operation of the No on 8 campaign recruited and mobilized quite possibly the largest number of volunteers of any LGBT campaign of any kind in California or in U.S. history. Despite its impressive size, however, it had a limited impact on the election results because of a mistake in focus.The field
The No on 8 field operation shows the potential of the LGBT community to do effective voter contact in a future campaign. With better focus, it could affect the outcome.
Sometimes overlooked in the face of the loss on Prop 8 is the remarkable number of volunteers recruited by No on 8 and the sacrifices the volunteers made. We should celebrate both and remember that the LGBT community and our allies are ready, willing, and able to work hard in the cause of LGBT equality. The terrific specifics are in the sidebar.
The achievements, however, had limited impact because the design of the field plan was flawed. It prioritized identifying supporters over engaging persuadable voters not yet with us. In a high-turnout presidential election year where so many of our supporters were already highly likely to vote, this was not a logical choice. Nor was late-stage sign-waving in our areas of greatest strength.
Field’s unique strength is the ability to have two-way conversations with voters. Maximizing these has the potential to help future campaigns both gain insight into voters’ thinking and win more votes.
Looking at No on 8’s Field Campaign
The scale of volunteer involvement represented an unprecedented outpouring of support from the LGBT community and among our allies. It only happened because a highly skilled field staff did the painstaking but necessary one-on-one recruitment on the phone and in person, not relying on e-mail or the Web or waiting for people to show up. This thoughtful cultivation of volunteers created the largest field effort ever not only in California’s LGBT electoral history but anywhere that the LGBT community has faced a hostile ballot measure.
Once No on 8 got volunteers engaged, they were asked primarily to phone bank: they called voters to identify supporters, and they called potential volunteers to ask them to come in and help. Other volunteer work included in-person volunteer recruitment in public places: on streets, at events, from organizations, at churches and synagogues, and from speaking engagements. Data entry, house party organizing, and office work (sign distribution, packet prep, etc.) made up a significant portion of the work as well.
Additionally, the field team raised $1,306,480 through systematic asking for money at all volunteer actions.
No on 8’s general consultant recommended from the start that the field team identify 3% to 4% of the votes it would take to win from a persuadable universe of undecided voters. This number was determined to be just over 200,000 voters.
Between the middle of June and the third week of October, the field operation therefore focused on calling a micro-targeted universe of voters who were deemed persuadable undecideds. Little persuasion was involved for two reasons.
First, volunteers spoke to few undecided voters. The targeted voter list provided to the field team overwhelmingly consisted of voters already planning to vote no. As a result, the field operation volunteers called, spoke with, and identified the views of 180,743 voters: 73% (132,245) were already planning to vote no, 11% (19,061) were undecided, and 16% (29,437) were planning to vote yes. The 11% of the list that was undecided was about the same as the incidence of undecideds among voters as a whole; the campaign could have just called a random list of voters and found about the same number of undecided voters. It may be that creating a voter list with more undecided voters would not necessarily include the types of undecided voters open to persuasion to our point of view. But calling the list provided was extremely inefficient—at least 89% inefficient—if the goal was persuasion.
The practical effect of this failure was that the body of work accomplished by the No on 8 campaign’s field operation essentially amounted to making a list of supportive voters that could be turned out to vote, which would generally not be considered a top priority in a high turnout election.
Second, volunteers were given a phone script that was not designed to persuade voters. The section of the script having anything to do with persuasion was minor. The two paragraphs below are what volunteers said to voters after they were asked how they felt about marriage for same-sex couples and how they would vote on Prop 8:
[IF UNEASY or UNSURE] I hear that. What we’re hearing from a lot of people across California is that regardless of how one feels about marriage for same-sex couples, it’s wrong to take away anyone’s fundamental rights. Many Californians agree and don’t want to single out one group to be treated differently. On this important issue, we need all fair-minded Californians to vote NO on this unfair proposition.
[IF UNSURE of Vote] It sounds like you’re really thinking about this. As the Election nears, may we send you some more information about Proposition 8? What’s your e-mail address? And is this your best phone number? Can I give you the Equality for All Web site? It’s www.NoOnProp8.com. We need all fair-minded Californians to vote NO on Prop 8. Thanks so much for thinking about a California where no Californian gets singled out for different treatment. Have a great day!
The complete voter contact script is in Appendix P. It was essentially unchanged throughout the campaign.
If the field campaign had continued with voter identification phone banks until the end of the campaign, the field team would have likely reached its goal of identifying 200,000 supportive voters. It is not clear, however, how many additional votes this might have produced for No on 8. In a high-turnout election year, identifying supporters is not as vital as it is in a year when many might fail to vote. Not voting was unlikely in 2008 for many of these voters.
