Page 7.38. Put a period at the end of the sentence to the right of "20".

Page 7.32. Research goals. The last green dot seems to be misplaced.

The table under "Methodolgy overview should be moved up so that it is directly under the heading instead of an inch or more below it.

Page 7.21. Instructions. Right most paragraph. The word "graphically" seems wrong. Delete it, or replace it with different language.

Page 7.13. User requirements, Conclusion. Change "interview" to "interviews" and insert the word "with" between the words "play" and "our".

Page 7.6. 4th bullet after the heading "Assumptions". 2nd line - delete the word "official".

Page 7.3, 2nd paragraph. I don't think the word "informed" on the third line is the right word. Suggest "influenced" or a similar word.

4th paragraph. The word "Reviewo" should be "Reviewing".

2nd bullet under heading "Guiding criteria" . I don't think that "2005 Voluntary Voting System Best practices (VVSG)." is correct.  Suggest "2005 Voluntary Voting System Guidelines (VVSG) best practices." Additionally, the word "informed" on the 3rd line does not seem to be the right word. Suggest replacing it with "influenced" or a similar word.

1.  Page 7.3, 4th bullet: remove "p" and end of bullet heading (Reviewp existing materials)

2.  Page 7.3, 6th bullet: capitalize "V" in "Help America Vote Act."

3.  Page 7.3, 2nd bullet under "Guiding criteria": As it is unlikely that the VVSG submitted to the EAC in 2007 will be effective in 2007, is there another way to describe the "2007 VVSG" references in this bullet?

4.  Page 7.4: In the text under the table, there are references to "triangles" but the table shows circles. 

5.  Page 7.15, "Experience" under "General findings summary": typo at end of first sentence ("one0").

6.  On page 7.8, the third recommendation for optical scan/full face ballots states that text for ballot questions and referenda is "most easily read in a two-column, side-by-side format."  The image on page 3.20, however, shows a one-column ballot question and the text under "referendum content" gives information about one-column, two-column, and three-column questions.  Assuming state law does not specify the format for ballot questions, is the recommended format one-column (as the image on page 3.20 would suggest) or two-column ballot questions (as the text on page 7.8 suggests)?  If the recommendation is two-column, I would suggest showing a two column ballot question on page 3.20.

On page 6.7 I agree from observations in my jurisdiction, including my own voting habits over the years, that posters on the wall are ineffective methods of informing voters of rights, duties and methods. The more that is added to the wall, the less it is looked at.

1.  On page 6.6, the report notes that Colfax County deviated from the recommended best practices for the bilingual ballot (one column for English and one for Spanish).  The Colfax County ballot (with one instruction in English and the Spanish translation underneath) (page 6.13) appears, however, to mirror the recommended two language optical scan ballot on page 3.35.  Is the image on page 3.35 the recommended best practice or is the best practice one column for English and one for Spanish?  If it is the later, could there an image of what that ballot would look like be added to the report?

2. Page 6.14: I think the heading for #5 and #6 should be "Contest titles" and "Contest instructions."  The term "contest" (rather than "Question titles" and "Question instructions") is used throughout the document, and no ballot questions are shown on the Nebraska ballots.

3.  Page 6.16, third paragraph under "State metrics/reporting": "questions" should be replaced with "contests."

I really question the comment (5.2) that 25 point font is more important than not scrolling.  The example on page 5.11 shows a ballot on which the voter has to scroll to a second page to see all of the candidates.  While I am sure it is true that ballots with large type are more usable, sticking some of the candidates on a "second page" is more likely to result in you explaining why you did so to a judge. 

On page 5.12, there is a screen shot for write-in voting.  The "typewriter" on the screen is in A-B-C order.  We also use the A-B-C order on our touchscreen for write-in voting but have received complaints and requests to use the "QWERT" order.  Was any research conducted on which letter order is easier to use?

On pages 5.14 & 5.15, paper receipts are mentioned.  While there are no best practices for the design of a paper receipt, it would be helpful if Design for Democracy could offer some best practices.

