draft with us
at the Standards Board Meeting in May is that election officials may not
have adequate data to answer many of the questions that ask for
information between Election Day 2004 through Election Day 2006. It
seems to me that if you want data collected for a two year period, you
need to tell the election officials about the expectation before the two
year period begins.

Thanks,
Michael

Michael Cragun, Deputy Director
Office of the Lieutenant Governor
State of Utah
PO Box 142325
Salt Lake City, Utah 84114-2325
(801) 538-1041
(801) 538-1133 fax
mcragun0utah. gov

"Carol Paquette"
.:paquear(gearthlin k. net~

09/29/2006 03:03 AM
Please respond to

paquear(gearthlink. net

To lotero(geae.gov
ee
bee
Subject comments on 2006 election administration survey

Laiza -

This survey has really improved. Someone has been working hard on this. I hope the attached comments are understandable and helpfuL. I was
sorr to see that there are no questions on types of
voting systems in use and identification of any problems, but maybe EAC is collecting that
data from some other source, You can reach me at 703.532.0524 if

awaiting the results!

Carol Paquette

paq ucaríaearthlink, n et

EarthLink Revolves Around You,

2006 Election Survey comments.

doc

you have questions, Good luck with the survey! There are many of

us eagerly

9/2812006

COMMENTS ON DRAFT OF 2006 ELECTION
ADMINISTRATION AND VOTING SURVEY

1. Recommend that "Voter Registration Applications" heading be changed to
"Registered Voters", and that new heading "Voter Registration Application
Processing" be inserted after question 2. Rationale: 1st 2 questions only refer to

voters; other questions deal with application processing. Adjusting the
headers helps the respondent navigate the instrument.
2. Recommend to move "Duplicate registration application..." description to
question 7 where it is referred to. Rationale: Use of ease for respondent by
providing explanatory material where it is referenced.
3. In question 1, would it be clearer to use "at the conclusion of registration
processing" or similar wording in place of "close of registration?" Rationale: Not
all registrations are processed by the close of the registration period. I assume that
the information sought is the final number of registered voters. This revised
wording would also encompass Election Day registrations.
numbers of

4. In question 2, should November 2, 2004 be added to get comparative data?
5. In question 6, what about applications generated by voter registration drives run

by civic organizations and political parties? These constituted a big processing
issue in 2004.
6. In question 7, it would be interesting to know the reason, for applications being

rejected (in addition to being a duplicate, as noted).
7. In question 8, does "new, valid registrations processed' mean entered in to the

voter database? Also, what is meant by "re-registrations across jurisdictional
lines"? Is this a change of address?
8. Recommend that the words "conducting voter registration" in question 9 be

replaced with "responsible for registering voters," as used in question i O.
Rationale: This just makes clear that both questions are talking about the same
function.
9. Capitalize Armed Forces in question i 1 to be consistent with usage in question 6
(or vice versa!).
10. In question 12, it would be interesting to know which agencies are provided
training for the "SOME, BUT NOT ALL" response.
1 1. In question 13, recommend that "offcial responsible for voter registration" be
reworded to something like "election offcial responsible for approving voter
registration applications." Rationale: This more clearly differentiates the election
folks and their role ffom the other agencies involved. Also recommend that
responses be divided into two categories: how delivered and what media used. For
example, a tape could be delivered by courier, by hand (what's the difference
between these two?), U.S. mail, inter-office mail, etc. Electronic delivery could be
by fax, email, VPN. Ftp is missing as an important electronic means. Power
Profile System needs to be explained. What are you trying to find out by this
'question? How quickly updates can be processed or how efficiently the process
takes place? Not clear what one would learn from this set of
responses. For

example, a tape delivered manually that is compatibly formatted to run directly
against the voter database would be much more efficient to process than an
automated ftp that requires data format conversion.

12. In question 14, recommend that "forms" be replaced with "applications."
Rationale: Consistent terminology. What is meant by "verifies"? Does it mean the
offcial with authority to approve application? What is meant by "processes"? Is
this merely data entry; does it include validation of any sort? Some local election
offices have administrative personnel enter applications into electronic pending
fie via keystroking the data or scanning the form, and then a senior official
reviews and approves for upload to voter registration fie.
13. In question 15, recommend that "processed voter registration form" be replaced
with "accepted (or approved) voter registration application." Rationale: Keeps
terminology consistent and clear.
14. In question 16, recommend that parenthetical explanation be reworded to read:
This refers to the process of verifying the information provided in the voter
registration application. It does not refer to the process of verifying voter
eligibility at the polling place. Rationale: Clarifies the language. Also, what does
"Link" mean in the responses? The word "records" should be added to the end of
the second and fifth responses. Social Security Administration should be
capitalized. What is a "disposition notice"?
15. Recommend that wording in question 17 be changed to read: "What data fields
are compared to identify duplicate registrations?" Rationale: This wording better
suits the response choices given.

16. In questions 20 and 21, suggest that questions be further refined to identify nature
of linkage, e.g., periodic batch ftp uploads, real-time transaction-based data
transfer, etc. This begins to provide some interesting data.
17. Don't understand what is meant by the terms "electronically" and "manually" in
the responses. Can you clarify?
18. In question 23, would be interesting to know if
there are differences in what the
state and local officials can do for the "Both" response.
19. Recommend that a heading such as REMOVING (OR PURGING) VOTERS
FROM VOTER REGISTRATION DATABASE be inserted after question 23.
Rationale: Ease of
use for the respondent.This is the topic that the next several
questions deal with.
20. Questions 24 and 28 seem to be asking for the same information with slightly
different wording.
21. In question 25, need a citation for parenthetical (Section 8(d)(2) confirmation). Is
this NVRA?
22. In question 28, suggest you clarify what "moved outside of jurisdiction" means in
first response. I assume it means left the state? Recommend that the word
"registrations" be replaced with "voters" in the last response.
23. In question 29, aren't "car registrations" and "lists of automobile registrations"
the same thing?
24. Recommend that next section heading be revised to "2006 ELECTION
RESULTS". Rationale: I assume that the data sought is the offcial certified

election results, not the preliminary election returs. Certified results are not

available until 30 or so days after the election, as defined by state law.
25. In first line, change "Ballots" to "Ballot" for grammatical consistency with rest of
sentence. Under Note:, delete "For" for grammatical correctness.
26. Is "Ballots Counted" going to be a problem for election officials? As I understand
it, normally ballots don't get counted, only votes get counted. Is EAC asking
election officials to collect new data? (There's nothing wrong with this, in my'

view) But if so, do election offcials understand what's being asked for? (And
maybe I'm just wrong in my understanding.)
27. Description of
"At the polls" is a bit confusing when the counting activity is
included. Some jurisdictions have central count systems for voting machine
results, so no ballots are counted at the pollng place. Some jurisdictions send
absentee ballots to the polling place to be counted, so all ballots are counted there.
If you are trying to distinguish between pollng place and absentee, maybe that
can be accomplished by just saying where the ballot is voted and cast.
28. Recommend that description of
"Domestic civilian absentee ballot" be revised to
something like the following. "Domestic civilian absentee voter refers to a nonmilitary citizen who is otherwise eligible to vote, but is temporarily absent from
their place of
residence on election day, and meets their State's requirements to
vote absentee."

