UNBIASED AUTOMATED REDISTRICTER UBARD Taking the bias out
UNBIASED AUTOMATED REDISTRICTER UBARD Taking the bias out of district lines Blue Group Feasibility Presentation CS 410 Spring 2017 March 15, 2017
TEAM BLUE Jack Doherty Chris Hartman Adam Hill John Rattz Kenneth Sutton Jacob Villafane DOMAIN EXPERT : Dr. Richard Murray, professor of political science, University of Houston CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 2
TEAM MANAGEMENT Roles • • Project Manager: Adam Hill Webmaster: Chris Hartman Web Version Supervisor: Kenneth Sutton Database Manager: Jacob Villafane Data Manager: John Rattz Algorithm Expert: Adam Hill UI Manager: Jacob Villafane Tester: Jack Doherty CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 3
DISTRICT LINES The redrawing of districts is subject to bias and impacts all government and public services affected by district boundaries. CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 4
MANIFESTATIONS OF THE PROBLEM • School districts and population • • Racial and economic segregation of students Government services such as Postal service • Sanitation • • Unfair election results via gerrymandering CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 5
IMPACTS OF BIASED DISTRICTS Gerrymandering is the process of drawing political district lines to favor a party or its members in an election. • Partisan gerrymandering creates many districts for the party in power and few for the opposition – resulting in an unbalanced voter base per district drawn. • Historical neighborhoods, minority voters, and socioeconomic factors may be misrepresented. • CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 6
PENNSYLVANIA DISTRICT MAP, 2012 Source: Vox, “Gerrymandering, explained” House Republicans won 49% of the popular vote in 2012 but won 72% of the state’s seats CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 7
NORTH CAROLINA’S DISTRICT MAP, 2012 -2014 Source: Daily Kos, “No, Maryland is not the most gerrymandered state. There is more to gerrymandering than ugly shapes” CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 8
NORTH CAROLINA’S DISTRICT MAP, 2012 -2014 Source: Daily Kos, “No, Maryland is not the most gerrymandered state. There is more to gerrymandering than ugly shapes” • NC started with 7 -6 Democrat majority before redistricting • After 2012, 9 -4 Republican majority • GOP lawmakers tried to pack in black voters into 1 st and 12 th districts, struck down as illegal racial gerrymandering CS 410 SPRING 2017 - BLUE GROUP Population columns show minority voters are grouped together in a few districts to reduce their overall influence by skewing the margins of victory to favor Republicans. March 15, 2017 9
2012 NC VOTE RESULTS Source: Vox, “Gerrymandering, explained” CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 10
WASHINGTON STATE’S BIPARTISAN REDISTRICTING Source: Daily Kos, “Here's what Washington state might have looked like in 2016 without congressional gerrymandering” CS 410 SPRING 2017 - BLUE GROUP • Uses a bipartisan redistricting commission • Often devolves into a gerrymander aimed at protecting incumbents from both parties • Uncompetitive races • Incumbent protection redistricting helped GOP because they made gains in 2010 elections March 15, 2017 11
VIRGINIA • GOP redrew district lines to secure them 8 -3 advantage in 2012 -2014 • Map struck down for unconstitutional racial gerrymandering • Black voters packed into 3 rd, made surrounding more likely to vote Republican • Requested an extension in 2015, denied by judiciary • Led to current map Source: Daily Kos, “Here's what Virginia might have looked like in 2016 without congressional gerrymandering” CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 12
VIRGINIA CONGRESSIONAL ELECTIONS, 2016 District Winner Party Margin of Victory 1 Robert J. Wittman * Republican 25. 1% 2 Scott Taylor Republican 23. 4% 3 Robert Scott * Democrat 33. 8% 4 Donald Mc. Eachin Democrat 14. 6% 5 Tom Garrett Republican 16. 6% 6 Robert Goodlatte * Republican 33. 4% 7 Dave Brat * Republican 15. 8% 8 Donald Beyer * Democrat 41. 2% 9 Morgan Griffith * Republican 40. 3% 10 Barbara Comstock * Republican 5. 8% 11 Gerry Connolly * Democrat Unopposed CS 410 SPRING 2017 - BLUE GROUP * Incumbent Source: Washington Post, “Who’s winning Virginia: Live election results” March 15, 2017 13
VIRGINIA CONGRESSIONAL ELECTIONS, 2016 Total Vote Percentage, 2016 0, 967 1, 125 Seats Won, 2016 4 49, 171 48, 737 7 Source: Virginia Department of Elections, “ 2016 November General Official Results” Democrat Republican CS 410 SPRING 2017 - BLUE GROUP Independent Write-In Democrats Republicans Independents March 15, 2017 Write-Ins 14
THE SOLUTION: UBARD • Districts must fairly represent all voters equally Districts in which a minority group constitutes the majority of voters are exceptions. • However, the 14 th Amendment requires that states do not discriminate on the basis of race. • • The solution should incorporate compactness and contiguity constraints. CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 15
