Earlier this week, Nate Silver coined the term “Broadus Effect” to describe a phenomenon he was seeing with the polling on California’s Proposition 19. He noticed that polls done via automated polling were showing higher support for marijuana legalization than polling done via live operator. The difference was particularly stark for minority groups.
This got me thinking back to some of the discussions that were happening late last year around pushing a ballot initiative for 2010. At the time, the ACLU (and Alison Holcomb specifically) was arguing against putting a full legalization initiative on the ballot this year. Their rationale was that the internal polling they’d done was not showing strong enough support for it. In an email to me, Holcomb indicated that their polling showed support for legalization was only between 33-40%. I found that figure to be hard to believe (considering that 44% of Nevada voters supported legalization at the ballot box in 2006) and wrote up a post about it.
As I-1068 was formed, Holcomb and the ACLU remained convinced that a marijuana legalization initiative couldn’t pass. The I-1068 folks largely left them out of the planning and then later requests for their support ended with them making a public refusal to endorse it. This lack of support eventually doomed the initiative’s ability to raise money from other Democratic groups who otherwise saw big benefits from getting it on the ballot.
So this week, I emailed Holcomb about Silver’s post. And it looks like the ACLU is now re-evaluating their previous pessimism over their internal polling in light of the “Broadus Effect”.
UDPATE: Governor Gregoire’s office responds to the fact that legalizing marijuana is still the top vote getter on the website they launched last week to take suggestion on how to fix the state budget.
UDPATE 2: Alison Holcomb wrote to me directly complaining that I didn’t properly characterize her email response that spurred this post, so I’m posting her follow-up email right here:
Your question was, “I’m curious if you’ve thought about the ACLU’s previous polling on marijuana legalization with respect to what Nate Silver has dubbed ‘The Broadus Effect.'” Indeed I have, and I’ve compared the margins our polls show on hypothetical proposals to WA voters with those described in Silver’s piece on the CA polling of Prop 19. What I’ve seen hasn’t given the ACLU reason to “re-evaluat[e our] previous pessimism.” Instead, we are thinking about how best to do necessary follow-up research that might, in part, test the existence and extent of a “Broadus Effect” in Washington – assuming the actual vote in CA provides additional support for the theory. This is what I meant when I said in that same email, “And it’s figuring prominently in thinking about future qualitative and quantitative research.”
I’ve also been examining our crosstabs to see whether sufficient samples of various races existed to draw any conclusions as to where, for example, African Americans were as a group on the questions we asked. I’m interested in testing messages about the racially disparate enforcement of marijuana laws, how that contributes to the shame and stigma Silver identifies, and whether we can do effective public education around this issue in a way that helps us build a broader coalition of support that includes our communities of color.