29. Recommend that "Domestic military citizen" be revised to something like:
"Domestic military absentee voter is statutorily defined as: etc.,etc."
30. Recommend that "Overseas military citizen" be revised to "Overseas military
absentee voter is statutorily defined as: etc.", and that "Overseas civilians" be
revised to "Overseas civilian absentee voter etc." These two descriptions should
be moved to immediately follow the Domestic military absentee voter, so all
descriptions of different categories of absentee voters are together.
31. What about the category of permanent absentee voters? Many states have this
designation for voters with medical conditions or disabilities that make it difficult
for them to get to the polls, residents of nursing homes, etc. Since they are not
absent from their place of residence they don't quite fit into the domestic civilian
absentee category above. This is also a category where voting fraud is often an
issue.
32. Under "Early voting" description I would add the word "special" in the second
line: "were no special eligibility requirements." Probably a nitpick, but avoids
possible (but somewhat sily) interpretation that normal eligibility requirements
don't apply.

33. Recommend that FW AB description be revised to something like: Federal WriteIn Absentee Ballot (FW AB) is an emergency absentee ballot available for use by
military and overseas citizens when they have properly requested but have not
received a regular absentee ballot from their local jurisdiction in time to return it
before the deadline. I would move this description to follow provisional ballot.
34. Would revise provisional ballot to something like: "Provisional ballot refers to a
ballot issued to a voter at the polling place when their eligibility to vote has not
been determined. Provisional ballots are prescribed by Section 302(a) of

America Vote Act."

the Help

35. Would it be possible to add further sub-questions to question 3l? For those who
answer yes, suggest you ask what is the time period, Also, what is the process, i.e.,
do voters have to go to clerk's office or are satellte locations provided, etc.?
Since this is defined by state law it shouldn't be too hard to answer.
36. In question 32, I would split the responses as follows: (A) Ballots cast in personat the polls, early voting, provisional ballots; and (B) Ballots cast absentee-

domestic civilian absentee, domestic military absentee, overseas military absentee,
overseas civilian absentee, FW AB

37. In question 35, I don't understand how "Ballot not timely received (absentee)"
would be a reason to reject a provisional ballot? Similarly "Incomplete ballot
form"? First of all, I don't really know what this means. If it means a ballot in
which no votes have been cast for some ofthe races, that's not a reason to reject,
because voters are not required to cast votes in all races for their ballot to be
accepted.
38. Suggest that description of absentee voting is not needed if previous suggestions
on rewording absentee voter categories are used. I think this description sort of
misses the point of this voting option, which is not voting before Election Day
(although true), but not being able to vote on Election Day due to inability to get
to the polls - either through physical absence or for other state-defined reasons.
39. I think the correct term is "advance ballot" rather than "advanced ballot." Also
appears in question 37.
40. In question 36, would add a column for Number Received and change Not
Counted heading to Number Rejected. Would also ask for how many of each of
these 4 categories of ballots were delivered to the voter and how many returned to
the election office by mail, fax, email or courier.
41. In question 37, column heading for responses should be Number Received and
Number Rejected.
42. Would use the same column headings for question 38.
43. In question 39, would change "Elector deceased" to "Voter deceased". Why
introduce a new term? Would change "Elector voted early at the polls" to
something like "Voter voted in person (either early voting or on election day)".
Would add the word request to "No ballot request application on record". What
does "Spoiled ballot" mean?

"Bonnie Glaser"

.: bg (gea re. berkeley. ed u:'
09/29/2006 02:24 PM

To lotero(geae.gov
ee
bee
Subject Comments on the 2006 Election Administration and Voting

Survey

Dear Ms. Laiza N. Otero:
Please find attached a letter to you with our comments on the 2006
Election Administration and Voting Survey, sent at 2: 30 EDT on September
29, 2006.

Sincerely,
Bonnie Glaser, Ph. D., Senior Researcher
Election Administration Research Center
University of California, Berkeley

510-642-8506

~

Comments on EAC 2006 survey. pdf

September 28, 2006

ATTN: Ms. Laiza N. Otero
US Election Assistance Commission
1225 New York Avenue, NW., Suite 1100
Washington, DC. 2005

Dear Ms. Laiza N. Otero:

We submit the following comments about the 2006 Election Administration and Voting
Survey, pertaining to "c) ways to enhance the quality, utility, and clarity of
the
information to be collected."
These comments are based on our experience conducting research with state and local
election offcials and in local election jurisdictions, both in terms of what data is possible

to collect and what would be useful for research. Questions are referred to as 'Q#' in
comments below.

Voter Registration Applications
Q4-Q6: It is likely that local election officials do not record if

registrations come in by

mail or in person, so it might be easier to make Q4 and Q5 optionaL. The same idea
applies to Q6, that is, it is unlikely that the local election offcials record where these
additional registration forms come from; even if they do record which ones come from an
outside agency, they wil not necessarily specify the agency in all cases.
California, each county has either a County Clerk who is the Registrar
QlO. In the case of
or a Registrar, rather than two titles that "share responsibility for registering voters" the
responsible party is called by different titles in different jurisdictions. The question could
be worded more clearly along these lines "mark all possible titles of the individual in
each jurisdiction that is responsible for registering voters."
this question is not clear. The choices provided are a
QI1. The usefulness and purpose of
mix of what is required by law and what is a state option, which could lead to confusion
in completing this and responses that do not reflect what the question is meant to
measure.

Election Day Results
Q31. Early voting wil likely vary across local jurisdictions and over time. For this
question to ilicit a more clear response, we recommend "Did any local jurisdiction in
your state conduct early voting for the November 7,2006 Election." It also might help to
ask the number of local jurisdiction that did conduct early voting.
ballots cast and of
ballots counted by different methods,
especially the five different forms of absentee ballots, wil provide extremely useful
information for understanding the different voting programs and their prevalence in each
Q32. This break down of

state. However, states that have a large number of
jurisdictions may have diffculty
getting this level of breakdown from each jurisdiction. While states should be
encouraged to collect this, perhaps they should have instructions for compilng the results
from less than all the jurisdictions (perhaps adding to their counts something like: "this is
the number of
ballots cast and counted in 40 out of 58 counties that provided this

information") .

