Two more counties have completed their random sampling according to today’s update from the Secretary of State’s office; one large (Riverside, with a validity rate of 73.3%), and one small (Santa Cruz, with a validity rate of 60.1%). I note that Riverside has a very high validity rate for a large county (74,478 raw signatures). No county with more raw signatures has a validity rate higher than 69.5%. (Of course, we still have to hear from Los Angeles County.) The largest county with a validity rate higher than 73.3% is Ventura (27,134 raw signatures, validity rate 82.2%).
The overall validity rate is 67.6%, up somewhat from the 67.0% validity rate in my previous report. That gives a projected valid signature count of 769,154 signatures, up sufficiently from the 762,328 in that previous report for Six Californias to qualify for a full count. This bodes well for Six Californias as long as the remaining counties average roughly a 67.3% or better validity count.
Sixteen counties still have to complete their random sampling. (According to the Secretary of State, they have to complete the process by September 12th, a week from this Friday.) The top ten (by the number of raw signatures they reported) are now Los Angeles (311,924 raw signatures), Orange (52,217), Fresno (38,382), San Luis Obispo (12,906), El Dorado (11,649), Humboldt (7,230), Tuolumne (4,732), Nevada (4,322), Yuba (3,720), and Lassen (2,066). The top three have to check 3% of their signatures; the others have to check 500 (unless they want to check them all).
(Note that I had made a mistake in Report #11, carried over in Report #12, where I had included Kern county in the list of uncompleted counties even though they had completed their random sampling. I have corrected those previous reports and triple-checked the list in this report.)
–Steve Chessin
President, Californians for Electoral Reform (CfER)
www.cfer.org
The opinions expressed here are my own and not necessarily those of CfER.
The counties that have completed their sample check represent 60.0% of the raw signatures. Los Angeles represents about 2/3 of the outstanding raw signatures that haven’t had a sample check.
If counties were of roughly comparable size, 60% would provide pretty good projection for the remaining 40%. But Los Angeles is such an outlier in terms of size and diversity, projection is unreliable. When every 1% change in the validity rate projects to 3,000 signatures, Los Angeles is clearly decisive.
The collection rate (raw signatures divided by population) is quite variable, with the highest rates mostly in the northern part of the state. This might mean it is easier to collect signatures in smaller counties, or possibly support for the idea of Jefferson.
The collection rate may have some relationship to the validity rate. Marin has a very low collection rate (0.5% vs the statewide 3.1%), but a high validity rate (76.9%). The validity rate might also reflect the high income and education levels. Signers would be more likely to be registered, have stable addresses, print their name neatly, etc.
If a county was being pushed for more signatures, circulators may have to repeat locations, risking duplicates. If someone crosses the parking lot to sign, their signature may be more likely to be valid. If the circulator tackles the signer and holds them in an arm lock, the signer might use a fake name and address.
The collection rate in Los Angeles is 3.1%, same as the statewide average. In Orange County it is quite low, 1.7% which bodes well for a high validity rate for over 50,000 signatures.
The duplicate rate remains troubling. It seems inconceivable that you would have a 10% duplicate rate, with professional circulators (distinguished from paid circulators). Some voters may sign every petition and forget which they had signed. But unless the circulator is really careless and doesn’t ask?
The counties that have completed their sample checks estimate that there will be 31,410 duplicates in their counties. But 42% of those are expected in two counties, San Diego and San Bernardino, despite the two only representing 26% of raw signatures. Meanwhile, Riverside, which is between the two, is estimated to have 3.5% of the statewide duplicates with 11% of the signatures.
San Diego which has a quite high simple validity rate of 76.9% estimates a 6.8% duplicate rate. Riverside, with a similarly high simple validity rate of 74.8% estimates a 1.5% duplicate rate. But if they both had a 4.1% rate (the average of the two), then we would expect 4 duplicates to be found in an average San Diego sample, rather than the 6 that were found; and 3 duplicates in an average Riverside sample, rather than 1 that was found.
Right now, it looks like there will be enough valid signatures, but not after duplicates are removed.
The current projected simple valid signature count is 822,000, just 15,000 over there threshold. But that only permits a 1.25% duplicate rate, and that is very optimistic.