An exit polling idea for NJ gov. race - please help brainstorm!

[SmileMaker]

I thought I'd give this idea its own thread here

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=104&topic_id=4938080&mesg_id=4943965

An exit polling idea for NJ gov. race - please help brainstorm!


Excerpts from posts I added to BradBlog discussion about voting rights conference. Please help brainstorm this. We could start this in NJ and, if it works, take it National by 2006! Basically, the adults of both parties have shown that they can't be trusted with counting our votes. It's time for the kids to keep an eye on them.

Please help brainstorm this and if you know how to connect with people who design these databases, let me know! Thanks... Sue (aka SmileMaker)

...Sue said on 9/30/2005 @ 2:44pm PT...

Please talk about making sure NJ votes count for Corzine in November!!!!


I think I’m ready to take leadership on coordinating this in NJ because I don’t think anyone else is. I’m still asking around and the answers are coming slowly. I don’t know what the best strategy will be to be certain our votes will count - or if they are already magically in place - but, I have a couple of ideas if they aren’t.

1 – Mail in absentee voting.

Voter keeps a copy of ballot and sends a copy or reports it to a centralized NJ database (Maybe we could use a database already in existence?).

2 – Have math and social studies students conduct exit polls outside of polling places located in schools with help from teachers and colleges students.

This would turn the polling activity into an authentic learning experience that would provide an opportunity to apply math / interviewing skills to a real situation that students have an enormous stake in. Just let the right wingers try to steal an election while our kids are watching!

Let me know what you think – I hope it will be a topic at the conference this weekend. I’ll try to listen in if I can. I hope that someone will archive the audio so we will be able to listen later though. There is a lot at stake.

---------

COMMENT #6
...Sue said on 9/30/2005 @ 3:13pm PT...

Today, as over 100 of us marched through Princeton Univ. campus to protest CondiLiestoUs Rice, I had an opportunity to talk to some of the progressive student leaders and leaders of the Coalition about the voting issue. There are more people I need to talk to, but it's so hard getting answers!! Anyway, I didn't want to miss the opportunity to get NJ's gov. election this November onto your conference brainstorming agenda.

COMMENT #7
...Sue said on 9/30/2005 @ 4:17pm PT...



Me again... I think I figured out how to make exit polling for kids REALLY simple, completely discreet, more transparent and a fun learning activity for the kiddies

1- Make paper copies of the sample ballots and
deploy activists to work with students of math/social studies in their schools and at other polling places all over NJ.

2-Have kids give the voters a sample ballot to complete as it was completed in the booth. Have the voter fold it up (when complete) and put it into a box (that is painted black??)

3- Have kids, teachers and activist helpers count up the exit poll results and see if there is any discrepancy between exit poll percentage and actual vote percentage. Try to get a huge sample by getting the kids to be as charming as possible when giving sample ballots.

4 - Getting the kids involved gives them the stake that they deserve in THEIR future, helps them understand what an election is and why it's so important to make sure our votes are counted correctly, gives the vote counting process an air of transparency because, of course, we'll want to invite the local media in to show the kids taking an interest in the election and learning all about the math involved when they are finished.

..Peggy said on 9/30/2005 @ 4:37pm PT...


Brad, you are a true American hero. Once again, thank you.

Sue, I really like your idea about the students polling voters. Could this idea somehow be promoted nation-wide???

COMMENT #20
...Sue said on 10/1/2005 @ 8:03am PT...



Peggy -

I will try to round up enough people in NJ to participate. The Coalition for Peace Action in Princeton (peacecoalition.org) is serving as a network for voting rights activism in NJ. Our biggest push has been to help get Rush Holt's bill passed.

Ideally, a database like those used to set up house parties could be used to set up kids polls at polling places around NJ and to report the results. Linking up with PDA, DFA and MoveOn could help get the world out.

Teachers of math and civics could brainstorm lesson plans that could tie the activity into standards in a related forum. If it works, I would think that by 2006 we could have an interesting new check on our vote auditing process and engaging lesson plans for teachers to use that would provide an authentic learning experience for the kids.