At the direction of the new campaign decision makers who took over in late September, however, the field operation stopped having its volunteers call voters during the last nine days of the campaign. The field operation leadership was informed that a separate paid phone program would continue to call the remaining universe and do any Get Out the Vote calling. The field team was instructed to move volunteers into the street to do visibility actions in supporter/base turf to clarify wrong-way voting and mobilize supporters to go to the polls. The Election Day action was part of this strategy, where field volunteers filled 11,000 shifts at 1,500 predominantly supportive polling locations throughout the state.
The new field tactics were no more closely tailored to the situation at hand than the ones they replaced. Wrong-way voting was a problem but not among all pro-marriage voters. The data show that wrong-way voting was not concentrated in the neighborhoods where most volunteers were placed on Election Day, where awareness of the issue and the election were highest. Quite the opposite; see the discussion of this topic in Appendix K.
Voter turnout was even less likely to be affected in a record-setting presidential election year like 2008, where voter turnout is already at its highest. Perhaps the campaign intended field at this point to have a much more limited effect, to boost the morale of the larger LGBT community (though it was not a morale boost to the field staff and volunteers already involved in the campaign, many of whom viewed the new activity as largely a waste of time). While many campaigns make this kind of choice with their field operations, it is wasteful. It yields a very low return in votes for so much volunteer capability, time, and energy.
Field in Future Same-Sex Marriage Campaigns
Field campaigns rely primarily on three types of tactics: (1) direct communication with voters (eg in order of decreasing value and increasing ease to get to scale, door-to-door canvassing, phone banks, and literature drops); (2) recruitment to build a big enough team to accomplish direct communication with voters on a big enough scale (eg in-person volunteer recruitment at an event or a high traffic public venue, recruitment phone banks, new media and email recruitment); and (3) actions that energize the campaign’s base (eg visibility events, rallies, house parties, yard signs). In a campaign where the goal of the field campaign is to affect the electoral outcome by securing a specific number of votes, most resources must be spent on the first two tactics: voter contact and volunteer recruitment. The third can easily be the least strategic; the morale boost it provides is certain, the production of additional votes much less so.
Part of the Obama campaign experience may be applicable to our situation.
The Obama field campaign illustrates this. Its principal focus: having volunteers in base communities call persuadable or swing voters. This is a typical tactic used when a campaign has a large universe of voters to contact in key geographic regions, and a high concentration of volunteers in different geographic regions. In California, Obama volunteers were mainly used to call voters in states unlike California, where the election was going to be close.
The No on 8 campaign faced the same dilemma on a miniature scale as the Obama campaign, and it turned to the same tactic. No on 8 asked its volunteers in the parts of California where the LGBT community and our allies have a large base (such as San Diego, Long Beach, Los Angeles, Sacramento, and the Bay Area) to call micro-targeted voter universes throughout the state.
The largest difference between the two phone programs was that the Obama model had a larger, more publicized phone-from-home program. However, No on 8 had a smaller phone-from-home program created to offer the opportunity for volunteers to participate in phone banking if they lived a certain distance away from a campaign office. A phone-from-home program offers the benefit of more people being able to get engaged in the work regardless of where they live.
However, phone-from-home also has three drawbacks that moved the No on 8 field team to do as many calls as possible not from home, but instead from campaign offices.
First, No on 8 recognized the lack of support a volunteer has when doing remote calls. In a candidate campaign, the worst experience that can befall a volunteer is occasional defamation of the volunteer’s preferred candidate. Although unpleasant, this is less traumatic than the worst that occurs when a volunteer is making calls about same-sex marriage. It’s harder not to take unkind calls personally; and the frequency of emotionally taxing calls can be higher. By contrast, when a team of volunteers get together in a collective space, the calls become much more tolerable. This is important, not just because a campaign needs volunteers to keep coming back throughout the campaign, but also because it is important that we do what we can to minimize emotional distress during an already damaging campaign.
Second, No on 8 noticed that phone-from-home offered the campaign less quality control and provided the volunteers less support and training. This may matter less when a campaign has a very consistent, relatively simple message (eg you are a Democrat; vote for the Democrat). However, in an issue campaign like marriage, when there are specific messages the campaign wants said or avoided in response to the opposition’s attacks, on-the-spot training and oversight helps volunteers navigate the more complex calls.
Third, No on 8 prioritized volunteer retention and leadership development. Presidential and other high-profile candidate campaigns often achieve good retention rates and get volunteers deeply involved through the celebrity factor inherent in their campaigns. It’s not automatic, but it’s easier. While an in-person request for the volunteer to return and to sign up for a specific future shift is valuable in all campaigns, it is indispensable in issue campaigns. In No on 8, the field team’s experience was that it was hugely important to have an organizer check in with each volunteer as they were wrapping up, ask how the calls went, and then make a strong recommit ask. When an organizer took the time to evaluate the experience with the volunteer and then asked the volunteer for additional help, many more No on 8 volunteers not only returned, but also chose ongoing and larger roles in the campaign.
