Preferential Voting

Dominion, Volume 28, Issue 285, 29 August 1935, Page 11

Preferential Voting

Sir, —There appears to be a mistake in the positions of the candidates in the article contributed by the New Zealand Legion appearing in your issue of August 24. hence the confusion of “T.T.” and others no doubt. I would draw “T.T.’s attention to the placings set out in my letter published in “The Dominion” of August 24 which are as follows: — A group: 13 ballot papers. J. Smith. 2. Jones. 3. Robertson. I? group: 12 ballot papers. 1. Robertson. 2. Jones. 3. Smith. C group: 11 ballot papers. 1. Joiics. 2. Robertson. 3. Smith. In the article contributed by the New Zealand Legion, the names Robertson and Jones were transposed in Groups B and C and in (he summary hut not in Group A, consequently the proof is “out.” This should read as follows:— First Trial Table.

Average of trial totals 108 divided by 3=36. Robertson and Smith do'not exceed trial totals and are rejected. Jones is elected. The most-preferred • man wins. Proof. 24 voters prefer Jones to Robertson. 12 voters prefer Robertson to Jones. Jones wins against Robertson. 23 voters prefer Jones to Smith. 13 voters prefer Smith to Jones. Jones wins against Smith. In a straight-out contest with Jones., both Smith and Robertson would be defeated. Therefore, Jones is elected. As the New Zealand Legion, in publishing the details of the Nanson-Hogben system, is endeavouring to enlighten the public regarding the advantages of that system of preferential voting, it is to be hoped that the article to appear in next Saturday’s issue in connection with a fivecandidate contest will be more accurately set out. —I am, etc., ANOTHER. DEMOCRAT. Wanganui. August 27.

Sir, —If ever there was a better example of “the blind leading the blind than was the case in the article contr.ibuted bv the New Zealand Legion, and which purported to explain bow preferential voting would work, then I would, like to see it. Your contributor makes a perfectly simple voting system seem not only complicated but absurd. The voting is supposed to be: — (a) Group (b) Group (c) Group 13 papers 12 papers 11 papers 1 Smith 1 Jones 1 Robertson 2 Jones 2 Robertson 2 Jones 3 RobertsonS Smith 3 Smith And then th e statement is. made in bold type. “Least desired candidate, elected.' which means, if it means anything.. that Robertson is eventually elected, which is nonsense, as he received the least number of votes and his second preferences go to Jones, who is thus elected. Now, apart from this mistake, the manner in which the electors are supposed, to have voted has no relation to probability. as we are expected to believe that electors who preferred Smith would all give theil second preference rotes to Jones, while those electors who preferred Jones would give not even a single second preference to Smith. Surely in realitv there must be some affinity between the political parties represented by Smith and Jones and at least some of the second preferences of Jones would go to Smith. . . , If a clear explanation of preferential voting is to be made the examples given should have some relation to reality and truth should not be sacrificed for a simplicity which only makes confusion more confounded. What would really happen is something like this: —(a) Group, 13 voting papers: Smith 13 first- preferences, Jones 9 second preferences. Robertson 4 second pre-, ferences, Robertson 9 third preferences, Jones 4 third preferences; and similarly with groups (b) and (c). If the parties in whose interest the candidates were standing had been stated thus: Smith (Labour). Jones (Independent), Robertson (Nationalist), the point I ant making would be more easily grasp-

ed. But it is when, your contributor endeavours to explain the “Manson” system that he makes his worst’blunder, as he states that Robertson was elected, which is an impossibility on the figures given bv him. See groups (a), (b), (c). This is how the example should have

Trial total 4S plus 34 plus 26 equals 108, which divided by 3 equals 36, which is the total number of voters. As Jones is the only one to obtain 36 preferences he is elected, and this :s the only possible result on the figures given whichever system of counting is used. If the New Zealand Legion were not more accurate and did not give more thought to the other proposals which it made some time ago then it is no wonder that it made no impression upon the _ political conscience of the electors of New' Zealand. —I am, etc., D. WILSON Wellington, August 26.

Preferred to: Jones. Robertson. Smith. Jones _ ' ~* 12 13 Robertson .. 24 — 13 Smith 23 — — ... — *■*-■*• Trial totals , 47 35 2C

read: — First. Trial Table. RobertJones son Smith Preferred to Jones .... — U 13 Preferred to Robertson 25 — 13 Preferred to .Smith ... 23 23 — 4S 34 2«