ContextualInquiry-Group:PollPrecision

From CS160 User Interfaces Fa06

Jump to: navigation, search

1 Contributors
2 Target Users
3 Problem and Solution Overview
4 Contextual Inquiry - Interview Descriptions
5 Task Analysis Questions
6 Analysis of Tasks
7 Interface Design
- 7.1 Sketches
8 Analysis of Approach
9 Images

[edit]

Contributors

Keenahn Jung
- Interviewed Nate (a survey designer) with Hiroki
- Helped develop interface design
David Eitan Poll
- Interviewed two voters (Mr. and Mrs. Schwartz), helped come up with Interface Design, and wrote "Target Users", "Interview Descriptions (of my Interviews)", "Problem and Solution Overview", and "Analysis of Tasks" sections.
Hiroki Terashima
- Interviewed three potential users and took notes during interview (Honors Society Nerds with Eric, Nathan Timberlake with Keenahn, and retiree with Eric), coordinated division of tasks/meeting times, wrote "Honors Society Nerds", "Similarities/Differences between interviews", "Task Analysis", gave input to various other tasks
Eric Vacca
- Interviewed two potential users (Honors Society Nerds and retiree), wrote "Analysis of Approach", "Interview Descriptions", helped with the "Task Analysis Questions", drew the storyboard of the voting process, and created a potential Voting Ballot.

[edit]

Target Users

Our initial assumptions about are target users are listed here.
Our target users fall into two categories:

Surveyors, Pollsters, and Data Miners
- These are the people who seek to acquire data, for whatever reason
- Most likely the direct client of our software
- Need:
  - System for easy and accurate collection of data from survey subjects and voters
  - Integration with data mining tools such as database software or other common scripting applications
- Want:
  - Verifiability of results
  - Highly customizable ballot templates
Voters and Survey Subjects
- These are the people who will respond to requests and surveys from pollsters and surveyors
- Most likely users of the system
- Need:
  - Method of data entry that will allow them to effectively communicate with their pollster
- Want:
  - Reassurance that their submissions were successful and received with high levels of accuracy
  - Easy data entry that doesn't requrie them to learn something new in order to provide feedback

We chose to interview the following for this stage:

Survey Administrator
Survey Designer
Voters
- Elderly
  - Fading vision, slower comprehension, very proud, politically active ideals/upbringing
- Mid Life
  - Engineer-type who considers himself to be tech-savvy
  - Highly educated woman who considers herself to be computer illiterate

The Survey Administrator and Designer fit our target user groups well. We intended to find out what their concerns are in doing their jobs, so that we might facilitate them through our system. By talking to voters, we hoped to identify potential weak points in current voting systems. We spoke to voters with a variety of handicaps and aptitudes to get an idea of the range of users we will need to cater to. Voters and Survey-takers are a very large group, and we need to have a design that will suit the vast majority of them, so a high variety of users need to be considered.

[edit]

Problem and Solution Overview

PollPrecision aims to provide electronic voting and polling with verifiable results and a shallow learning curve for voters and survey takers. Using the Anoto Digital Pen and Paper system, PollPrecision will provide voting and surveying that produces ink-on-paper records while providing the benefits of electronic counting and analysis. It reduces the overhead of data entry for users and lets them have quick and accurate access to their data. The ink-and-paper byproduct helps to ensure that black-box tampering with the system does not ruin the integrity of the survey or vote, and the use of pen and paper allows voters and survey takers to use an interface they are already familiar with, reducing confusion.

[edit]

Contextual Inquiry - Interview Descriptions

Interviewing process:

Prepare interview questions.
Designate one team member to ask questions, and the other to take notes.
Ask user to enact the process that he usesnormally.
Keep the user focused on the task with pointed questions.

[edit]

Retired Voters

After calling all the local retirement homes, we were able to find a retired person who still goes to polls to vote (most retirees file absentee ballots). Our team showed up on Bancroft And Sacramento at the Berkshire Retirement Home. We were escorted to the retiree's room, where she sat on her bed and we sat opposite her and conducted our interview. We were forced to use this environment because being almost 90, she has limited mobility. Since we were unable to enact her process of voting, we presented various ballot designs from past elections, while asking her questions on how she interpreted its design.

The retiree we interviewed, was a 89 year old woman, who worked as a stewardess. She was raised emphasizing the importance of voting and has never missed an oppotunity to have her voice heard. She seemed to be very confident in her actions despite some of her failing senses. She has never used an electronic voting machine, and is only familiar with pen and paper and punch card ballots.