While it is arguable how necessary it was to have volunteers predominantly call from central offices during the No on 8 campaign (since the main phone tactic was simple voter ID), in any future campaign that involves voter persuasion, it will be crucial to have a large, well-trained, well-supported team of volunteers who return repeatedly, develop expertise, and take on increasingly demanding leadership roles. A future campaign is more likely to achieve this relying minimally on phone-from-home and maximally on collective action that takes place with large numbers of volunteers coming together in central locations.
It is a terrific accomplishment that the No on 8 field team recruited, trained, and mobilized our community in record numbers. We now know that with a team of comparably skilled organizers in the field, the LGBT community and our allies have the ability to build a large-scale grassroots one-on-one voter contact operation.
What needs to improve in a future campaign is to get an even larger number and wider range of people involved between campaigns starting now, and to focus such a field team on the work most likely to meaningfully affect the electoral outcome.
No on 8 lost by 600,000 to 1,000,000 votes (including wrong-way voters). Under these circumstances, simply identifying 200,000 who already agree with us will not change the outcome. Identification of supporters may be very useful, depending on the election year, the likely turnout, and the role to be played by a Get-Out-the-Vote operation targeting those who otherwise might not vote. But turning out those who already agree with us will not be enough to win. We will also have to persuade a large number of undecided and unsupportive voters to stand with us as well.
Encouragingly, several voter persuasion experiments are now underway in California. It is beyond the scope of this report to assess all of the work currently happening in California, but see Recommendation 3 for a description of a few promising projects.
Just as with media, any future campaign will need to focus field efforts in areas that produce or retain pro–same-sex marriage votes. If door-to-door canvassing turns out to be an efficient and successful way to communicate with our base voters and/or key groups of undecided or persuadable voters, it would make sense that a future campaign would locate its field offices in areas with:
· Base voters who are potentially highly susceptible to the anti-gay arguments raised by our opponents;
· Unsupportive and undecided voters with whom we have the best chance of persuading to become supportive;
· A large concentration of potential volunteers.
Finding 1 offers data identifying many in the first category. We currently lack adequate information to describe the second category. It is possible, however, that current experiments may show that unsupportive voters who live in communities with more supportive voters are more moveable than unsupportive voters who are surrounded by voters who share their opinion. If this proves true, it would be a compelling reason to invest most field resources is in relatively supportive or evenly divided areas of the state.
The incompleteness of our current knowledge should also motivate us to use the time we have before and between campaigns to invest in communities that are unorganized or have been under-organized previously to better understand their potential. It is likely that many communities have not been as involved in LGBT issue campaigns or educational efforts because they have not been asked to be or because they have not seen themselves represented in the work. Once a more remarkable investment is made, they may get involved in high numbers and meaningful ways.
On the other hand, some communities may simply not have the critical mass of pro-LGBT populations that would justify field offices and full-time field organizers. Even then, however, the next campaign will surely want to consider how to provide off-site support in every area and training to local leaders across the state as a long-term commitment to community building. We must create better ways for all communities to get involved in the campaign. It seems realistic that any campaign could and should devote a portion of field staff time to ensure that volunteers in areas without field offices have access and support to do the most valuable voter contact work. This could take the form of an organizing kit that can be e-mailed, online support, and/or dedicated staff and phone lines designed to help supporters in these areas have easy lines of communication with the campaign and actively participate in campaign actions and briefings. In addition, since campaigns have limited resources and can’t provide every type of support volunteers or all communities want, volunteers ought to initiate activities of their choosing without support from the campaign, and campaigns ought to feel good about uncoordinated activity. Few activities will do more harm than good.
Understanding the Yes on 8 Field Operation: A Weakness in This Report
There was no independent source of data that would allow evaluation of the scale and functioning of the Yes on 8 field campaign. This report would be stronger with such an assessment. Given the massive involvement of the Church of Latter-day Saints and religious institutions and networks in the fundraising for Yes on 8, it is a reasonable guess that their involvement with the field side of the campaign was also extraordinary.
Suggestive evidence to support this hunch can be found by comparing the vote on Prop 8 to the vote on the same day on Proposition 4. Prop 4 was the third recent vote in California on whether to require minors to notify their parents before they can obtain an abortion. Both Prop 8 and Prop 4 received a very similar number of No votes: 6,401,482 for the pro-LGBT No on 8 and 6,728,478 for the pro-choice No on 4. But the two propositions did not receive the same number of Yes votes. Yes on 8 received 7,001,084 votes, while Yes on 4 received 6,220,473 votes.
The much higher Yes on 8 vote—and the much smaller number of voters who simply skipped it—could result from the Yes on 8 field campaign, its paid media campaign, its success in the earned media, the generally higher visibility of Prop 8, the relative novelty of voting on same-sex marriage, or (most likely) a combination of some or all of these. A strong field operation has strong potential to reduce ballot drop-off; in most campaigns, it would be the first theory I’d consider to explain it. So while these numbers don’t amount to proof of an effective Yes on 8 field campaign, they offer strong warning that one existed in 2008 and could come to life again in future campaigns.