7 f,XMMM mL4$xM *s .c.Appendix .c2.Glossary from Samuel Merrills Making Multi-candidate Elections More Democratic. Princeton N.J.: Princeton University Press, 1988. Glossary, pages 133 to 138. Admissible strategy. A feasible voting strategy is admissible if there is no other feasible strategy that is preferred to it for at least one contingency [any feasible set of ballot preference orders from other voters] and at least as good as it in all other contingencies. Collusion. Collusion is any effort by a coalition of voters to alter, by agreement on how to vote, the outcome of an election. Condorcet candidate. The Condorcet candidate in a multi-candidate elections that candidate, if one exists, who could beat each of the others in separate pairwise contests,i.e., is preferred to each of the others candidates by a majority. Condorcet completion method. Condorcet completion method is a voting system that chooses the Condorcet candidate, if one exists, and specifies a contingency rule if one does not. Condorcet efficiency. The Condorcet efficiency of a voting procedure is the proportion or percentage of a class of elections (for which a Condorcet candidate exists) in which the voting system chooses the Condorcet candidate as winner. Contingency. A contingency is any feasible set of ballot preference orders from other voters. Correlation. The correlation between two random variables is a statistical measure of the tendency of the two variables to vary in concert. together / with each other Decision under risk. A decision under risk is a choice of strategy when the probabilities of possible outcomes are known. Dimension of a spatial model. The dimension of a spatial model denotes the number of coordinates designated in each point in space. Each such coordinate may be intended to represent the position of a participant (voter or candidate) on a specific issue or characteristic. Dual culture. A duel culture is an electorate in which all or most voters have one or two preference orders, usually opposite in nature, or alternatively, an electorate following a bimodal distribution in a spatial model. Impartial culture. An impartial culture is a model of an electorate in which all preference orders (for a set of candidates) are equally likely. Insincere voting. A voters ballot is insincere if his reported preference order differs from his true preference order. Kemeny distance. The Kemeny distance between two preference orders is the number of adjacent pairwise switches needed to convert one preference order to the other. Monotonicity. A voting system violates monotonicity if a voter can raise a candidate in the social ordering by lowering that candidate in his individual ordering. Multi-candidate. A multi-candidate election is an election in which there are three or more candidates. Negative of distance utility. Normal distribution. The normal (or Gaussian or bell-shaped) distribution of probability specifies that probability follow the density f (x) = (1/s 2) exp[ - (x - )/2s2]. [Where s = the standard deviation, and = the mean.] Polarized society. A polarized society is an electorate in which two or more (usually disparate) preference orders predominate. Such a society with exactly two dominating preference orders is called a dual culture. Random society. A random society is a model for an electorate in which, for each voter, candidate utilities are drawn independently from a uniform distribution. = an impartial culture? Relative dispersion. In a spatial model of voting, the relative dispersion of candidates to voters is the ratio of the standard deviation of the candidates positions to that of the voters. [See figure 3 and 4] Simulation. A simulation is an experiment run as a model of reality. The simulations in this book are computer simulations, i.e., are run on a computer using mathematical models. They are also stochastic, that is they involve input generated to follow probability distributions. Social utility. The social utility of a candidate is the total (alternatively, the average) utility of the candidate over all voters. Will Merrill OK this? Social-utility efficiency. 2 The social utility efficiency of a voting system is the normalized ratio between the expected social utilities of the candidate selected by the system and the candidate maximizing social utility. 