Q35. This question wil produce valuable information. However, it wil be difficult for
states to collect the number for each reason for each
jurisdiction. We see that in the
previous survey the question was simply "Identify the five most common reasons that
provisional ballots were rejected," which did not require explicit counting of
the reasons.
We suggest offering an alternative way to report these reasons if the state is not able to
count the incidence of each one. Also, "administrative error" could mean different things
to different survey respondents. For that choice, we recommend something a little more
specific about the kind of administrative error, such as "poll worker error," or a space to
list the type of administrative error.

Absentee Ballots

"absentee ballots," because in California any
registered voter can request an absentee ballot without a reason or special qualification.

We are confused by the definition of

Q40. This data is only useful in the context of each state's laws. For example, do all
states require that the ballot be notarized? Some require the registration form for military
and overseas (FPCA) is notarized, but do they require that for the voted ballot as well

Poll Workers
Q43-Q45. Collecting information on poll workers is very important; however, these three
questions are too general to provide useful data. The previous survey asked for the
required number of
poll workers per precinct as established by law, and this question
should be asked again, but with an option to give more than one answer. For example, in
California the law is 3 workers per precinct, but for various reasons, such as language
needs, voting technology, size of pollng place, and precincts per polling place, the
required number is more than 3 for certain precincts.

Also the counties in California have their own preferred level of staffng (often 4 or
more) based on various needs at different polling places. In many cases, the legal
minimum is not considered adequate staffing. For practical purposes, the success in
staffng polling places is better reflected by evaluating how well they met their own
perceived needs. The states could ask local jurisdictions to report some numbers on that

question. For example, "how many precincts in your jurisdiction should have 3, 4, 5, and
6 workers at all times, and how many precincts did have that number."

there are "split shifts" where two poll workers are hired for one day, then the number in
Q44 wil not be useful compared to the number in Q43. Q44 could read the number of
"poll worker days served" so that two workers who split the day are counted as one.

If

Q45 wil not necessarily produce useful answers because states wil try to minimize this
number; poll worker staffing is an extremely sensitive topic. Asking about actual staffing
relative to preferred staffing (as suggested above) might be more productive than asking
about compliance with the laws, and wil stil reflect difficulties with recruitment.
It would also be beneficial to ask more from states about poll workers than their total
number hired and the adequacy of staffing levels. Some other important topics could be
. the five most common methods of recruitment,

. the five most common barriers to recruitment,

· the extent of problems with last minute dropouts and no-shows (such as the
number per jurisdiction of individuals who were signed up but declined to work in
the last week before the election or failed to show on Election Day), and
· whether any jurisdiction allows half day or multiple shifts.

Voting Jurisdictions and Polling Places
Q50. This question wil be very difficult to answer, as it requires a "yes accessible" or
"no not accessible" evaluation of thousands of pollng places, rather than a degree of
accessibility. Many polling places may be substantially more accessible than other

pollng places, but not completely ADA compliant on every measure. Also, many polling
places are not accessible on their own, but are adapted by the elections office just for
Election Day, and its unclear how to tease those out from others in this number. Q50
needs to be broken down into several questions or needs to include a more explicit
definition of

"accessibility" for the purposes of

the question.

Q51. It is unclear why there is this question on visual impairment and not any other type
of disabilty. It seems as though there should be a question on each major disability

category or none at alL. It is also unclear why any questions about voting equipment were
dropped from this version of
the survey. Asking about the number of
"accessible voting
machines" per polling place would be a better measure of accessibility to independent
voting for voters with various disabilities. A measure of

the prevalence of non-electronic

accessibilty devices must also be collected, in order to account for those jurisdictions
which meet this requirement without "machines." Note that all states should answer in
Q51 that all their polling places have the capacity for a visually impaired voter to cast a
private ballot, as required by law. Because some jurisdictions use the "accessible"
machines for all voters, the average number of accessible machines per polling place (for
those jurisdictions which use machines for accessibilty) should be more than one. That
average would be an interesting number to compare across states, capturing both the
uptake of accessible voting machines as well as the extent of accessibilty. While we
understand that the "accessibilty" of different machine types is not absolutely confirmed,
especially for all disabilities on all machines, collecting this data may stil give a fairly
good indication of the intent of jurisdictions to provide accessibilty.

Thank you for the opportunity to share our thoughts about this important survey of
election administration. Please contact us if
you have any questions about these remarks.
Sincerely,

Bonnie Glaser and Karin Mac Donald
Election Administration Research Center
UC Berkeley
111 Moses Hall
Berkeley, CA 94720
510-642-8506

"Wittman, Bradley S"

.:Wittman B(gmichigan .gov~

To lotero(geae.gov
ee "Thomas, Christopher M" -:ChristopherT(gmiehigan.gov;:,

09/29/200602:36 PM

"Anastor, Rayan" -:AnastorR(gmiehigan.gov~

bee
Subject Comments: DRAFT 2006 Election Administration & Voting
Survey ( Michigan)

Greetings Ms. Otero: The following comments are offered with respect to the

proposed 2006 Election Administration and Voting Survey released by the U. S.

Election Assistance Commission (EAC) in August. We are directing our comments
to your attention pursuant to the instructions published in the Federal
Register on August 1, 2006.
In general terms, we find the data requests provided under the section of the
survey entitled "2006 Election Day Results" overly detailed and ambiguous in
places -- a problem noted in the 2004 Election Day Data Survey released by the
EAC and the subj ect of much discussion at the meeting convened by the EAC on
April 4, 2006 to review and critique the survey tool.
It bears note that despite the fact that Michigan's Qualified Voter File (QVF)
system has been programmed to capture many of the data elements requested in
the proposed EAC survey, it will still be necessary for this office to contact
every local clerk in the state (i. e., Michigan's 1, 51 6 city and township
clerks) to collect key portions of the requested information.

This task will entail the analysis of the final version of the EAC survey to
identi fy the data elements which must be collected from the city and township
clerks, the development and design of our own survey tool, the distribution of
the reformatted survey tool to the state's 1,5l6 local clerks, the compilation
of the responses we receive, and the correction of any errors that we are able
to identify in the compiled data. This is essentially the same procedure we
followed when preparing our responses to the EAC' s 2004 post-election surveys.
As you may be aware, we aggregated the data we collected after the 2004
general election by county given the complexity and burden such data
collection exercises pose for Michigan. It is our intent to aggregate the
data collected after the 2006 general election in the same manner when
responding to the EAC' s 2006 post-election survey.
In view of the above concerns, we strongly urge the simplification of the
EAC's proposed 2006 post-election survey. As stands to reason, the more
complex and ambiguous the data requests, the greater the chance for errors,
confusion and unreliable results. Specific observations and recommendations
are outlined below:

ITEM 32
1) Item 32 asks for the number of ballots "cast" at the polls and the number
of ballots "counted" at the polls. Under the definitions, "ballots cast"
means those ballots that have been submitted but not "verified and/or

counted." Aside from the fact that a unique -- and therefore confusing --

definition has been given to the term "ballots cast," what would be an example
of a ballot that has been "cast" at the polls but not "counted"? Would
provisional ballots held for later evaluation fall into this category? (We
note that there are separate data entry lines for "provisional ballots" later
in the survey.)
2) In many cases, it is impossible for election officials to distinguish those

ballots mailed to "domestic civilian" absentee voters from those ballots
mailed to "domestic military" absentee voters. Consequently, any data that
relies on the accuracy of this distinction would be highly unreliable.
3) Similarly, it is often impossible for election officials to distinguish
those ballots mailed to "overseas military" absentee voters from those ballots
mailed to "overseas civilian" absentee voters. Again, any data that relies on
the accuracy of this distinction would be highly unreliable.
4) As you are aware, a Federal Write-In Absentee Ballot (FWAB) can be rejected
for a variety of reasons -- including the fact that the voter returned the
absentee ballot issued by his/her voting jurisdiction in time to be counted.
Given the definition given to "ballots cast" (i. e., submitted but not
counted), it would appear that the number of FWAB' s that will fall under the
"ballots cast" category will be alarmingly high and very misleading.
5) Again, given the definition of "ballots cast," would the "ballots cast"
entry for provisional ballots under Item 32 be the same as the entry for the
number of provisional ballots "rejected" under item 34?
ITEMS 36 and 38 (ITEM 37 does not apply in Michigan)

See comments above regarding the difficulty of accurately distinguishing
ballots sent to "domestic civilian" absentee voters verses "domestic military"
absentee voters and "overseas military" absentee voters verses "overseas
civilian" absentee voters.
ITEMS 39 and 40

Requesting breakdowns of the specific reasons why absentee ballots mailed to
"domestic civilian" absentee voters and "military and overseas" absentee
voters were rej ected would be tedious and burdensome data to document.
Consequently, the accuracy of any data collected would be questionable.

Recommendation
Given the issues identified above, the simplification of the data requested
under the "2006 Election Day Results" portion of the survey is strongly

recommended. The following is suggested:

A) Number of voters who participated in the election (include all voters who
attended the polls including those voters who were issued a provisional
ballot; all voters who returned an absentee ballot which was counted; and all
voters who submitted a FWAB which was counted).
B) Number of absentee ballots distributed to all voters.

C) Of the absentee ballots distributed to all voters, the number distributed
to I) voters wi thin the territorial limits of the United States and 2) voters
outside of the territorial limits of the United States.
D) Number of absentee ballots returned in time to be counted.

E) Of the absentee ballots returned in time to be counted, the number returned
by I) voters within the territorial limits of the United States and 2) voters
outside of the territorial limits of the United States.
F) Number of absentee ballots which were rej ected and not counted including

those that were returned late.

G) Of the absentee ballots which were rej ected and not counted, the number

returned by 1) voters within the territorial limits of the United States and
2) voters outside of the territorial limits of the United States.
H) Number of provisional ballots issued to voters.

I) Of the number of provisional ballots which were issued to voters, the

number which were counted and the number which were rej ected and not counted.
J) Number of Federal Write-In Absentee Ballots (FWAB's) submitted by 1)
domestic military voters and 2) overseas military and civilian voters.

K) Of the number of Federal Write-In Absentee Ballots (FWAB's) which were
submi tted, the number which were counted and the number which were not

counted.
Thank you for extending the opportunity to comment on the proposed survey.
Please do not hesitate to contact this office if you have any questions
regarding our comments.
Best Regards,
Bradley S. Wittman
Director, Elections Liaison Division
Michigan Department of State
Bureau of Elections
P.O. Box 20126
Lansing, MI 48901-0726
Phone: (517) 373-2540
Fax: (517) 241-4785

"Nicole Trella"

To lotero(geae.gov

.:ntre Iia (gelectio ns .state. md. U

ee

s~

bee

09/29/2006 04:06 PM

Subject FW: Comments on 2006 Draft Survey

From: Nicole Trella
Sent: Friday, September 29, 2006 1:09 PM
To: 'lkoterotQeac.gov'

Subject: Comments on 2006 Draft Survey

Please see attached comments.
Thanks,
Nikki Trella

Maryland State Board of Elections

410-269-2843
-:..Comments to Proposed 2006 Elec Admin & Voting Survey.pdf:;~

~

Comments to Proposed 2006 E lee Admin & Voting Survey. pdf

MARYLAND

STATE BOARD OF ELECTIONS
P.o. BOX 6486, ANNAPOLIS, MD 21401-0486 PHONE (410) 269-2840
Gilles W. Burger, Chairman

Linda H. Lamone
Administrator

Bobbie S. Mack, Vice Chairman
Joan Beck

Andrew Jezic

Ross Goldstein
Deputy Administrator

A. Susan Widerman

September 29,2006

Via Electronic Mail Onlv
Laiza N. Otero
U.S. Election Assistance Commission
1225 New York Avenue, N.W. Suite 1100
Washington DC 20005
Dear Ms. Otero:
Thank you for the opportnity to provide comment on the proposed 2006 Election

Administration and Voting SuI'ey. My comments relate only to the "2006 Election Day Results"
portion ofthe draft survey.

Applying the proposed definitions of "early voting" and "absentee voting" to Maryland's
absentee voting laws!, the State has "early voting" and does not have "absentee voting." As a
result, we would not be required to submit the information requested under the "Absentee Ballots"
section of
the survey i(i.e., the number of
requested and not counted absentee ballots, the number of
rejected absentee ballots and the reasons for the rejections, etc.). I assume that this is an
unintended consequence of

the definitions.

If early voting is generally considered "in person" voting and absentee voting is typically
conducted "by mail," perhaps the distinction between the two terms could be based on those
criteria, instead of whether eligibility requirements exist. Alternatively, a footnote or parenthetical
comment could be added under the "Absentee Ballots" section that requires states with "early

voting" that is conducted generally by mail to complete this section, even though the definition is
not technically met.
Question 32 asks for a breakdown of the ballots cast and ballots counted for domestic

military citizens, overseas military citizens, and overseas citizens. It is important to note that
statistics for these categories of voters can only be provided if the voters voted by absentee ballot.
A voter who falls into one of these categories, is in Marland on Election Day, and wishes to vote
in person at the pollng place wil be counted as "at the polls." To clarify that these voters voted
by absentee ballot, the report could refer to these categories as "Domestic military citizen absentee
ballots," "Overseas military citizen absentee ballots," and "Overseas civilian absentee ballots."
These descriptions would be similar to the description of
"Domestic civilian absentee ballots."