Key Points:
- The retiree was blind in one eye and had to use a magnifying glass in order to read the ballot.
- In examining concise questions she usually had very little problems with comprehension
- For ballots with many choices (california recall ballot) or a small font, the retiree encountered problems.
- The retiree felt that it takes a lot of effort (even more in her old age) to stay informed.
- If the retiree ever felt like she didn't understand an issue, or was uniformed about a candidate, she would normally choose not to vote, although she admitted to making wild guesses on occasion.
- She never goes over her choices, or has had to fix a mistake during the process of voting.
- Before entering a polling station, she always has a pretty good idea of the selections she is about to make.
- She has never recieved a receipt for voting, but is confident her vote has always counted.

[edit]

Middle Aged Voters

Both Mr. and Mrs. Schwartz were interviewed in their home. We lacked the facilities to actually observe them in the process of voting, but came up with a strategy that we think helped us gain some insights. They were asked to walk us through the process of voting in the last election from the time they got out of their cars at the polling place until they left the polling place. Through questioning, we were able to get a detailed description of each of their tasks.

At any time where Mr. or Mrs. Schwartz were unsure of what the process required, we probed them with questions to learn more about their confusions.

Note: "Schwartz" is a pseudonym to protect the identities of the interviewees.

[edit]

Mr. Schwartz

Mr. Schwartz is an engineer who considers himself to be tech savvy. Going into the polling place, he felt confident that he would be able to successfully vote using the electronic voting machines. Retrospectively, he feels that he was indeed able to accomplish his tasks, but felt that there were problems with the interface that forced him to be especially diligent in checking that he was voting correctly.

Key Points:
- Mr. Schwartz found transitioning from issue to issue and position to position using the electronic ballot was unnecessarily difficult, since it forced users to scroll through every selection on the page to reach the "Next" button.
- Mr. Schwartz also felt that too often it was difficult to see all of the candidates or choices. When he voted for governor of California, and there were more than 100 candidates, he had to scroll through many pages of names to find the person he wanted to vote for. Every time he changed pages, he lost the context of the vote that he was trying to cast, and so he found this process to be quite frustrating.

[edit]

Mrs. Schwartz

Mrs. Schwartz is a highly educated (two doctorates), middle-age woman, but considers herself to be computer illiterate. She used an electronic voting machine in the last election, and had some axieties about the process.

Key Points:
- Using a dial for entry was cumbersome, as Mrs. Schwartz was unfamiliar with this type of control. She was able to make her selections successfully, but felt that she never received enough feedback to tell her that what she was doing was correct.
- Mrs. Schwartz went into the voting booth with a sample ballot to help her remember what she wanted to vote for.
- Mrs. Schwartz was especially anxious that her vote wasn't counted. She missed the feeling of actually placing her ballot in the ballot box. The confirmation screen on the electronic ballot did not reassure her that her vote was submitted, only that her selections had been made properly. To this day she is not sure that her vote was actually committed to the computer system.

[edit]

Surveyors, Pollsters, and Data Miners

We interviewed a very savvy survey designer, as well as some honors society kids who had experience administering and collecting surveys, but no design experience.

[edit]

Nathan Timberlake

Nathan is a grad student who actually did a dissertation on the importance of question ordering in surveys. We thus consider him to be an expert in this field. He gave us some very valuable insight into what is needed to design/administer a good survey. We interviewed him in his office where he was able to draw pictures for us on the whiteboard. He was not allowed to show us the actual interface of the tools used for designing the surveys (WISE).

Key Points:
- Survey design
  - Need to be able to maintain a question repository, so that surveyers can re-use questions. Need to be able to search past surveys.
  - Need flexibility in layout design.
  - Need a large variety of answer types.
Administration:
- need to authenticate users, but preserve anonymity
Data collection:
- SUrveyers need raw data, which they can then play around with and create aggregate data,

[edit]

Honors Society Nerds

The Honors Society Nerds distribute what is called "course evaluation surveys" at the end of the semester. They don't create the surveys themselves, and do not get to keep the surveys. Instead, they give the filled out surveys to the department, and in return they get data back from the surveys. They then post this data online on their website to help students see which courses are difficult and which TA's are good.

We interviewed two nerds in their office where they described to us the gist of the survey process and where they fit in (see above). They also provided us with sample survey forms: a course evaluation form and TA evaluation form. (Seeb images below)

Key Points:
- Survey design
  - Not in charge of survey design
- Administration:
  - Need to contact professors beforehand to figure out what time is best to administer the surveys
- Data collection:
  - Receives data back from the department so that they can post it on the web
  - Be careful not to post results of GSI's who don't want their results shown to the public

[edit]

Comparison

We found that all the interviewees were very concerned with the correctness of the votes/data. All of the voters were concerned that their votes will be correctly counted, and the survey users (Nathan and Honors Society Nerds) were concered that the questions on the survey are clear so that the survey takers would have no misinterpretations. They all agreed that the questions/directions should be easy to read and understand.