1 The social utility of a candidate is the total (alternatively, the average) utility of the candidate over all voters. Squeeze effect. The squeeze effect refers to the reduction in electoral success of a candidate when, in a spatial model, nearby candidates draw support away from the focal candidate. Standard deviation. Standard deviation is a measure of the variation of a random variable; namely, the square root of the average squared deviation of the mean. Strategic voting. Strategic voting involves any decision by the voter in marking his ballot intended to improve the outcome from his point of view. [ In addition to insincere voting, it includes, under approval voting for example, expansion or truncation used to optimize a voters effect on the outcome. ] Transitivity. A voters preference order is said to be transitive if whenever the voter prefers A over B and B over C, he also prefers A over C. A similar definition applies to a social preference ordering. [Utility. The expected utility of a voter sees in a candidate is the amount the voter expects to gain or lose if the candidate wins. This amount may be measured in safety, status, dollars, happiness, or any other measure but most often no unit of measurement is specifiedeach voter may choose his own.] self-rightiousness, self-esteem, optimism, opportunities, Criterion Definitions from Phillip Straffins Topics in the Theory of Voting. Boston: UMAP, 1980. Chapter 2. Condorcet winner criterion: If there is an alternative X which could obtain a majority of votes in pairwise contests against every other alternative, a voting rule should choose X as the winner. Majority criterion: If a majority of voters [1 out of every 2] have an alternative X as their first choice, a voting rule should choose X. Monotonicity criterion: If X is a winner under a voting rule, and one or more voters change their preferences in a way favorable to X (without changing the order in which they prefer any other alternatives), then X should still be a winner. Pareto criterion: If every voter prefers as alternative X to an alternative Y, a voting rule should not produce Y as a winner. Definitions used in electing multi-seat legislatures Coalitions: A ruling coalition of legislators must continue to agree on policies or else risk a collapse of government that requires a new election. A working coalition may form simply to pass a particular bill and then disappear. Cumulative-vote plurality gives each voter as many votes as there are offices to be filled. He may give all of his votes to one candidate or spread them out among several candidates. The first strategy is always better than the latter. Multi-seat districts: Non-dominated strategy: Proportional representation (PR): Most European parlemants have used PR since early in the 20th century. Such constitutions help small parties to earn some representation giving them a voice, but rarely power, in their legislatures. Single-seat districts: Single Transferable Vote (STV): In its multi-winner form, Hares STV becomes a bit more complicated. Weighted Votes: .c2.Blank ballots and worksheets Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Ballot Fill a total of 5 boxes. Fill only 1 box below each number and only 1 box on each candidates row. Fill #1 for your favorite, #5 for least favorite. Rank Numbers . Candidates 1st 2nd 3rd 4th 5th . . . . . Worksheets for calculating election results You can run a CSTV election any time you have a pencil and paper. It is the quickest way for a group to find its majority opinion or the leading solutions on which to focus their discussions or negotiations. The limited usefulness of CSTV to small, informal, or unitary groups was discussed on page22. Computer users, see page 28. [Approval voting is easier to calculate and probably quicker for voters also. But approval does not tell us the manjority opinion necessarily it actually may miss a candidate preferred by a majority of the group.] None or the status quo should be a candidate in any election. If it wins, the committee or chairperson may have the power to decide to put the issue aside, for some time or to continue discussions and hold another election with different candidates, or to accept the result. Why? Voters tend to favor the top/ first name listed on any ballot, particularly the first.