On page 5.30, the voting unit displays the "inactive alert" screen shot after 5 minutes of inactivity and resets the ballot after 6 minutes of inactivity.  Was this 6 minute timeframe based on research or an existing voting system's limit?

On page 5.34, the screen shows how a user can change the font size.  Are there any recommendations on what font sizes should be offered?

The instructions in section 5.21 regarding the audio settings imply the voter will make the adjustments as to turning on/off or setting the volume levels by touching this screen.  A sight-impaired voter will be unable to do this.  Should this not be at the beginning of the voting process so that the poll worker or someone assisting the voter could begin the audio process for the sight-impaired voter?  On our system the screen goes blank once the audio ballot is selected and all selections are made using the ADA buttons.  Then the voter could be instructed how to set their own volume level. 

Page 4.2 5th goal under "Full-face DRE ballots". The discussion regarding more or fewer pages seems inappropriate since a full-faced DRE ballot is by definition a single page. Additionally, while the minimum font-size recommendation is sound, long names and column widths often force Election Officials to use a smaller font.

Page 4.21, under heading "Referendum title" change "seperate" to "separate".

Page 4.11, right column next to last line, strike the space between the "i" and the "m" in the word "simply".

Page 4.6, on the third line after the heading "Design" strike one of phrases "are based on".

Page 4.2. The first sentence under the heading "Full-face DRE ballots" may need some work. I would think the phrase "Voluntary Voting System Best practices (VVSG)." should read "Voluntary Voting System Guidelines (VVSG) Best Practices." As an alternative - Best Practices should not be capitalized.  

3rd goal under the "Full-Face DRE ballots" - add a period following the word "instructions".

With the two language ballots there is an indication that the number of sheets each voter must handle also doubles. The more sheets that a voter handles the greater the chance of error for the voter, such as missing one side of eight different sides thereby not voting in certain contests. There is also a greater chance of confusion at the check-in table for the poll workers who are handing out ballots, checking names, assisting voters with affidavits of current and continuous residence, and assisting new voters. It would likely be better to have a ballot dedicated to a single language so as to keep the ballot length smaller. If a second, third or more language requires ballots to be printed in those languages, then entirely separate ballots should be available for each language. Further, the more ballots there are the greater the chance that the ballto box will fill during the day. Although states have clear procedures for removing cast ballots from full ballot boxes and securing the ballots in plain view until the polls close, voters have a general discontented feeling when they see an election official handling cast ballots while the election is in process. Any action that creates suspicion among voters should be eliminated. With the running columns of black boxes on the left, the vertical and horizontal borders, teh grey and the blue colors, the instructions and all else, the ballot can be confusing due to the variety of information printed on each sheet, it can be "loud". This can be avoided by simplifying the information on a ballot using the design principle "keep it simple". 

On page 3.26, ballots which require an arrow to be connected should be discouraged. There are few comparable situations anywhere else, if any, where a choice is indicated by connecting two ends of an arrow. However, the option of filling in a circle or oval is quite universal from high school tests to consumer information surveys. The instructions state that you should not mark outside the arrow. What is outside the arrow? There are no defined boundaries as to where "outside the arrow" would be. With oval or circles the form of the particular shape clearly define where the marking should begin and end.

On pages 3.16 and 3.17 relative to the candidate separation line, I prefer the line below the candidate's name, such as in the U.S. Senator race "Dennis Weiford - Blue", to be extended completely to the left and right so as to form a complete box around the candidates name, the party, any other relevant information (such as address) and the vote indicator bubble. This makes it absolutely clear to the voter that the vte indicator bubble in the box next to that name is the vote indicator bubble for that candidate. The border (see 3.15.1), the contest title (see 3.15.2) and the contest instructions (see 3.15.3) with their distinct borders, colors and shading serve as a definite border between contests. This should eliminate any confusion to the voter that the box around the candidate's name, party, address and vote indicator bubble is a separate race.