During the 2005 Legislative Session, the Marland General Assembly passed legislation repealing the eligibility

requirements for a voter who wishes to vote by absentee ballot. Under the new law, any voter may request an absentee
ballot.
FAX (410) 974- 2019

MD Relay Service (800) 735-2258

Toll Free Phone Number (800) 222-8683

http://www.elections.state.md.us

151 West Street Suite 200

Annapolis, Maryland 21401

Letter to Ms. Otero
Page 2

September 29, 2006

Question 38 requests information about the number of Federal Write-In Absentee Ballots
(FW ABs) received. Question 36 requests the number of requested and rejected ("not counted")
absentee ballots but specifically excludes FW ABs. Question 40 is silent as to whether the reasons
why any FWABs were rejected should be included, although the lack of responses specific to
FW ADs suggests that these ballots are not to be included. If Question 40 is not intended to report
the rejection reasons for FW ADs, there is no place to report the number of FW ABs either counted

or rejected. This wil likely result in misleading absentee bal10t statistics.2

To avoid this result, I suggest either: (1) including a separate question requesting the number
of FW ABs received, the number of FW ABs rejected, and the reasons for the rejections; or (2)
specifical1y including FW ABs in Question 40. If the Election Assistance Commission includes a
request for the number of
rejected FW ABs and the reasons why the ballots were rejected, rejection
reasons specific to FW ABs (i.e., "not registered" and '4no bal10t application on record") need to
included.3
Than you again for the opportunity to comment on the proposed 2006 Election
Administration and Voting Survey. If you have any questions about these comments, please

contact Nikki Baines Trella at 410-269-2843.
Sincerely yours,

¿d:// ¿/J /J
Linda H. Lamone

State Administrtor

2 In responding to the 2004 Militar and Overseas Absentee Ballot Survey, Maryland reported that the total number

of absentee ballot returned was 9,538. As requested, this number did not include 1,768 FWABs received by local
election offcials in Maryland. We also reported tht the total number of absentee ballots counted (including FW ABs)

was 10,205. By not including FW ABs in the number of received absentee ballots, it looks like more absentee ballots
were counted than were received.

3 The Federal Voting Assistance Program's 2004 Post-Election Voting Survey of Local Election Offcials included a
specific question about the number of FW ABs ilat were rejected and the reasons why the ballots were rejected. (See
Question 8 of the survey.) Another rejection reason would be that the voter's State absentee ballot was voted and
timely received.

"Mike McCarthy"
.:Mike. Meearthy(gstate.mn .us
~

09/29/200604:28 PM

To "'Iotero(geae.gov'" -:Iotero(geae.gov::
ee "Jim Hansen" -:Jim.Hansen(gstate.mn.us::, "Alberto
Quintela" -:Alberto. Quintela(gstate .mn .US"

bee
Subject 2006 Election Administration and Voting Survey

Ms. Otero,
Please find enclosed the comments of Secretary Kiffmeyer regarding the 2006 Election Administration
and Voting Survey.

Mike McCarthy

Minnesota Secretary of State's Offce

-:.:EAC Comment 09.29.06.doc;:;:

the Secretary of
State of
Minnesota
may be public data subject to the Minnesota Data Practices Act and/or may be disclosed to third
parties
NOTICE: E-mail correspondence to and from the Office of

EAC Comment 09.29.06.doc

MINNESOTA SECRETARY OF STATE OFFICE
Mary Kiffrneyer, Secretary of State

MEMORANDUM

Date: September 29, 2006
To: United States Elections Administration

From: Mary Kiffmeyer
Minnesota Secretary of State
Re: Comments on Draft 2006 Election Administration and Voting Survey

Minnesota, I wish to
commend the hard work of everyone who has contributed to developing the proposed
Information Collection Activity.

First, as Minnesota Secretary of State, and on behalf ofthe State of

I have provided comments on areas of the collection activity that I believe could benefit
from revision. My hope is that you wil find my commentary helpful, and that it wil
serve to enhance this activity. In response to the four questions presented, I make the
following comments.

(a) Whether the proposed collection of information is necessary for the proper
performance of the functions of the agency, including whether the information shall
have practical utilty;
· Much of the information requested may have relevance for election
administration. However, even though Minnesota has a well developed
Statewide Voter Registration System, having had no notice that some of this
information would be collected this election cycle means that some of this
information has not been tracked in the past in a form which would provide an
answer to the question.

· As examples, question (32) asks for a separate count of "Domestic Military
Citizens," and "Overseas Military Citizens." Because the information
requested on the Federal Post Card Application permits sending voting
materials to a forwarding address, it is not always clear if a service member is
overseas or not. In addition, question (36) asks for number of
ballots "Not

Counted." Ballots "Not Counted" is not defined. Does it include ballots
mailed out, but returned as undeliverable or not retured at all, along with
rejected ballots? The answer to this question, as it is written, may give a

misleading result.

· Question (39) requests information as to the reasons for the rejection of
domestic civilian absentee ballots which is beyond the specific information
maintained by the local election offciaL. The reasons for rejection are noted
by this state but placed into fewer categories in the state registration system
than the expectation of this question. With notice, this information can be
categorized as requested, but it cannot be done after the fact.
· Although question (45) asks the number of polling places "that did not have
the required number of poll workers" for the election, it seems implausible
that answers to a question of
this type, i.e., asking a local election
jurisdiction
to admit it did not comply with state law, wil result in an accurate answer.
· Questions (41) and (42) are to the number of

under

votes and overvotes

reported. However, currently, this information is not tracked in the system
and cannot be tracked without redoing part of the system. With notice, this
can be done, but it wil require a substantial systemic change in the state
which cannot be accomplished on such short notice, and so close to the 2006
election.
(b) the accuracy of the agency's estimate of the burden of the
proposed information collection;

· I believe this estimate underestimates the cumulative burden of the proposed
the
information proposed to be collected has not been required to be tracked in the
past so it may be unavailable. In addition, NVRA reporting requirements are
now being placed upon all states as required reporting with this proposal, with
no notice being given so states and local jurisdictions could begin collecting
this information in a timely manner. In fact, as some of these requirements
may increase election administration costs or conflict with current state
statutory requirements; it may be necessary to amend current statutes to
become consistent with these revised reporting requirements.
information collection on all levels of election offcials. Some of

· In particular, proposed questions (25), (26), and (28) are not consistent with
current state statutes and would place a heavy burden upon local election
officials to research the answer.

· In addition, this state permits precincts to have combined pollng places which
would not readily appear in the Statewide Voter Registration System; so this
total would have to be manually prepared this election cycle. If we had notice

that this information would be requested; the data could be entered into the
system as part of routine data entry in the future.
(c) ways to enhance the quality, utilty, and clarity of

the information to be

collected; and

· Provide longer notice of information which is to be sought so that systemic
changes can be made prior to the request for the information.

· Permit states that do not come under the NVRA the option to respond to a
question that information only required under the NVRA was not collected in
that state, if that was the case, and waive the reporting requirement for that
question at least until the succeeding election cycle. Those states can then
begin collecting that information for the next reporting cycle.
(d) ways to minimize the burden of the information collection on

respondents, including through the use of automated collection techniques or
other forms of information technology.