(ref.) Multi-candidate ballots give less advantage to the top position than do ballots with only two candidate. [the average amount of advantage] This advantage should not go to any one of the major candidates. For a large electorate, you can print several versions of the ballot with each candidate in the first position on her share of the ballots. If you cannot afford printing variations, or if they would confuse your voters, then None should take the top position on the ballot list. Use a tic-sheet like this the one below example 1 for recording the pairwise preferences from each ballot. Pairwise compairisons . None A B C D . voters voters voters voters voters voters voters voters voters voters for for for for for for for for for for ROW : COL. ROW : COL. ROW : COL. ROW : COL. ROW : COL. None A B C D Write the final totals as printed numbers Arabic numerals on a similar table. Reverse the pairs of numbers for cells as shown.If one row has only ratios greater than one to one (1:1), for example, 12:7, 10:9, 11:8, and 14:5, then that rows candidate is the Condorcet winner. (The ratios in that candidates column should each be less than 1:1, in this example 7:12, 9:10, 8:11, and 5:14.) If no Condorcet winner exists, then use a tic-sheet like the one below to total the first-place votes. for each candidate. Pairwise compairisons . None A B C D . voters voters voters voters voters voters voters voters voters voters for for for for for for for for for for ROW : COL. ROW : COL. ROW : COL. ROW : COL. ROW : COL. None A B C D If no Condorcet winner exists, then use a tic-sheet to total the first-place votes. for each candidate. First-Place Votes . None A B C D The committee or chairperson may be given the power to choose to table a cyclical issue, or to continue debate, or to accept the result of eliminations. If the later is chosen, then find the candidate who has the fewest first-place votes. Draw lines through that candidates row and column to eliminate her. Does any one of the remaining candidates win all her comparisons? If not, eliminate the candidate who is now the weakest in first-place votes. Continue until one candidate beats each of the remaining ones. .c2.Do it yourself worksheet; You can run a CSTV election any time you have a pencil and a piece of paper. It is the quickest way for a group to find its majority opinion or the leading solutions on which to focus their discussions or negotiations. The limited usefulness of CSTV to small, informal, or unitary groups is discussed on page22. The Appendix has blank ballots and worksheets which you may want to photocopy before marking. For computer spreadsheets, see page 29. [Approval voting is easier to calculate and probably quicker for voters also. But approval does not tell us the manjority opinion necessarily it actually may miss a candidate preferred by a majority of the group.] None or the status quo should be a candidate in any election. If it wins, the electoral rules or the chairperson may decide to put the issue aside for some time or continue discussions and hold another election with different candidates. Why? That choice shoud be taken decided made before the votes are counted. If a voting cycle occurs and there is no strong, clear-cut (Condorcet) winner, the same options may be applied. A voting cycle indicates a manipulation of CSTV may have been attempted. The options might help an electoral system resist manipulation. So there are 3 uses for thses options: none wins, no one wins a clear victory, and manipulation may have distorted the election result. Voters tend to favor the top/ first name listed on any ballot, particularly the first.(ref.) Multi-candidate ballots give less advantage to the top position than do ballots with only two candidate. [the average amount of advantage] This advantage should not go to any one of the major candidates. For a large electorate, you can print several versions of the ballot with each candidate in the first position on her share of the ballots. If you cannot afford printing variations, or if they would confuse your voters, then None should take the top position on the ballot list. Ballot for 5 Candidates Rank Numbers Choices 1st 2nd 3rd 4th 5th . Fill a total of 5 boxes. None . Fill only 1 box below each number. Name A . Fill only 1 box along each candidates row Name B . Fill #1 for