I have to agree that the design of the optical scan ballot with the instructions and color looks great, however I also have to agree that they would be very difficult to implement with the limited ballot space we have, conflicting state laws on ballot design, and limited funds.  In primary elections now we are running the largest ballots that we can get through our M100's and AutoMarks.  While I realize that the convience and cost to the agency takes a back seat to making sure the voter has the opportunity to cast their ballot accurately, making maximum use of the ballot and, if at all possible, keeping the ballot length to one ballot card is also very important.  I cannot recall any instance were voters misunderstood the ballot as it is now designed for our state - while we can improve the design we have to take into consideration election budgets.

My comments are more general in nature. I may be repeating what others have said, because I haven't read through all of the comments submitted so far. I really appreciated how you defined the project purpose and the methodology used. It is helpful in knowing the logic for why certain strategies were recommended. I also appreciated the limited scope of 4 areas for review. My one concern would be limiting the use of color on a ballot for instructional purposes only. In my county we use color for multiple reasons, i.e., legislative districts. We do use color to assist with instructions, but a broader definition for the use of color would be recommended.

We have used numerous strategies that have been suggested in this best practices report.  The feedback from voters on the use of directional arrows, large font, signs, symbols and color and comment cards is generally positive.  

Ballot space, the final frontier.  While this section contains great principles and goals, I would agree with some of the other comments voicing concern over the amount of space available for instructions.  Multiple ballot cards are a bear to balance and audit, not to mention the expense.

Best practices are always appreciated unless they become de facto standards.  We need to be careful of setting unattainable expectations.  If the researchers would consider the realities of crowded ballots and tight budgets along with the overarching goal of informing voters, we will better manage those expectations.   

I was amazed at how easy these ballots were to understand using the pictures in the instructions and the colors for the different office headings.    I have two concerns; one being the additional cost of printing with the color and pictures and the other is the additional space used to accomplish these things.   The additional space used would not allow us to get all races and ballot propositions on one ballot in General Elections.   Maybe a shortened version could be used on the ballot and a larger version could be posted in the voting booths using the pictures during General Elections.

I like the readability of having issues and measures read across multiple columns instead of single very long entries.  If thurst is still reading this, would this work with our state practices?  I also enjoy the use of color but not sure how expensive that will be vs. current black and white.  Costs are always a very real issue in smaller and more rural jurisdictions.

Space on optical scan ballots is precious and these recommended examples waste way too much of it.   I doubt many jurisdictions have room to eat up a column of an optical scan ballot with instructions.  Instructions can be printed either on a ballot stub or on the secrecy sleeve.  Placing the candidate's party on a separate line also wastes space. 

Recommendations in this document must be reasonable.  Studies may show these examples are the most easy to read, but the resulting three or four pages per ballot would cause our costs to skyrocket and present tabulation issues I don't even want to contemplate.  

I have no background in graphics which qualifies me to disagree with the folks who prepared this, but my common sense and election experience tells me there is not sufficient consideration of the practical implications of some of these recommendations.  We need guidance that helps us prepare a ballot that is easy to read and affordable (and tabulatable, if that is a word)...not just one that is easy to read.

Section 3 Optical Scan Ballots.  I would like to see some recommendations for best practices for primary ballots.  Specifically for states with open primaries where the ballot contains candidates for all major parties.  We currently place each party in their own column and with 3 major parties couldn't use the first column for instructions.

I would like to see research on the impact of instructions in different locations between the primary and general ballot.  Is it better to be consistent for both elections to have the instructions across the top, or is it better to have the instructions across the top for the primary if necessary and use the left column for instructions for the general election?  Is there any research/recommendation for open primary ballots with the parties starting in mid-column after the last candidate of the previous party vs. starting each party in their own column?  Is there any research/recommendation for instructions for open primary ballots?

The write-in format and instructions are great!  Iowans have the right to write in anyone they want for every office on the ballot, but our instructions (which are specified in the law) have never been as clear as the examples I found here. 

On page 3.2, the 5th "key criteria" is that ballots with larger fonts and more pages were more usable than those with fewer pages and smaller type.  In Maryland, we typically use one ballot card (with printing on front & back).  Does the recommendation to use a larger font and therefore have more ballot cards hold when the decision is between using one ballot card with smaller font or two ballot cards with larger font?  From an election administration perspective, managing two ballot cards is more difficult, more costly, and more wasteful.