CURRENT VIRGINIA PROCESS CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 16
PROPOSED VIRGINIA PROCESS CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 17
COMPETITORS • Brian Olson: program that draws districts to be “optimally compact” based on 2010 census • Center for Range Voting: uses shortest splitline algorithm BDistricting CS 410 SPRING 2017 - BLUE GROUP Center for Range Voting March 15, 2017 18
COMPETITION MATRIX CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 19
THE CUSTOMER • Voters • Non-partisan/non-profit groups • Political activists • The press • Academia • State legislators CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 20
MARKETABILITY • Depending on the solution in place, costs for some states will vary • The costs to replace their current system could be significantly lower, which will more than likely sway the states opinion on the solution • The states that do benefit from UBARD will see an increase in their budget and fair districting • Possibly little cost for teaching the tool CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 21
ADVANTAGES OF UBARD • Unbiased districts based on available data • Can be extended to • • School districts • Governmental services such as postal and sanitation • Emergency services Public awareness of the problem CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 22
MAJOR FUNCTIONAL COMPONENTS Databases • • US Census Data Previous District Maps State and City Boundaries Previous Election Results Internet UI Local Cache • Frequently Used Data • Historical Sites • Archived District Map Solutions CS 410 SPRING 2017 - BLUE GROUP UBARD District Maps • • • Electoral Districts District Compactness Population Equality Legal Requirements Algorithms Ranked Priorities March 15, 2017 23
COSTS Hardware • • Minimal • No real time applications, time to run the algorithm is not paramount • Any relatively new personal computer would suffice Software • • • Also minimal • Census data freely available Maintenance CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 24
LIMITATIONS AND PROBLEMS • Won’t redistrict, just make suggestions • Algorithm limited in scope and scale • Supervising body could have a political agenda, modify code to produce biased districts CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 25
CONCLUSION • Redrawing of districts is subject to bias • Impacts all government and public services affected by district boundaries • Results in unbalanced distribution of public resources • Unbiased redistricting can be achieved by a software solution CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 26
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. Article V, Virginia Constitution. (n. d. ). Retrieved February 19, 2017, from https: //ballotpedia. org/Article_V, _Virginia_Constitution Baas, K. (n. d. ). Auto-Redistrict. Retrieved March 24, 2017, from http: //autoredistrict. org/index. php Gerrymandering and a cure - shortest splitline algorithm. (n. d. ). Retrieved March 02, 2017, from http: //rangevoting. org/Gerry. Examples. html Impartial Automatic Redistricting. (n. d. ). Retrieved March 02, 2017, from http: //bdistricting. com/2010/ Redistricting in Virginia. (n. d. ). Retrieved from https: //ballotpedia. org/Redistricting_in_Virginia Splitline districtings of all 50 states DC PR. (n. d. ). Retrieved March 02, 2017, from http: //rangevoting. org/Split. LR. html 2016 November General Official Results. (2017, January 6). Retrieved from Virginia Department of Elections: http: //results. elections. virginia. gov/vaelections/2016%20 November%20 General/Site/Congress. html Prokop, A. (2015, May 15). Gerrymandering, explained. Retrieved February 9, 2017, from http: //www. vox. com/cards/gerrymandering-explained/what-is-gerrymandering Who’s winning Virginia: Live election results. (2016, November 8). Retrieved from Washington Post: https: //www. washingtonpost. com/2016 -election-results/virginia/ Wolf, S. (2016, December 27). Here's what Virginia might have looked like in 2016 without congressional gerrymandering. Retrieved from Daily Kos: https: //www. dailykos. com/story/2016/12/27/1610276/-Here-s-what-Virginia-mighthavelooked-like-in-2016 -without-congressional-gerrymandering Wolf, S. (2016, December 30). Here's what Washington state might have looked like in 2016 without congressional gerrymandering. Retrieved from Daily Kos: https: //www. dailykos. com/story/2016/12/30/1612731/-Here-s-what. Washington-state-might-have-looked-like-in-2016 -without-congressional-gerrymandering Wolf, S. (2016, June 1). No, Maryland is not the most gerrymandered state. There is more to gerrymandering than ugly shapes. Retrieved from Daily Kos: http: //www. dailykos. com/stories/2016/6/1/1532608/-No-Maryland-is-not-the-mostgerrymandered-state-There-is-more-to-gerrymandering-than-ugly-shapes CS 410 SPRING 2017 - BLUE GROUP March 15, 2017 27
- Slides: 27