· Due to the close proximity of the 2006 election to the time when the proposed
information collection activity was released for comment, the burden on
respondents is significantly increased because not all of the information has
been previously sought, nor collected. Consequently, when this collection

activity is finalized, it wil be too late for some of the information to be
collected from voters by local election officials, placing an undue burden upon
them for providing this information. This wil be particularly burdensome on
local offices with smaller staffs.

· Automated collection techniques, when coordinated with the information
maintained in state registration systems, wil enhance this process, and wil
relieve the burden on all offcials affected.

CONCLUSION
On behalf of the State of Minnesota, as well as the citizens of the United States
of America, I would like to thank the Elections Assistance Commission for allowing me
to comment on the proposed Information Collection Activity. Your assistance in this
process is to be commended. The integrity of not only our elections process, but that of
our democracy is contingent upon setting forth standards that are based upon reliable
information. This should be taken as an opportunity to bolster a system that wil stand
strong for generations to come. With that said, thank you for putting this process into

motion, and I wish the Election Assistance Commission much success in these, and all
future efforts.

"Scott Novakowski"
c:s nova kowski(gdemos. org~

To lotero(geae.gov
ee

09/29/200604:20 PM

bee
Subject Comments on Notice of Information Collection Activity;
Proposed Information Collection; Comment Request

Dear Ms. Otero,
I attach a cover letter and comments regarding Notice of Information
Collection Activity; Proposed Information Collection; Comment Request
published at 71 Fed. Reg. 43477 (August If 2006). Do not hesitate to contact

me with any questions.
Sincerely f
Scott Novakowski

Scott Novakowski
Policy Analyst

Demos: A Network for Ideas and Action

220 5th Avenue, 5th Floor

New York, NY 10001
Phone: 212-389-1415
Fax: 212-633-2015

Email: snovakowskiCCdemos.orq

http://ww.demos.orQ

Visit Demos' redesigned, web-based e-joumal Democracv Dispatches for a daily news-feed on
democracy-related issues, a blog analyzing current trends, and in-depth commentary.

~

~

EAC Cover Lelter.pdf EAC Survey Comments - FINAL.pdf

September 29,2006

VIA E-MAIL
Ms. Laiza N. Otero
U.S. Election Assistance Commission
1225 New York Avenue NW, SuÌte 1100
Washington, DC 20005

Re: Notice of Information Collection Activity: Proposed Infonnation Collection:
Comment Request
Dear Ms. Otero:

Demos welcomes the oppoitunity to submit comments in response to the Notice
ofInformation Collection Activity; Proposed Information Collection; Comment Request.
The notice was published at 71 Fed. Reg. 43477 (August 1,2006).

#¿~~'

Respcctfully Submitted,

Scott C. Novakowski
Policy Analyst
Demos: A Network for Ideas & Action
220 5th A venue, 5th Floor
New York, NY 10001

Attachment

220 5th ,\venue', 5tb fi.
New York. NY 10001

j', 2.J 2.65:$ l/íOc)
E 212.633.2015

inf ,ürg
\-\;\V\V. d C IJ10:'. \) r'g.

Before the U.S. Election Assistance Commission

Notice of Information Collection Activity; Proposed Information Collection;
Comment Request

Comments provided by Demos: A Network for Ideas & Action
I. Introduction

Demos: A Network for Ideas & Action submits these comments in response to the
Election Assistance Commission's Notice ofInformation Collection Activity; Proposed
Information Collection; Comment Request published at 71 Fed. Reg. 43477 (August 1,2006).
Demos is a non-profit, non-partisan think tank and advocacy center dedicated to helping America
achieve its highest democratic ideals. Demos' primary interest in the EAC's Election
Administration and Voting Survey is to create a survey instrument that accurately collects muchneeded and usable data on elections.

Demos has a strong record of producing high quality research and conducting advocacy
the topics covered on the proposed survey. Through the NVRA
Implementation Project, a collaboration to improve states' compliance with the public assistance
provisions of
the National Voter Registration Act, Demos has made extensive use of
the EAC's
(previously the Federal Election Commission's) biennial report on the impact of
the NVRA.
Demos also has assumed leadership in the national debate on Election Day registration,
published several reports and mounted advocacy campaigns on felony disfranchisement and
published Placebo Ballots: Wil "Fail-safe" Voting Fail?, one of
the first reports to address
HA VA's provisional ballot requirement. Additionally, Demos staff
members have had
significant graduate-level instruction and experience in survey construction. This work puts
Demos in a unique position to speak to the types of data needed for the understanding of the
electoral issues sought by the survey.
campaigns around several of

We offer the following recommendations with the hope that the EAC can further develop
its role as a centralized clearinghouse for election-related data.

II. Data Gathered with Respect to the National Voter Registration Act

A. Additional Information Required
The EAC has the responsibilty to administer a survey on the impact of
the National
Voter Registration Act. With respect to public assistance agencies, the current survey gathers the
basic information, i.e. the number of registrations received from public assistance offces.
However, additional information is necessary for more than a surface understanding of
registration activity. Collection of
survey's purpose to ascertain the impact of

this voter

this additional information would greatly advance the
the law.

Specifically, we encourage the EAC to request the following additional information:

· The number of applications, recertifications, and changes of address processed by public
assistance agencies;

· How often a recipient of public assistance is required to recertify or renew her benefits;
· The percentage (or raw number) of agency interactions that are conducted via telephone,
mail, or internet;
· The individual public assistance programs (e.g. Food Stamps, Medicaid, TANF)
administered by each office designated as a public assistance agency; and
· The number of completed voter registration applications submitted by each public
assistance offce or agency.

The current survey does not provide us with the information necessary to contextualize the
raw number of applications. We recommend that the survey ask for the number of applications,
recertifications, and changes of address processed by each public assistance agency - a figure
that represents the total number of transactions in which voter registration should have been
offered under the NVRA. i

Raw numbers can be misleading without this information. For example, assume that a state
registered 10 individuals in its public assistance offices. If that office only had 11 client
interactions requiring an offer of
voter registration services, it did a fantastic job. However, if
100,000 individuals completed NVRA-covered transactions, registering only 10 of
these
individuals may be indicative of a serious failure to implement the law. Furthermore, it is wholly
possible that as office traffic rises faster than the number of applications completed, an increase
in the raw number of applications is actually accompanied by a decrease in the percentage of
clients completing voter registration applications.
In order to most accurately gauge an agency's NVRA-covered traffic flow, two
additional pieces of information would be quite helpfuL. It is important to know what percentage
(or the raw number) of an agency's interactions are conducted via telephone, mail, or Internet.
While some states require voter registration to be offered to clients interacting remotely, many
do not. Additionally, information on how often a recipient is required to renew or recertify
benefits will be useful in understanding how many times a single recipient is offered the
opportunity to register.
Collecting the above information relating to office traffic is not likely to place an undue
burden on public assistance agencies. We know that the federal government already requires

this information for the Food Stamp program and we suspect that it
Food Stamps was the only program for
which this data were available, its traffc flow could serve as a proxy measurement for general
agencies to track much of

may be tracked for Medicaid and WIC as welL. Even if

public assistance use.