your favorite. Name C . Fill #5 for your least favorite. Name D . Use a tic-sheet like the one below example 1 for recording the pairwise preferences from each ballot. Score the top-ranked candidate on a ballot with tic marks under her ROW all the way across her row. (SeerowC.) The lowest rank on a ballot gets her row marked with tics for the columns candidates all the way across. (Seerow None.) Each ballot will add one tic to each box. The final number of tics in each box must equal the number of voters. When counting a ballot, just fill the upper right boxes above the hash () marks. Then flip those row:column ratios to fill the lower left boxes. (See the box for row A , column B compared with the box for row B , column A. Pairwise comparisons . None Name A Name B Name C Name D . votes votes votes votes votes votes votes votes votes votes for for for for for for for for for for ROW : COL. ROW : COL. ROW : COL. ROW : COL. ROW : COL. . None I I I I Name A I IIII I : II I I Name B I II : IIII I I I Name C I I I I Name D I I I I Write the final totals as printed numbers Arabic numerals on a similar table. Reverse the pairs of numbers for cells as shown.If one row has only ratios greater than one to one (1:1), for example, 12:7, 10:9, 11:8, and 14:5, then that rows candidate is the Condorcet winner. (The ratios in that candidates column should each be less than 1:1, in this example 7:12, 9:10, 8:11, and 5:14.) If no Condorcet winner exists, then use a tic-sheet like the one below to total the first-place votes. for each candidate. First-Place Votes . None 0 A III 3 B II 2 C IIII I 6 D II 2 The chair may choose to table a cyclical issue, to continue debate, or to eliminate. Find the candidate who has the fewest first-place votes. Draw lines through that candidates row and column to eliminate her. Does any one of the remaining candidates win all her comparisons? If not, eliminate the candidate who is now the weakest in first-place votes. Continue until one candidate beats each of the remaining ones. .c2.Computer spreadsheet to calculate CSTV; Computer spreadsheets to calculate CSTV are available for $5.(US) from the author. Please specify MS-DOS or Macintosh disk format. Lotus 123, Excel, SYLK. Robert B. Loring Robert B. Loring 5911 Sprinfied Dr. 5114 Sunset Beach Drive Bethesda, MD 20816 Olympia, WA 98502 USA USA (206) 866-1619 .c1.References; Bordley, R. F. A Pragmatic Method for Evaluating Election Schemes through Simulation. American Political Science Review. 77 (3/83) 123-141. The author graphs utility efficiencies for several voting systems in simulations with varying assumptions. Chamberlin J. R. and Cohen J. L. Toward Applicable Social Choice Theory: A Comparison of Social Choice Functions under Spatial Model Assumptions. American Political Science Review. 72 (12/78) 1341-1356. The authors used computer simulations to estimate the Condorcet efficiencies of 5 voting systems. John; 5601 Haven Hall, Dept. of Political Science; U. of M., Main Campus; 48109 Chamberlin J. R., Cohen J. L., and Coombs C. H. Social Choice Observed: Five Presidential Elections of the American Psychological Association, Journal of Politics. 46 (1984): 479-502. The authors continued computer simulations to find Condorcet efficiencies, violations of subset rationality, and manipulability of 5 voting systems. Chamberlin J. R., and Featherston, F. (1986 in Merrill p 74) or? (1985)- Selecting a Voting System. Journal of Politics. 48(1985):347-369. The authors present a method for mimicking the voting patterns of a particular electorate. Condorcet, Marie-Jean-Antonie-Nicolas de Caritat, marquis de Fishburn, Peter. Condorcet Social Choice Functions. SIAM Journal on Applied Mathematics. 33: 469-489 Gibbard. Manipulation of Voting Schemes, a General Result, Econometrica, 41: 587-601. Gudgin, G. and Taylor, P. J. Seats, Votes, and the Spatial Organization of Elections. London: Pion, 1979. Thomas Hare, Treatise on the Election of Representatives, Parliamentary and Municipal. London: Longmans Green, 1859 Mansbridge, Jane J. Beyond Adversary Democracy. Chicago: University of Chicago, 1983. Two case studies leading to brilliantly original insights on the normative view roles, limitations, and effects of democratic systems, including voting, in small groups and society. Merrill, Samuel, Making Multi-candidate Elections More Democratic. Princeton N.J.: Princeton University Press, 1988. 