01/19/2018. Attorneys for Plaintiffs SUPERIOR COURT OF THE STATE OF CALIFORNIA COUNTY OF SAN FRANCISCO

Transcription

1 0 SSAMERICAN CIVIL LIBERTIES UNION FOUNDATION OF NORTHERN CALIFORNIA MICHAEL T. RISHER () RAUL L. MACIAS (0) Drumm Street, nd Floor San Francisco, CA Telephone: ()- Facsimile: ()- AMERICAN CIVIL LIBERTIES UNION FOUNDATION JULIE A. EBENSTEIN (To Be Admitted Pro Hac Vice) Broad Street, th Floor New York, NY Telephone: ()-00 Facsimile: ()- COOLEY LLP WILLIAM P. DONOVAN, JR. () REBECCA L. TARNEJA () nd Street, Suite 00 Santa Monica, CA 00 Telephone: (0) -00 Facsimile: (0) -00 Attorneys for Plaintiffs ELECTRON ICALLY F I LED Superior Court of California, County of San Francisco 0//0 Clerk of the Court BY: RONNIE OTERO Deputy Clerk 0 SUPERIOR COURT OF THE STATE OF CALIFORNIA PETER LA FOLLETTE; and THE AMERICAN CIVIL LIBERTIES UNION OF NORTHERN CALIFORNIA, v. Plaintiffs, ALEX PADILLA, in his official capacity as Secretary of State of the State of California; and WILLIAM F. ROUSSEAU, in his official capacity as Clerk-Recorder-Assessor- Registrar of Voters for the County of Sonoma, Defendants. COUNTY OF SAN FRANCISCO No. CPF - DECLARATION OF PAUL MITCHELL IN SUPPORT OF PLAINTIFFS MOTION FOR WRIT OF MANDATE Date: March,0 Time: :0 a.m. Dept: 0 Judge: Hon. Harold Kahn RES ID:

2 0 0 I, PAUL MITCHELL, declare that:. My name is Paul Mitchell and I am Vice President of Political Data Inc. (PDI), a voter file company based in California with offices in Norwalk and Sacramento.. I am over the age of years and qualified to make this declaration. I make this declaration based upon my personal knowledge, unless otherwise specified. I have no direct or indirect interest in the outcome of this case for which I am offering observations, analysis, opinions or testimony.. PDI has been in business in California since. The core of PDFs business is managing a voter file which is comprised of state- and county-level voter file information, along with data from other data sources such as county assessor records, census data, and commercial data sources.. Clients of PDI include both major political parties, the California Democratic Party and the California Republican Party, along with candidates at the statewide and local level, ballot measure committees, independent expenditure groups, and state and local government, including contracts with local city clerks and even the Secretary of State.. PDI provides clients a deeper understanding of the voter file through processing of data which is not readily available in a raw voter file. One example of this is ethnic coding.. To identify ethnicity, PDI utilizes a proprietary surname database built initially from the U.S. Census surname database. This is supplemented with information from the voter registrations themselves, including non-u.s. place of birth and requests for non-english language ballots. Surname, birthplace, and non-english language ballot requests are all used to help define ethnicity. In some cases, surnames which could be multiple ethnicities are cross-referenced with the U.S. Census data from the American Community Survey, which provides the ethnic composition of a census block group.. Another core function PDI provides to clients is a rich history of past voter behavior. PDI collects and manages a system that tracks past votes in statewide and local contests, along with information about how voters cast their ballot (in person at the polling place, by mail ballot, or by another method), and when an early voter casts a ballot. From some counties, we obtain data on.