I don't know how to set up those kinds of data collection systems, that is why I'm reaching out to the voting rights activists meeting this weekend to help with brainstorming.

From my experience with offering enrichment programs in schools and to after school programs, I know that change happens SLOWLY. Another complication (or expense) could be getting the ballots copied and in place. I'm thinking that highlighter pens could be used to mark the sample ballots so they wouldn't need to be adapted to make space for marking.

I will be unavailable tonight during the BradShow online blogging - I hope that this idea can be tossed around with people who could help make this idea a reality.

[Febble]

1. Suggestion:

Look at the BYU exit poll exercise.

http://exitpoll.byu.edu/

Their methodology is excellent, and they use student pollsters.

Also look at the San Diego exercise and perhaps learn from the errors there (my guess is that asking people to sign affadavits was probably a major problem).

Edit:

BTW the math is relatively easy (though there are some pitfalls). The hard part is getting a representative sample - the actual data collection.

[SmileMaker]

2. Thanks! Sample ballots completed privately seems most efficient

What I like about the sample ballot idea is that it can be kept anonymous and the voter can give an identical response to the one that they chose in the privacy of a voting booth.

Smart college students and professional pollsters can come up with other measurements. What happens though is that we get the 'reluctant Bush voter' excuse that serves to discredit the results of the survey.

When I have more time, I'll check out the BYU exit poll project. Thanks.

[Febble]

3. You are right.

And without wanting to be controversial, I think that there is good evidence that extreme care is required to avoid a Democratic bias in polling samples. The better your response rate the less your bias is likely to be, although you have to make sure that methodological procedures designed to improve response rate are not going to improve response rates from one group more than another.

You might like to contact the San Diego group for lessons learned. It seems very likely that they ended up with a biased sample, given the recount results.

One problem, I imagine, is presenting the project as non-partisan. After all, by far the most noise about election irregularities has come (with reason) from Democrats, and I suspect that a factor in the San Diego study may have been mistrust of the project on the part of Republican voters.

But anonymity is vital, as you point out! As is the assurance that the data will be kept confidential and not released in any form that could allow the respondents to be identified.

[diva77]

4. Check out parallel elections -

http://www.ecotalk.org/ParallelElections.htm

to quote from Lynn Landes:

Parallel Elections serve three purposes:

Introduces Americans to real voting

Provide evidence to challenge election results and request a new election

Pressure election officials to ban voting by machine, absentee, or early


Two major concerns expressed about this project have been - participation and signing the ballot.  First, it is not necessary to have 100% voter participation. If the Parallel Election (PE) vote tally indicates a candidate won over 50% of the votes at your precinct, but the official tally doesn't, then you can challenge the results.  Or, if we count more votes for any candidate, than what the official results show, then we can challenge. This could be particularly helpful for third party candidates. Challenges can be made on the precinct, county, or state level.


Second, if voters don't want to sign their ballots, they could sign a register or leave their ballot unsigned.  However, their ballot will not constitute direct evidence of how they voted, and therefore, will be less valuable to the project.


Anyone can organize a Parallel Election. A single person can conduct a Parallel Election at their own polling place on Election Day. Or, people can be more ambitious and organize their town, county, or state.  With every election, more voters will get the idea. If Americans want every vote to count, then Americans must be at their polling place on Election Day to count every vote.

[Febble]

5. That's interesting

So there's a bit of a damned-if-you-do, damned-if-you don't problem here.

The parallel election in San Diego got a discrepancy between the precinct totals and the signed parallel election ballots, which was why they requested a recount.

However, if they hadn't got the discrepancy, or if the ballots hadn't been signed, are you saying that they wouldn't have got the recount?

And yet it looks as though the discrepancy was due to a biased sample (because the recount matched the precinct totals). Which may have been due to the fact that people were asked to sign the parallel election ballots.

hmmmm.

Nice methodological problem to solve here!

[SmileMaker]

6. Great idea - I'll look into that

I met Lynn Landes a few years ago at a documentary showing of one of the election movies - Unprecendented I think.

Is anyone listening to the Voting Rights Conference in OR? It's streaming live at kpftx.org - direct link can be found at bradblog.com site.