150 pages. Dept. of Math; Wilkes Col. Wilkes Barre, PA 18766, Chapter 1, pages 3 through 14, introduces the types of problems found in multi-candidate systems. Chapters 2 - 4 give his findings on Condorcet and utility efficiencies in computer simulations. Chapters 5 - 7 look at the effects of strategic voting on the major voting rules. Merrill wrote the main text for lay readers and included 6 appendixes of math proofs for professionals. Mueller, Dennis; Public Choice 2. Cambridge: Cambridge University Press, 1989. 518 pages JF 1001.M78 Mueller reviews the breadth roots of the public choice in economic philosophy, field with less emphasis on specific voting systems than the other authors here. Niemi, R. G. The Problem of Strategic Behavior under Approval Voting. American Political Science Review 78:(1984): 952-958. WA State Lib Niemi, R. G. and Riker, W. H. The Choice of Voting Systems Scientific American. June 1976, 234:21-27. Plott, Charles R., and Ferejohn, John A. of Cal Tech present a case in which deletion of the last-place candidate reverses the order of finish calculated by Borda. (and probably by standard score.) Straffin, Phillip D. Topics in the Theory of Voting. Boston: UMAP, 1980. Dept. of Math; Beloit Col. Beloit , WI 53511. WA State Lib Straffins short book is the clearest introduction to the study of voting systems. He limits his analysis to the axiomatic search for faults in each system. Satterthwaite. Strategy-Proofness and Arrows Conditions, Journal of Economic Theory 10:187-217. Taagepera, Rein and Shugart, Mathew Soberg. Seats and Votes, the Effects and Determinants of Electorial Systems. New Haven: Yale, 1989. Tideman, T. Nicolaus. 1981. The Relative Attractiveness of Voting Rules. Presented at the meeting of the Public Choice Society 3/13-15, 1981 New Orleans La. UW? Suzz Per 994.05 AU The Australian Quarterly Main collection 328.94 Au78 The Australian Parliamentary Handbook WA State? Main collection HA3001.B68a Official Yearbook of Australia Aux Stax HC257.16P86 Ireland Public Affairs/Leargas Feld, Scott L.; Grofman, Bernard. Necessary and Sufficient Conditions for a Majority Winner in n-Dimensional Spatial Voting Games: An Intuitive Geometric Approach. American Political Science Review 31 (1987) 704- Rosenberg, Shawn W. The Structure of Political Thinking. American Political Science Review . 32 (1988) 539 .c4.Figures 3 to 9. Distributions of Voters and Candidates .c4.Electoral Process Cartoons Back cover. Misc. Points This is still a majority-rule system. If a majority coalition votes for candidate A then A will win with no eliminations via MSTV, CSTV, CIEVE. Candidate A might not win under approval, standard-score, or Clark tax voting systems. MSTV, CSTV, CIEVE decide voting cycles. So the 64% rule is not important here. Majority rule probably evolved from and co-existed with consensual democracy in classical Greece, particularly in Athens between the seventh and fifth centuries B.C.E. for page ~5 In simulated elections, Merrill found the frequency of cycles ranged from 47.5% for elections with 10 candidates and 25 voters randomly distributed on issues, to just 1% with 5 candidates and 200 voters normally distributed in a bell-shaped curve. This distribution simply means most voters are moderates. [Me:20,24] Chamberlin and found similar percentages for their simulation assumptions. On ballots from actual elections of the American Psychology Association, Chamberlin, Cohen and Coombs found Condorcet winners, therefore no cycles, in 5 out of 5 elections. Those were five-way races using rank-order ballots. Did Tideman report the frequency of cycles? I suspect STV is less subject to non-mon than alternative vote is. A candidate who exploits non-monotonicity to win the first seat is probably a close second to the Condorcet candidate - who will win the next seat. Even so, STV should check for and never eliminate a Condorcet candidate. If a candidate wins by Condorcet and has the fewest first-place votes, then skip that candidate and eliminate the one with the next fewest first-place votes. The winner of the last seat can be picked by the CSTV system introduced in this article.) [See page A1 & 2] From electorates simulated with random or uniformly distributed preferences, only 76% of elections had Condorcet candidates. But ninety-nine percent of spatial model elections had Condorcet candidates.