The Arizona Elections Database

The Arizona Elections Database A Proposal for Discussion 1 Overview A free Internet resource containing:  Election results and voter registration data, current and past  Downloadable data files, by precinct, district, and census tract  Downloadable maps in three GIS (geographic information system) formats  State, county and city-level election data for primary, general, runoff and special elections  Vote counts for candidates and propositions 2 Why build this database? In the near term, and throughout the decade: To fulfill the political mapping and data analysis needs of anyone involved in: Campaign strategy  Political science  Public administration  Districting/redistricting  Education  Grant-writing  News reporting  3 By the end of the decade:  To have an accurate and accepted database for the 2011 round of legislative, congressional, supervisorial, community college and city council redistricting To avoid the delays and extra costs, caused by use of improper data, that have beleaguered candidates, elected officials and Arizona taxpayers over the past three decades of redistricting  4 Redistricting in the 1980s  The Legislature hired a private consultant to assemble prior election data The Legislature‟s computers were too small to handle the large data files produced by the consultant Because of a late start, there was not enough time to reformat the databases or obtain larger computers Redistricting technicians had to rely on paper maps, printouts and adding machines 5    Redistricting in the 1990s  The Legislature put off getting started on the redistricting process A private consulting firm was hired to design redistricting software – but the program operated far too slowly On tight deadline, the Legislature‟s clerical staff and temporary workers data-entered prior election data from county election department printouts Data errors were rampant and the database was abandoned 6    Redistricting in 1990s, (continued)  Democrat and Republican leaderships had to rely on redistricting resources from the national political parties and their consultants Once again: paper maps, printouts and adding machines  7 Redistricting after the 2000 Census  Proposition 106, on the 2000 General Election ballot, created the Arizona Independent Redistricting Commission (IRC) to draw new legislative and congressional maps State provided an incomplete set of computerized precinct maps to the IRC  IRC had to hire a consultant to finish the work  Two month delay, data-linking problems, and 8 additional taxpayer expense  Redistricting after the 2000 Census, (continued)  The very limited selection of prior election data provided to the IRC reduced the reliability of the analyses undertaken by its consultants Faulty data provided to the IRC about the number of Democrats and Republicans skewed the analyses of competitiveness measures. (The Congressional Districts plan was approved anyway) 9  Magnitude of errors in IRC data on the number of voters in Maricopa County legislative districts 10 For future redistricting, isn‟t all the data already available from the Secretary of State? Well, sort of… …but not really 11 Sort of…  The Secretary of State maintains precinct-level voter registration files on its website http://azsos.gov, but a direct link to the files is not provided. Visitors must know the correct Internet path to access the files, or be good at guessing a sequence of menu choices  County elections departments deliver to the Secretary of State an “electronic canvass” of precinct-level election results after each primary, general, and runoff election 12 …but not really  Those “electronic-canvass” files of precinct-level election results are in various formats and layouts Many of the files are not useable without specific software or expertise, and most are not readily useable with analysis or mapping software   The Secretary of State does not maintain results data for city or county elections and does not have electronic maps of precincts or election districts 13 Whose idea is this?  This idea is based on the California Statewide Database, which is supported by a permanent appropriation from California‟s State Assembly and managed by the Institute of Governmental Studies at UC Berkeley. Visit them at http://swdb.berkeley.edu Phoenix-based redistricting consultant Tony Sissons used California‟s database during his work on the Merced County, California, Supervisorial Districts plan 14  What‟s so great about the California database?    Developed during the 1980s with redistricting in mind Thoroughly vetted against error or partisan bias by its daily availability to the public Securely housed in a public educational institution 15 Is the California database a good model “as-is”, or can it be improved?  The California Statewide Database staff has recognized that adding county and city election results would greatly expand the usefulness of their political database and increase the number of stakeholders However, their funding from the California State Assembly only allows them to handle federal, statewide and legislative contests 16  If this is to be a free resource, how will it be supported? That‟s an open question at the moment… 17 A non-partisan, non-profit organization could be established to build and maintain the database, however…   fund-raising would be a constant activity private contributors may become vulnerable to partisan influence 18 A for-profit organization could operate the database, charging a subscriber fee to campaigns, political parties, and other institutional users, but…  the subscriber base would probably be too small to keep the fees low enough for the data to be accessible to a wide audience  this important resource, intended for the public‟s use and benefit, needs to be unquestionably impartial, and really should not be privately owned 19 A publicly funded institution could house and maintain the database, but...  in this era of declining public revenues, it is likely that legislative bodies might, in some budget years, not view this Database as having high enough priority for adequate funding. 20 A fourth funding approach to consider is a public cooperative  Given that the database will contain information from and about multiple jurisdictions, establishing an entity which receives funding from both state and local governments might be appropriate There may even be an existing multijurisdictional organization that could “adopt” responsibility for the database 21  Setting aside funding issues for a moment, who are the likely stakeholders?        The Arizona Legislature and State Government National, state and county-level political parties Political Action Committees Campaign consultants Political scientists, think-tanks, public policy analysts Candidates for political office Election-process advocacy groups Wait!! There are more… 22          Voting-rights advocacy groups University and college political science departments High school civics programs News reporters Public relations firms and lobbyists Foundations and philanthropic organizations Public administrators and elected officials Urban geographers and social service providers The general public 23 How can this database be used? 24 One example is thematic mapping of political party registration: 25 Another example is mapping precinct-level support for propositions: 26 Here is an example of how election precinct and city council maps and data are maintained in the Database 0061 Prop 1 Prop 1 District 1 1 1 1 Precinct 0061 0181 0481 3061 YES 694 600 963 475 NO 116 96 141 84 1 3361 487 61 27 What are the benefits to implementing the Database now?   Better, more-reliable data, available sooner Many opportunities for savings:  The Database could supplement a portion of the Secretary of State website  State and county elections staffs can simply refer some callers to the Database, rather than producing and mailing copies of maps and data  Users will no longer have to spend time resolving year-to-year, and county-to-county differences in 28 reporting formats A Case Study in Savings: Census Precinct Mapping THE PROJECT: Census Bureau‟s Redistricting Data Program – In preparation for the Census, states provide electronic maps of precincts so that the Bureau can report accurate population data for redistricting LAST TIME: Preparing for the 2000 Census, the State and counties spent six months and an estimated $400,000 on the precinct mapping task. Despite these expenditures, half of Arizona‟s counties ultimately lacked the technical or staff resources to complete the work CONSEQUENCE: The Arizona Independent Redistricting Commission had to hire a consultant to finish the work, at extra cost and considerable delay 29 FOR NEXT TIME:  Electronic precinct maps from the Arizona Elections Database can be used to “tag” each 2010 Census Block with the identity of the precinct in which it falls An afternoon‟s work for the Database staff, at a tiny fraction of the previous cost!  30 We have already digitized the precinct maps used for the 2002 and 2004 Primary and General Elections From this point, keeping the maps updated makes much more sense than waiting until 2009 to digitize a decade‟s worth of precinct maps 31 Okay, what now? This short presentation is meant as a starting point 32 We are making this presentation to a number of groups that we think might see…  the immediate value in having useful political data to analyze in preparing for upcoming elections  the long-term value of a dependable consistent database that can be relied upon throughout the decade the significant cost avoidance of not repeating the timeconsuming and costly database „start-overs‟ that have been a troublesome feature of the state‟s last three redistricting cycles 33  What funding, setting, or support models do you think would be appropriate? Please send your comments to Tony Sissons at tsissons@aol.com or call at (602) 230-9580 Thank you! 34