1 This information is different than agency caseload, a number that does not capture recertifications, changes of

address or initial applicants who were denied benefits but who are stil required to be offered the opportunity for
voter registration upon application. Nevertheless, caseload data would be preferable (as an indicator of covered
transactions) to nothing.

2

Finally, we urge the EAC to seek two additional pieces of information to ensure that
voter registration services are being offered at public assistance agencies: (1) identification of
individual programs (e.g. Food Stamps, Medicaid, TANF) administered by each offce
designated as a voter registration agency; and (2) the number of completed voter registration
forms submitted by each public assistance agency or office. Collecting such information would
help ensure that all covered programs are offering voter registration and facilitate identifying
offices that may be having difficulties or are especially successful in implementing the law.
B. Existing Questions

Drawing on our knowledge of effective surveying, we offer the following additional
comments about the survey's format and language used in questions relating to the National
Voter Registration Act.

Question 6 asks for the number of registration applications received from or generated
by several categories of designated voter registration agencies including "All public assistance
agencies that are mandated as registration sites under the NVRA." Question 11 asks elections
offcials to identify "each and every" state "office or agency" that is designated as a voter
registration agency.
We recommend moving Question 11 toward the beginning of
the survey so that it appears
before Question 6, as it did in the survey instrument used in the previous EAC report. This
proposed ordering wil require officials to list relevant NVRA agencies before tallying the total
number of applications received from such agencies.

In addition, the format of the list of categories that follows Question 11 looks more like a
than a template for a listing of designated agencies. The current
placement of
the question next to others in which similarly formatted answer templates actually
are multiple choice check-offs adds to the assumption that Question 11 should be answered
similarly. We suggest that the answer to this question be reformatted to make clear to
respondents that they need to list all agencies for each category listed.
multiple-choice check-off

Finally, the request for a listing of "each and every state and local government offce or
agency designated as a voter registration agency" in Question 11 is vague and subject to
multiple interpretations. For example, a respondent could list the name and address of each
offce (meaning the physical building) that is designated as a voter registration agency.
Alternately, a respondent could simply list "Department of Social Services," meaning the
abstract agency with multiple physical offce locations that administer assistance. In order to
maximize the usefulness of

the survey, we recommend'that respondents list the overarching

agencies designated as providing public assistance (i.e. Department of Social Services) as well as
the address of each physical offce building at which registration is conducted.

Question 7 asks for the "Total number statewide and by county/local jurisdiction of
registration applications identified in response to Question 6 (regarding different designated

3

voter registration agencies) that were: Duplicates of other valid voter registrations; Changes of
address, name, or part; or Invalid or rejected (other than duplicates)."
Prom the wording and context of Question 7, it is not clear if the category for "Changes
of address, name, or part" is meant to include all such changes or only those making these
changes within their local
jurisdiction. Since Question 8 (asking for the number of
new, valid
registrations) specifies that name, address changes, and changes of part within the local
jurisdiction are not to be counted as "new, valid registrations," it seems that a voter changing an
address outside of their jurisdiction would be counted as "new" under Question 8 rather than a
change of address under Question 7. If this is the case, Question 7 of the survey should clarify
that respondents should report only the number of address, name, or part changes within the
local jurisdiction.

Without this clarification, it is entirely possible that the same application might be
counted in more than one category, e.g. a change of address outside the local jurisdiction might
be counted as both a new registration and also a name, address, and part change. The
possibility of
this double-counting wil reduce the value of
the data collected. Furthermore, this
ambiguity may lead to different interpretations in different states, hindering meaningful crossstate comparisons.
Question 12 inquires about the voter registration training provided to employees of

designated voter registration agencies. The proposed multiple choice answers are helpful in that
respondents must clearly state the level of
training that is provided. Por the second answer listed
(stating that the office conducts training for "some, but not all" agencies), we recommend
inserting a follow-up question asking for which agencies training is provided. A common theme
in the NVRA Implementation Project's work in public assistance agencies is that front line
agency workers are not provided adequate training on voter registration procedures. A question
asking to whom training is provided would be helpful in understanding why a state mayor may
not be successful in implementing the NVRA.

III. Data Gathered with Respect to Election Day Registration

As registration deadlines may pose a barrier to a significant number of citizens who wish
to cast a ballot, it is important for policy reasons that we have an accurate count of the number of
citizens who utilize policies allowing them to register and cast a ballot on the same day.
Question 2 asks for the total number of persons in each county/jurisdiction who

registered to vote on Election Day. The survey states this question is "Only applicable to states
with Election Day registration (Idaho, Maine, Minnesota, New Hampshire, Wisconsin, and
Wyoming)." We suggest that Montana be included among the list of states to answer this
question. In 2005, Montana passed a law allowing its citizens to register and vote on Election
Day at the county election administrator's offce. See Mont. Code Ann. § 13-2-304 (effective
July 1,2006). This law wil be fully implemented in the November 7,2006 General Election. A
comprehensive count of the number of citizens who utilize Election Day registration should
include the number of
voters in Montana who register and vote on Election Day.

4

i.
i

In addition, several states, including five states that have Election Day registration, also
allow what we have termed "same day registration." Under this system a citizen may register to
vote and cast a ballot on the same day during periods of early voting. To further gain an
understanding of
how many voters register and vote in the same day, it would be helpful to add
the following question:

your state allows early voting, how many citizens registered to vote and cast a ballot on the
same day during the period of early voting?"
"If

iv. Data Gathered with Respect to Felon Disfranchisement

Felon disfranchisement laws are often complicated and a recent study2 indicates many
elections offcials themselves do not completely understand the laws. We are concerned that the
wording of
the series of questions relating to felon disfranchisement wil cause more confusion
among those completing the survey and those utilizing its results. Our comments and
recommendations to avoid such confusion follow below.

Are the following classes of persons eligible to vote?
At) Those who have been convicted of a felony?
o Yes

o No
A simple "yes" or "no" response to this question cannot accurately capture most states'
laws. "Those... convicted of a felony" is a "class of persons" that may include both those able
and unable to vote within a single state.
In New York, for example, only those incarcerated or on parole for a felony conviction
are barred from voting. Those on probation, even if for a felony conviction, remain eligible to

vote. Those "convicted of a felony" who have completed their sentence of incarceration also can
vote. So how should a New York official answer this question? Answering "yes" (as would
seem to be required since a certain number of those convicted of a felony can vote) would void
Question A2, which we do not believe is the intention of
the Commission nor the best way to
ascertain information about a state's felon disfranchisement practices.