(Me:24) Five of five APA elections did also. And of 1,000 elections from electorates simulated to resemble the APA voters, all had Condorcet candidates.(Ch,1984) In such an electorate, CSTV will be 100% monotone. Probably only 10% of real-world elections lack a Condorcet winner. (Merrill observed 24% of random and 2% of spatial model elections lacked Con winners. Chamberlin and Featherston found voter preference patterns intermediate between the random and spatial models. So the data suggest......)Probably less than <25% [Ch,C, & C] of those elections have a voting group with the position and incentive to manipulate non-monotonicity -if it could get the necessary information about other groups preferences and coordinate its own members ballots. Such manipulation of that 2.5% of elections will always pick a candidate who scores a very close second under (sincere) Borda calculation. So we may conclude that non-mon is very rare and if it ever occurs, the outcome is not much different in terms of [policies,] popular support or legitimacy. One case can not prove that CSTV is monotonic more often than MSTV is but logic can. The non-monotonicity of alternative vote results from the elimination process. (Straffin:26-27,30) The Condorcet criterion requires eliminations less often than the majority criterion does. With fewer eliminations, CSTV will be less subject to non-monotonicity. See page D. Notes 2 & 3 on Matt Smiths copy page 3 Note 1b on Matt Smiths copy page 3 To prove that CSTV sometimes needs fewer eliminations in elections with cycles (but not that it is monotone) we could start with a Condorcet election of A. Add a candidate F who beats A but loses to C and so creates a cycle. Give F fewer first-place votes than either A, B, C or D. No one wins by Condorcet so we eliminate F. A then wins by Condorcet but not by a majority because A, B, C, and others divide the first-place votes. This voting pattern is almost certainly rare. The fact is, most multi-candidate elections have Condorcet winners but not majority winners. So MSTV often requires eliminations and CSTV does not. Because alternative voting is not monotonic, its outcomes are not independent of irrelevant alternatives The ...relative standings of candidates ... [can] be altered by the entry of additional candidates into the race.(Me:10) or changing candidates M and N make a forth candidate win. But few elections are manipulable under MSTV. (Me:Ch 6) CSTV is probably better because it is much more often monotone. Merrills spatial-model simulations had Condorcet-criterion winners in about 98% of the elections. So in 98%, CSTV and Black will pick the same winner. CSTVs social utility efficiency will be almost 98% of Blacks plus about 2% of Hares, that leads a utility efficiency between 95% and 97%. Following Merrills analysis in the paragraph above, I have assumed Blacks efficiency when there is no voting cycle (and Black uses Condorcet s criterion) is lower than when there is a cycle (and Black becomes Borda). So Ive dropped Blacks efficiency from 98% to 97%. I have had to assume that the utility efficiencies of Borda and Hare stay about the same regardless of voting cycles. 98% x 97% + 2% x (40% to 98 %) = 95% to 97% utility efficiency. Would most (50/50) splits of this electorate (example 3) into 2 electorates result in 2 victories for the sincere Con winner? I dont think so. But I would state the positive arguement this way: The opportunities for this rarely occur and contain great risks. If too many of As supporters, one more in this example, change their first preferences, they elect C, their least favorite candidate. In addition to that danger the small target (of manipulation) has another effect. Most 50/50 splits of this electorate into two electorates results in 2 Condorcet victories for the sincere winnerB. So when a cycle occurs we can use the laborious calculation of many subelections to determine the sincere winner. A more likely method: Find the left-right sequencial ordering of candidates that fits the proximities of the candidates on most ballots. This is like finding the probable location or sequence or ? of genes on a chromosome. Then eliminate the ballots which do not