Also, the various sample ballots in the report use "Vote for 1" and "vote for one."  Is there a preference between the two instructions (i.e. spelling out the number or using the numeric character)?

Pages 3.51 and 3.52. The examples do not always have lines between the candidates. The example should be correct.

Page 3.6. Strike the first use of the phrase "are based on" on the 3rd line following the heading "Design: one language".

Page 3.50 on the third line in the first paragraph change the word "a" to "an".

Page 3.43 in the 3rd paragraph under the heading "Discussion" strike the first instance of the phrase "The use of political".

Page 3.4

Add a colon at the end of the first line under the heading "Planning goals".

In the 4th paragraph under "Planning value by role/response" change the phrase "online translations online" to read "online translations are often".

In the last paragraph uner "Planning value by role/response" change the phrase "accurate are" to "are accurate".

To SWeir:

Thank you for the time you are investing in helping EAC produce Best Practices for Effective Designs in Election Administration.  I am responding to your question about whether or not EAC has issued a Best Practice that encourages separate one language ballots.  EAC has not issued a Best Practice on that topic.  However, through our work on developing Voting System Management Guidelines and the study conducted by Design for Democracy that produced the draft document posted on this site, we are considering all aspects of that issue.

My question to you:  Do you have a recommendation about the use of one language ballots?

Gracia Hillman

Simple question, I read in the introduction to this section that a "one language" per ballot is preferable.  It goes on to state that two language ballots are possible and such possibilities are demonstrated.  Is there an EAC Best Practice that encourages separate one language ballots?

I ask because I have two languages, and we were forced into a two (19") ballot option in November, 2006.  We found this to be difficult for all concerned.  Additional postage, two long ballots to navigate, and since we have to hand process folded ballots (back fold ballots prior to tally), I think I want to go to a single language ballot model.

2.24 General Information and 2.25 Special Voting Rules are both noted as HAVA required.  Understanding that the majority of states still conduct poll elections, these are both written for poll elections.  The language does not work well for Oregon with all mail elections.  Would these files be able to be amended by Oregon to conform to mail elections as indicated in 2.9 for files ending with ".indd" or files ending in ".eps"?  And if not, will there not need to be such forms developed for such differences between poll and mail elections?

2.10, 2.20, 2.34, & 2.44 all have an example of 140 Vote by paper ballot (with arrows). The examples do not include a line being drawn.  From a county that uses arrows, if it must be clear that someone is to connect the front and back of the arrow with a single line.  In the example shown for bubbles, it does show shadowed area for filling in the arrow. 

I think this information will be extremely helpful to many Election Administrators in my State that do not have access to a printer.  The graphic instructions are excellent.

Section 2. There are recommendations that some signs/posters may/should be hung from the ceiling. While I agree that this may be the case, I do want my poll workers doing that because there are too many things that can go wrong. I suggest that the document include warnings as to the danger of hanging things from the ceiling and the possibility of damaging some types of ceilings.

Pages 2.50 and 2.51. Put periods at the end of the 1st line following the heading "State seal" and after the 1st sentence following the heading "Discussion".

Page 2.38. Put periods following the 1st line following the heading "Background", the 2nd line following the heading "Precinct/Ward" and at the end of the 3rd paragraph following the heading "Discussion".

Page 2.32. Strike one of the phrases "are based on" on the 2nd and 3rd lines of the paragraph under the heading "Design: two languages".

Page 2.28. Add a period at the end of the 2nd paragraph under the heading "Discussion".

Pages 2.23, 2.24, 2.25, 2.47, 2.48 and 2.49. Replace the comma with a period at the end of the line under the heading "Background".

Page 2.18. Under the heading "Discussion" - left justify the word "Primarily" at the beginning of the second paragraph and add a period at the end of the 4th paragraph.

Page 2.12. At the end of the fourth paragraph under the heading "Discussion" change the comma to a period.

Page 2.8. On the second and lines under the heading "Design: one language" strike the first instance of the words "are based on".