A recent studl has documented a widespread belief in affected communities in several
states (including New York) that a felony conviction permanently bars an individual from voting
even when that is not the case. We fear that the inclusion of
this question wil further perpetuate

2 See "Boards of

Elections Continue Ilegally to Disfranchise Voters with Felony Convictions" (2006) by the
Law and Demos: A Network for Ideas & Action, available at
http://www.demos.org/pubsINY Survey Report031506.pdf.
3 See "Studies of Voting Behavior and Felony Disenfranchisement Among Individuals in the Criminal Justice
Brennan Center for Justice at the NYU School of

System in New York, Connecticut, and Ohio" (2005) by Ernest Drucker and Ricardo Barreras, available at
http://www .sentenci ngpro i eet.org/pdfsl drucker. barreras. pdt'

5

misinformation about felon voting laws. We recommend alternative language for the question,
which we have provided at the end of
this section.
A2) If "No", are they eligible to register or re-register upon pardon, issuance

of eligibilty, or restoration of their Civil Rights?

of certificate

As with AI, the phrasing of this question does not capture the nuances of state felon
disfranchisement laws. As a result, different respondents are likely to interpret questions
differently, thereby reducing meaningful use of
the survey results.
First, the question refers to the "restoration of Civil Rights" but, in many states,
restoration of civil rights is much different than, and not necessarily connected to, restoration of
voting rights. In New York, for example, a felon's voting rights are automatically restored upon
completion of the sentence of incarceration or discharge from parole, whereas restoration of civil
rights would be applied for at a later time. At least in the case of
New York, asking about
restoration of civil rights does not seem to make much sense.
Similarly, there is a good deal of variation among state laws on certificates of eligibility
and other similar certificates that the survey's language does not capture. Using New York as an
example once more, those on parole can apply for a Certificate of Relief from Disabilties, which
would allow them to vote while still on parole. This is much different than a policy that requires
application for a certificate of relief after completion of
the individual's entire sentence.

Finally, the inclusion of pardon, issuance of certificate of eligibilty, and restoration of
Civil Rights together does not provide any valuable information when presented as a "yes" or
"no" question. A pardon, which is generally granted by a governor, is much different from a
certificate of eligibility in which an individual must often navigate a complex set of bureaucratic
procedures to obtain. Finally, in a state like New York, asking about restoration of civil rights is

largely meaningless, as a felon would likely have finished serving her sentence (and thus have
had her voting rights automatically restored) prior to applying for restoration of civil rights.
The remainder of the questions regarding felony disfranchisement laws is also confusing.

We propose the following clearer and more user-friendly set of questions:

~ Does your state currently disfranchise:
o Individuals who are currently incarcerated for a felony conviction? Yes No
o Those currently on parole for a felony conviction? Yes No
o Those currently on probation for a felony conviction? Yes No
~ Can an individual disfranchised for a felony conviction ever regain the right to vote? Yes
No
o If
yes, what are the requirements for voting rights restoration?

· Completion of sentence of incarceration
· Discharge from parole and/or probation

· A pardon
· Other

6

~ Does an individual disfranchised due to a felony conviction have to pay all fines, fees,
and/or child support prior to having her or his voting rights restored?
~ Does an individual whose voting rights have been restored have to produce
documentation of her status when registering to vote?
~ Who notifies elections officials to remove an individual from the voting rolls due to a
felony conviction?

~ If applicable, are elections offcials notified when an individual convicted of a felony is
once again eligible to vote?
o If yes, how are elections officials notified of this change in eligibility?

V. Data Gathered With Respect to Provisional Ballots
According to the EAC, in the 2004 election, elections officials ultimately rejected over
half a milion provisional ballots. To date, we know very little about why so many provisional
ballots were rejected. There has been no systematic analysis of
the specific reasons why
provisional ballots were not counted. It is essential that this information be compiled to allow
clear insight into the effectiveness of provisional ballots.

While it is a significant improvement to ask respondents for the number of rejected
ballots categorized by the reason for rejection, rather than asking for the five most frequent
reasons why a ballot was rejected (as was done on the previous survey), the response categories
proposed under Question 35 are not mutually exclusive. A fundamental principle in creating
survey questions of
this sort is that each response should fit into one, and only one, category.
Here it is possible for the same provisional ballot to be accurately counted in multiple categories.
For example, it is not clear what the difference is between the categories labeled "Name missing
from voter list" and "Not registered." If a person's name were missing from the voter list, it
would also follow that the same person could also be considered to be "not registered." In
another case, if someone does not produce ID at the polls and fails to return with ID within 24
hours, do they belong in the category of
"No identification provided" or "Non-appearance within
24 hours"? Also, some states may allow a voter longer than 24 hours to appear with ID. Such
ambiguity wil make it difficult to create meaningful cross-state comparisons.
Finally, we are doubtful that some of
the categories listed are indeed reasons to lawfully
reject a provisional ballot. For example, can a provisional ballot be rejected because the voter is
a "First time voter registration on Election Day"? In such an instance, the ballot would likely be
rejected because the registrant did not have the proper ID to register at the polls on Election Day
and was not able to subsequently provide adequate ID, rather than simply because the voter
registered for the first time on Election Day. Another suspect category is that of "Elector
challenged." An elector being challenged would be a reason to cast a provisional ballot, not to
have that ballot rejected.

VI. Conclusion
Demos is encouraged by the EAC's dedication to becoming a national clearinghouse on
election data. We applaud the Commission's efforts in combining and revising its various

7

survey instruments in a manner that captures accurate information. We urge the EAC to adopt
our suggestions to clarify the questions asked of elections offcials and to provide the data
necessary to accurately evaluate and improve our electoral system.

8

"Neal McBurnett"
.:neal(gben .boulder.eo. us~
09/30/2006 01 :30 AM

To "Laiza Otero" -:Iotero(geae.gov;:
ce
bee
Subject Comment on EAC's 2006 Election Information and Voting

survey

This comment is submitted in response to the request in the Federal
Register, Document 06-6602
http:// eac. gov leav _ survey. asp

Questions about audits are completely missing from the proposed 2006
Election Information and Voting survey. Audits are critical to the
securi ty of elections and voter confidence, and insights and feedback
on current audit practice is sorely needed so they can be improved.
The following information, at a minimum, should be provided for
each county or other voting jurisdiction.
I)
2)
3)
4)
5)
6)
7)

Was a post-election audit of election results performed?
How many electronic voting devices were used in the election?
How many electronic voting devices were audited?
Were all contests per device audited, or just some of them? How many?
How many problems or issues arose during the audits?
How were the issues resolved?
Was your audit designed to help also detect problems with
your central tally equipment?

More information on problems with current audit procedures is

available at

http://www.coloradovoter.net/moin.cgi/ManuaICountAudi t
Thank you,
Neal McBurnett

http://mcburnett . org/neall