fit that ordering. Part I: Quote others on: [Straffins chapter 2 on rank order; Merrills Introduction; Chamberlins 1984 introduction] Introduction to the importance of choosing a voting system How much do voting systems effect us? Effect on outcomes quality of decisions Effects on political styles and strategies - hence on our world view and social behavior manipulation effects on legitimacy of gov; (from political theory not voting theory) importance of participation, legitimacy, , the logical limitations of voting systems Arrows impossibility theorem Gibbards and Satterthwaites proofs the problems of voting systems Agendas need complex rules which their effect outcomes. All voting systems require some agenda but multi-candidate systems require fewer steps. Manipulations by nominators introduction of irrelevant alternatives divide and conquer squeeze by voters punish bullet voting or plunking: voting only for ones first choice. skip decapitate: do not vote for ones first choice give an approval vote to someone you dont approve of sequence for preferences changed Inherent problems monotonic Smiths rule Condorcets rule ballot position of a candidates name (the top left name gets more votes than other positions do) the measures of voting systems information (time) costs of voting: different kinds of ballots manipulability efficiencies Condorcet efficiency utility efficiency Work Order 10 ) Edit: 11 ) TOC renumber pages. 11 ) Bold best phrases, edit around those. 12 ) Spell check each file 14 ) Sensible Grammer checks 1x ) Delete commas from most introductory clauses. 13 ) Check indenting of long quotes. 20 ) Write 21 ) Test preface on readers. 22 ) Test worksheet on readers. 23 ) Rename and rewrite the normative sections: 24 ) Copy values statements to discussion and recommendation sections. 25 ) Why do we need more than 2 candidates to choose from? - Pres 1. orgs 26 ) Why do we need this way of counting votes? It is the least manipulable and it predictably elects centrists. But Hare elected the Condorcet winners in 5 of 5 APA elections. Plurality would have elected 85% of them. 27 ) How does it work? - See Basics parts of CSTV and worksheets to empower readers. 28 ) How does it effect a community? - It picks centrists from multi-party slates. The strong centrist tendency reduces incentives for extremism on the part of politicians. The multi-party aspect helps start-up & splinter parties which keep major parties open to change from below. This effectively combines the primary and general elections into one - so more people vote in the primary than is usual. 29 ) Type my outline of who wins and who loses from each political invention for Matt Smith, myself and WCL. Use it as a table of contents for the political section of Out of My Mind [OOMM]. 2x ) Where is it approved? by review committees for political sci Journals, Tilth org, 2x ) How has it worked in use? - [Tilth, UUA, ] 2x ) Change MSTV CSTV for final published? Not so many people will work to promote it. 29 ) Type in notes on CSTV: incl pages A... F. 29 ) Type in my notes on approval voting and a prcis of Niemes article. 29 ) Review notes on other voting systems. 30 Appendix: 37 ) Refs from Dad on normative poli sci 38 ) Describe each simulation by annotating its citation in the reference section footnote: Their methods are described in the glossary and annotated bibliography. Contrast 1) approval, 2) plurality ?= agenda? 3) standard score, 4) CT 39) Rewrite glossary Straffins or .. Strategic, monotone 39 ) Appendix on simulation quoted from Chamberlin and Cohen (1978, page 1348). 36 ) Computer SS cell formulas and flow chart. 40) Draw 42) An opinion shift on a scattergarm. Its effect on the choice of a utility maximizing candidate. 44 ) One Excel chart for each voting systems results from the table Nearness to the Center. Add PR and notes on the need to research STV and SNTV. 45 ) Cartoon of the campaign and election process 47) For PA S-curve: A Pharoe as God & King Henry III, signing of the Magna Carta & a council of nobels, guild sign and meeting and Rembrandts (Night Watch?), North American & French Revolutionairies: groups of men, no women, in discussion at a pub (no mobs and no uniforms), Abolitionists with banners? & a black brigand in US Civil War uniforms, Sufferagettes with banners? in US & Europe. 