The retiree was different from the rest of the interviewees because she was blind in eye and needed magnifying glass to read the ballot. She required special services (time, magnifying glasses, assistance) to vote. Also, she considered voting a big priviledge and thus never missed to vote.

Nathan was the most vocal about the way the questions are being asked in a survey to insure accurate responses from the survey takers. He was especially concerned with making the data that are retrieved from the survey easy to access, organize, and analyze. He had a lot of opinions on the matter of survey analysis because it pertains to his profession.

The Honors Society Nerds did not have much control over the surveys- they only got back the data from the surveys. However, they had to be careful with what data to present on their website, as some TA's and GSI preferred that their evalution not be posted online.

[edit]

Task Analysis Questions

All voters will use this system. People over 18 residing in America. It is assumed the voter has some knowledge and is at least literate.
The task varies from county to county, and state to state. Recently people have been adapting to electronic voting. Most voters go into the voting area with a good general idea of who to vote for in mind. However ancient legislation specifies the voting ballot and the layout often leaves voters confused. For example on ballots, the candidates are presented in a random alphabetical order, whatever that means. More recent problems have come with the sensitivity of touch screen voting machines. I voter pushes the “Next Screen” button, and moves 2 or 3 screens. In the 2000 presidential election, elderly voters in Florida had to distinguish if a vote for Kerry was the second hole down or the 3rd hole down and it was somewhat ambiguous.
Ideally the voter should be able to walk into the voting booth with a general idea of their selection and make the selections with confidence. Most voters even bring a sample ballot into the poll station. Voting is a simple process and has been stupidly overcomplicated. The task should not confuse the users, and allow the users to find their candidates with ease.
There are various sources where the user can turn to for guidance on voting. The voters can learn to vote from their parents, friends, assistants at the voting booth, and online.
Volunteer voting stations across the nation. Usually schools, churches, and other public places. The retirement home we conducted one of our interviews is also a voting station. "Absentee voting" allows immoble voters to vote without having to go to voting stations.
For voting, the user IS the data, so any process that occurs limiting the user from understanding, hinders the process.
Tools that the user can use range from punch card ballots, to electronic voting ballot machines, to mail in ballots.
While voting, users are prohibited from communicating with each other. Voters can communicate freely before and after they vote.
The voting takes place whenever there is a need to select a candidate for a position. This could imply local mayoral voting, state legistration, as well as voting for the president of the United States. They all occur on different schedules.
We have not found any evidence of time constraints on voting. The only constraints would occur if the user is unable to decipher the interface, and begins to feel ignorant while other people are waiting to cast their vote.
When things go wrong, votes are miscast, uncounted, and unrepresented. The result is a nation of apathetic youths and adults alike, who are slowly losing faith in democracy and the prospect of having a voice in the democratic nation.

[edit]

Analysis of Tasks

Easy
1. Make selections on a ballot or survey
  - Voter or survey taker makes selections on a survey or ballot. He weighs his options, and responds to various questions that have limited choices (a list of names, a place to write-in, yes/no, etc.). Selections are made by punching the ballot, writing with a pen, or interacting with a digitally displayed interface.
2. Submit a vote or survey
  - After making his selections, the voter or survey taker must submit his vote or survey for counting and analysis. He does this by completing a wizard on a computer and/or handing his paper ballot or survey to the administrator.
Moderate
1. Analysis of results
  - The pollster or election administrator needs to be able to do something useful with the results of the survey or vote. He must count votes, acquire statistics about surveys, and interpret written information. Transmission of this information might be digital or constrained to compiling information from raw ballots and survey submissions.
2. Correcting errors
  - Pollsters and administrators need to be able to cope with errors made in casting ballots or submitting surveys. In elections using paper ballots, improper chads must be considered and counted based on certain arbitrary rules for qualification. Surveys that have been filled out incorrectly might still contain useful data. It is up to pollsters and administrators to resolve, based on their own critereon, errors made in voting and survey-taking.
Difficult
1. Create a ballot
  - Ballots and surveys must be designed prior to administration of the poll or election. These might be designed in any sort of publication application ranging from Microsoft Word to Publisher, etc. Ballots have fields that must be specified. For example, holes to be punched must be specified in punch-card ballots. Digital interfaces might have buttons or controls that the user can interact with to make their selections. In all of these cases, the ballot-maker must somehow make it clear to the voter how to make a valid selection.
2. Verify results
  - Voting, in particular, requires a certain level of verifiability. Usually, this involves recounting votes for a particular region. Recounted votes need to be reprocessed or hand counted, helping to ensure accurate results. Hand counting involves poll workers visually examining votes and manually entering new data.