3 0 0 rejected ballots, including damaged ballots, late ballots that were received after the deadline, ballot envelopes lacking signatures, and ballots rejected because the signature on the ballot envelope does not match the voter s registration signature.. At the request of the ACLU we ran counts from our database of voters in the 0 California general election for whom we had identified as having their vote-by-mail ballot rejected due to a signature non-match based on the data provided by the county. This data was not available statewide; the data was only available in the counties' that use the DFM Associates software which collects and reports the reason a vote-by-mail ballot was rejected. While this means we only have data from of counties, these counties span the range of urban and rural, coastal and inland, big and small, and provide enough quality data to allow for detailed evaluation of ballots rejected for a signature non-match by ethnicity, age, partisanship, and other factors found on the voter file.. Data provided from the 0 California general election is consistent with what we have seen in prior election cycles. We find higher rates of vote-by-mail ballots rejected for a signature non-match among voters who are Latino (0.%), Asian-American (0.%), or bom outside of the U.S (0.0%), compared to voters who are non-latino and non-asian (0.%), and U.S.-born voters (0.0%). 0. We also find varying rejection rates by county that are outside the normal range that would be expected by normal variation. Attached as Exhibit A is the data for the counties for the November 0 general election, showing (i) the total number of non-match voters, (ii) the total percentage of non-match voters, (iii) the percentage of non-latino, non-asian non-match voters, (iv) the percentage of Latino non-match voters, (v) the percentage of Asian non-match voters, (vi) the percentage U.S.-born non-match voters, and (vi) the percentage of foreign-bom non-match voters. The counties for which we have data on vote-by-mail ballots are: Alpine, Butte, Calaveras, Colusa, Contra Costa, Fresno, Humboldt, Inyo, Madera, Mariposa, Merced, Mono, Napa, Placer, Riverside, Sacramento, San Benito, San Joaquin, San Mateo, Santa Barbara, Santa Clara, Santa Cruz, Shasta, Sierra, Sonoma, Sutter, Tehama, Ventura, and Yuba Counties..

4 0. Latino, Asian-American, and non-u.s. bom voters also had a higher rejection rate for late vote-by-mail ballots and had higher use of provisional ballots in the 0 and 0 elections. This could speak to an unfamiliarity with the voting process or issues regarding the instructions on election materials.. The variation in signature non-match rejection rates by county is not correlated with counties which had higher rates of rejection for late vote-by-mail ballots or provisional ballots. Anecdotal evidence suggests that some counties contact voters who have a signature which was deemed non-matched and give them an opportunity to rectify the problem in time to have their vote counted, while other counties do not.. The counties where data was available for the 0 general election had a 0.% rejection rate for vote-by-mail signature non-match. If that rate is applied to the,, voteby-mail ballots cast by all California voters in the 0 general election, that would project to,0 votes being rejected statewide for signature non-match.. There is no evidence to suggest that a significant number of vote-by-mail ballots rejected for a signature non-match are the result of attempted voter fraud. I declare under penalty of perjury under the laws of California that the foregoing is true and correct. Executed at Sacramento, California on this th Day of January, 0. 0.

5 EXHIBIT A

6 Exhibit A: Data from November 0 California General Election re: Vote-by-Mail Ballots County Non- Match Total % All Voters % Non-Latino and Asian Voters % Latino Voters % Asian Voters Non- Match % US Born % Foreign Born Alpine 0.% 0.% 0.00% 0.00% 0.% 0.00% Butte 0.% 0.%.%.0% 0.%.% Calaveras 0.% 0.% 0.0% 0.00% 0.% 0.% Colusa 0 0.% 0.0% 0.% 0.00% 0.% 0.0% Contra Costa 0.% 0.% 0.% 0.% 0.% 0.% Fresno,0.% 0.%.%.%.00%.0% Humboldt 0.% 0.% 0.%.0% 0.% 0.0% Inyo 0 0.% 0.% 0.0% 0.00% 0.% 0.% Madera 0.% 0.% 0.% 0.% 0.0% 0.% Mariposa 0 0.% 0.% 0.00% 0.00% 0.% 0.% Merced 0.0% 0.0% 0.% 0.% 0.% 0.% Mono 0.% 0.%.%.% 0.% 0.% Napa 0 0.% 0.% 0.% 0.% 0.0% 0.% Placer 0 0.% 0.% 0.% 0.% 0.% 0.% Riverside,.% 0.%.%.%.%.% Sacramento, 0.% 0.% 0.0% 0.% 0.% 0.% San Benito 0.% 0.% 0.% 0.% 0.% 0.% San Joaquin 0.% 0.% 0.% 0.% 0.% 0.% San Mateo 0.% 0.% 0.% 0.% 0.% 0.% Santa Barbara 0.% 0.%.0% 0.% 0.% 0.% Santa Clara, 0.% 0.% 0.% 0.0% 0.% 0.% Santa Cruz 0 0.% 0.% 0.% 0.% 0.% 0.0% Shasta 0.% 0.% 0.% 0.% 0.% 0.% Sierra 0.% 0.%.0% 0.00% 0.% 0.00% Sonoma 0 0.% 0.% 0.% 0.% 0.% 0.% Sutter.0%.0%.%.% 0.%.% Tehama 0.% 0.0%.% 0.% 0.%.0% Ventura 0.% 0.% 0.0% 0.% 0.0% 0.% Yuba.%.0%.%.%.%.0% TOTAL, 0.% 0.% 0.% 0.% 0.0% 0.0% Source: Political Data Inc.