49) a voters (black dot with white V) distance to each of 5 candidates (white dots with black italic letters). 60 List of Examples: 61) Find a case where voters raise a candidates rank and CSTV is monotone but MSTV is not. Start with Straffins case of non-mon 68) x cases where Borda is or is not Condorcet but MSTV and CSTV would be. Should I replace mine with Merrills 3 X 3 on p 66 showing 2 cases in one? Misc. notes 65 ) v Check examples {in Excel SS}. 69 ) examples of non-mon 70 Letters: 71 )** Write to the Australian (and Irish) embassy for copies of preferential ballots. 72 ) Merrill Does standard-score voting reward punishing voters? 73 ) Chamberlin Are the charts on Nearness to the Center of the Theoretical Electorate or the table Sensitivity to Incomplete Ballots acceptable interpretations or extensions of your research ? Or do these strain or break the rules of inference? 80 Readings: LRC first reading; Steven, David & Barbara, Ken, 75 ) Where is it most clear? Where is it most convincing? Where is it least convincing? Where is it most confusing? 70 Read: 71 ) * Non-monotonicity hyphen? 71 Embassies addresses and phone numbers 77 ) Gibbard, and Satterthwaite on theoretical manipulability 75 ) Brams 1983 Approval Voting for strategies under Hare; 76 ) also ?s 1975 math article listed in Ref.s Check facts in Lib 79 ) Reviews of Merrill 79 ) Any (80 or 90?) years of experience with MSTV in Australia (not STV) and 70? with STV in Ireland. Thomas Hares Treatise 1859. 76) * Austral Pol Sci J ? Indexed in ? 77) * Name(s) then address(es) for Aussie political sci writers in books here. 79) First name of Copeland in my books 72 ) municipal govs systems of representation and legislative voting? 73 ) Check citations page numbers: 73 ) How to deal with multiple references to one point and/or page? 90 ) Misc 99 ) loss (laus) - an instance of losing 99 ) lose (luz) - to suffer loss lost - not won 99 ) loose (lus) - not rigidly fastened Goals My highest goal is to match jurisdiction borders to problem areas. And to gaurenttee fredom of speech, information, and emmigration. (with repayment of debts to local businesses and governments. The second highest is a utility voting system or parlementary procedures or both which give minorities thier share of power. This would make some majority interests anxious to sepparate and gain independance from the minorities. Of course some minorities have long wanted independance from the majority, so their relations would become symetrical. Equality under the law through equal division of funds for prosecution and defense lawyers.  Loring Condorcet, single transferable vote  Loring Condorcet, alternative vote  Loring Condorcet, alternative vote  v wxhyvxh 74) I suspect Tideo the importance of choosing a vot /_GZ&6Q7BZvly K \ W f  & ) <  # z23LOQHZ+A@AZabkly TY%&(>DE.VWkl=?IW[\} NO[]_ A@Y_acegikmopqr 9 g h t v x z | ~ !R!!!!!!!!!!!!!!!!!!!!!"k""""""""""""""""""""########@Y##############$$$$$$$$$$$$$$$$$$%%%%%%%%%%%%%%%%&&&&&&&)&*&+&1&2&&&' ' ' ''''''''''' 'B'C'E'q(4(D(o(())*B@W*B*P*******++++++- - -)-B-------.Q.T.X.a.e.q.s.|.................................///-/^/e/h0001#1$1%1)1*1?111111111@@Z11112 2 2 22222222 2!2"2'2(2/20212225262?2@2A2B2C2E22222233333 3 3 3333445"5&5>5@666J6b667 7778|888899O9U999::::;;; ;t;<<<=='@@@X='=,=8=<=F=H=J=L=N=P=R=T=V=X=[=]=v={===============>>>>#>)>+>,>4>6>;>\>b>g>h>l>n>t>v>>? ?$?%?(?O?P?c?o????@@@@AAA"A#A'A(A,A_Aa@!@MAaAbAfAgAABBBKBWB^BbBcBdBeBoBsBBBBBBBBBBBBBBBBBBBBBBBBCC!C7ChCoCrDDDDDE6E7E<E?E@EGEHENEOEPETEUEYEZEaEcFmFqGGGHTHWHZH]HrHsHwHHHHHIII@@ @VIJJ:JJKKK KKL^L_LLLLMKMLMMMMMMN4N@NONPNoNNNNOO.O3O4OJOdOOOPEPFPYPPQRRRRRRRRS SS$S<SSSSTTTT\ToTTUSUTUkUUUUUUVVVVVVVWWZWtWuWWXX@^XXX0XIXeXXXYYZZ:ZKZZZZ[[``b_bfbcddg4gHhWhhhhhiiiii4i5i[i`iaibigihiiiiiiiikkll$liljmmo o!o"ocodp~pppppqqr4sTsVt1uu3u^uuv]x@x_xxA@@Yxxxxxxyy-y@yEy^y`yyyzzz"z;zDzOzQzzzz{||+|1}}z[]$%):GHLT-0Ya./^_@$]"'45IQY^npqrs}=?LSZbs@@7 cG67Zl K W )  z3Haby(I[\}Oq˫(@@ @!@& Tp $ @ \x(@@ @!@T th<q  8 h !!Q!!!!!!"5"j"""""#M######$$f$$$%%%%%%%&&)&*&+&3&&&''B'C'Dټټټޕټټټټ(@@ @!@& Tp $ @ \x(@@ @!@T t (@ @!@(@ @!@Tp ,:'D'E'q)*- - -|--....01#1?12C2D2E22{g{g[ 4@@ @!@@@ @!@ (|$"@@ @!@/p ( @|l$t>u>vA'A(A`Aҭ{pS@@ @!@#   ,8 @@ @!@@@ @!@|d X@@ @!@| X @@ @!@|L(4@@ @!@8 \xH4<!$AABBBBDDE6EbEcGG=G>GGHqHrHHIIKK L^L_MKMLݸxrme]WRHHHH ` ,p4(@@ @!@8< ( dP$ #x @@ @!@$@@ @!@2p @,l