[edit]

Interface Design

We determined five basic interaction points with our system and came up with the initial UI design for them.

Designing the Ballot:
- Poll designer uses favorite text editor to produce an XML document with the required questions.
- Using our custom tag libraries, designer adds answer fields.
- Answer fields can be of the types:
  1. checkbox
  2. choose-n (radio button is choose-1)
  3. write in (text field) which can be OCR'd, or left as a blob
  4. rank ordering
  5. continuum (line scale)
- We will use QTI (Question and Test Interoperability, the XML standard for surveys) tags. We chose XML as our file format for several reasons. This will allow us to easily archive questions in the future (keep a "question repository"), search our questions/surveys, track metadata, etc. In the future, this may be transparent to the poll designer, who would use a GUI to design the survey, and just export the file to XML.
- Poll designer prints the survey on Anoto paper, with unique page ID's.
Voting:
- As voters walk into the voting area, after handing the poll worker their Personal ID, they are handed a pen and the poll worker checks off the name on the main polling computer with a list of registered voters. They sign their name on a “sign in” sheet giving double verification of the voter and thus ensuring that they do not vote twice. They will be identified initially by their name and address (unique identifier) but this will be recorded as a hash of this ID. This hash will be associated with the penID until it is docked again. Poll workers will not be able to map the hash back to the name.
- Next, they walk into a voting booth, where there is (optionally) a computer monitor.
- They read over the ballot, then begin filling it out. The computer monitor displays instructions, indicates mistakes, and shows a summary of their voting so far. The display will look like a ballot, but will show electronic check marks (or text for write ins), not handwritten strokes. This computer does not record the vote for the voter, this happens later. It is merely provided as a convenience for the voter.
- After filling out the ballot, the voter drops the ballot into a box and hands the pen to a poll worker. The poll worker docks the pen and the vote is recorded.
- The voter receives a printed receipt, acknowledging that their vote was recorded.
- If the voter makes a mistake during the voting process, the monitor in the booth would warn him and advise him to get another ballot. Only the last ballot written on counts.
Adding the vote to the computer
- Our system will add the votes to a DB once the pen is docked.
- A receipt will be printed.
Resolving conflicts/verification
- human interpretation of voting errors and write-ins
- in the case of a recount, paper trail allows greater accuracy
- After the vote is recorded, the system with detect and alert poll workers to potential errors. The poll worker will be able to compare the actual strokes recorded by the pen vs. the recognized input.
Examining the voting results:
- Simple counts/strokes (summary, high level info)
- Raw data in DB
- Poll workers will be able to access raw strokes if necessary.

[edit]

Sketches

Dataflow diagram:

Voting Storyboard:

Sample Ballot:

[edit]

Analysis of Approach

Our application takes advantage of many of Anoto's affordances. People are most familiar using pen and paper as an input device and so now when voting, the voter need not worry about a clunky touch screen interface, or confusing punchcard ballots. The direct digital input makes the menial, but important, data entry of elections/surveys, into an automatic process. Using the Anoto pen's unique ID numbers, privacy issues and vote verification are both easily handled. Also because the Anoto system uses (nearly) normal paper, ballots can potentially be both unique to each voter, and mass produced.

Electronic voting has been a hot topic of late, and seems to be the front running candidate to take the place of punchcard and pen and paper ballots in the near future. In fact a law has been passed that by 2006 every state will have to have converted to some sort of electronic voting mechanism. Our voting system is markedly better for many reasons:

Many voters, especially older ones, have little to no experience with visual user interfaces, let alone a touch screen. What may seem trivial to many voters, is impossible to a large base of voters. Using pen and paper, a medium that everyone is familiar with, eliminates this technological gap.
Electronic voting stations are large and difficult to set up, while our system is portable and easily set up by anybody.
Electronic voting systems are non-transparent. They often use in house software that requires an expert, while our voting system works easily with any PC.
Each voting machine must have its counter reset and touchscreen callibrated before every use, while our system requires very little configuration.
Write in votes must be taken seperately, but with our system the voter can write directly onto the ballot.

Other portable writing devices like the tablet PC could very well be used for voting as well, but are quite expensive, and have processing power beyond what is needed for the simple task of voting. The stylus interface is also foreign to many people and takes time getting used to.

Summary:

Pros: