More on US Elec­tion

Fiji Sun - - Comment - Feed­back: jy­otip@fi­ Ja­son A John­son Ja­son John­son is a pro­fes­sor of po­lit­i­cal sci­ence and pol­i­tics edi­tor for The Root.

The Amer­i­can pres­i­den­tial elec­tion is the long­est, most ex­pen­sive and ar­guably most watched elec­tion on the en­tire planet. While the com­plex­i­ties of the Amer­i­can elec­toral sys­tem are not nec­es­sar­ily any worse than those of South Africa, or Mex­ico or the UK, the amount of money spent prog­nos­ti­cat­ing the elec­tion can al­most over­whelm any­one who is not in­ti­mately fa­mil­iar with the in­tri­ca­cies and nu­ances of the sys­tem.

Likely the most dif­fi­cult part of Amer­i­can elec­tions to un­der­stand is the flood of polls that come out pre­dict­ing which can­di­date is ahead, be­hind or more likely to win.

Some­times it is dif­fi­cult to de­ter­mine which are re­li­able and which are just more pre­elec­tion noise. As the Amer­i­can elec­tion sea­son winds down to the fi­nal days and hours here are three sim­ple rules for be­ing a good poll watcher of Amer­i­can elec­tions.

It’s State Not Na­tional Polls that Mat­ter

The US pres­i­den­tial elec­tion is a two-tiered sys­tem; there­fore the polls that you pay at­ten­tion to vary greatly in im­por­tance. There are polls for the na­tional pop­u­lar vote, which is how many Amer­i­cans vote for a par­tic­u­lar ma­jor party pres­i­den­tial can­di­date - in this case ei­ther Don­ald Trump, the Re­pub­li­can, or Hillary Clin­ton, the Demo­crat.

Then there is the elec­toral col­lege, which is es­sen­tially a points sys­tem whereby the pres­i­den­tial can­di­dates earn a cer­tain amount of points for ev­ery state where they win that state’s par­tic­u­lar pop­u­lar vote. Elec­toral col­lege votes are based on the state’s pop­u­la­tion, so for ex­am­ple Cal­i­for­nia has 55 elec­toral votes while Virginia only has 13. The win­ner of the pres­i­den­tial elec­tion is not ac­tu­ally who wins the most pop­u­lar votes; it is the can­di­date who can earn

270 elec­toral votes from the states. Usu­ally the win­ner of the pop­u­lar vote wins the elec­toral col­lege, but these two num­bers aren’t nec­es­sar­ily that cor­re­lated.

So, how does that change how you look at polling?

While the na­tional num­bers are im­por­tant to look at, to see if Trump or Clin­ton is lead­ing, it’s al­most more im­por­tant to look up how the polls are do­ing in so-called com­pet­i­tive “bat­tle­ground” states like Florida, Ohio and North Carolina. Clin­ton could win the na­tional pop­u­lar vote by the slimmest of mar­gins, say 1 per­cent, but if she wins big states like Ohio, Florida and New York she could get over 300 elec­toral col­lege votes and still be­come pres­i­dent.

Al­ways look at state polls be­fore con­sid­er­ing the na­tional num­bers. Now there is a cau­tion about state polls: they are not al­ways as re­li­able as na­tional polls. Na­tional polls taken by ma­jor Amer­i­can me­dia out­lets like

NBC, or CNN or Gallup have been tried and tested over the decades but state polling tends to be done by lo­cal univer­si­ties or small news­pa­pers.

When look­ing at a state poll, ex­am­ine whether or not it’s been taken by a univer­sity, or a lo­cal news sta­tion, or a pri­vate com­pany that may be funded by ei­ther of the two ma­jor par­ties. This will give you a bet­ter idea about that poll’s re­li­a­bil­ity.

Trends are more im­por­tant

In the last few days of the cam­paign it is easy to be overly ex­cited or overly con­cerned about one poll or another. For ex­am­ple, a Rasmussen poll came out this week show­ing Trump lead­ing the pop­u­lar vote by 3 per­cent, but 5 other polls by ma­jor firms showed Clin­ton with any­where from a 1 to 5 point lead.

These dif­fer­ences do not oc­cur be­cause poll­sters are frauds try­ing to tilt the race to one can­di­date or another, it’s

be­cause they all use dif­fer­ent data anal­y­sis tools.

Some poll­sters call mostly land­lines, some call mo­bile phones, some polling com­pa­nies use au­to­mated calls and some use live in­ter­view­ers. In other cases some polling com­pa­nies are bet­ter at get­ting re­sponses out of mi­nori­ties or young vot­ers, and in the end these dif­fer­ences in polling strat­egy and col­lec­tion af­fect the fi­nal polling out­comes.

Due to this vari­a­tion in polling it’s best to not take any in­di­vid­ual poll too se­ri­ously and in­stead look at sev­eral polls cov­er­ing the same state, or the na­tional race and see what the re­sults are. Pop­u­lar Amer­i­can polling sites that ag­gre­gate dozens of polls from across the coun­try are FiveThir­tyEight and Real

ClearPol­i­tics. These sites al­low you to look across sev­eral dif­fer­ent polling com­pa­nies, states and of­ten times pro­vide anal­y­sis for how to read and weigh their polling as well.

Up­sets In Amer­i­can Pres­i­den­tial Elec­tions Are Not Com­mon

The pres­i­den­tial elec­tion may seem ex­cit­ing with var­i­ous twists and turns but un­like sports and ac­tion films there are very few sur­prises and up­sets.

In fact even this elec­tion year fea­tur­ing two of the least pop­u­lar can­di­dates in Amer­i­can his­tory, the pref­er­ences of the vot­ers have been sur­pris­ingly sta­ble. The can­di­date who wins the pop­u­lar vote usu­ally wins the elec­toral col­lege. There has only been one elec­toral col­lege pop­u­lar vote split in the last 120 years of Amer­i­can pres­i­den­tial elec­tions, and that boiled down to a con­tro­ver­sial recount of votes in Florida, a state then gov­erned by the brother of even­tual win­ner Ge­orge W Bush. The can­di­date who is lead­ing in the polls head­ing into Elec­tion Day usu­ally wins the elec­tion. Even if the num­bers

ap­pear to be close, “un­com­mit­ted” or un­de­cided vot­ers tend to break for the in­cum­bent party. It has be­come trendy for Amer­i­can po­lit­i­cal pun­dits and even some Trump sup­port­ers to cite the Bri­tish “Brexit” vote as an ex­am­ple of a vote where all of the polls were wrong and a right-wing can­di­date/idea up­set the estab­lish­ment. While on the sur­face this ap­pears to be cor­re­la­tion to Trump break­ing the estab­lish­ment just like Nigel Farage and Brexit the sys­tems and the elec­tions are vastly dif­fer­ent.

First, Brexit was trending to­wards pass­ing for the days and weeks head­ing into the fi­nal vote, and when polls are trending in one di­rec­tion that is a safe bet that they may con­tinue. As of right now there is some sta­bil­ity in the na­tional pop­u­lar vote in America even as the state num­bers are more volatile.

Fur­ther, Brexit was a once-ina-gen­er­a­tion vote that poll­sters may have been ill-equipped to prop­erly track.

Pres­i­den­tial elec­tions hap­pen ev­ery four years. Poll­sters at Amer­i­can news net­works and po­lit­i­cal cam­paigns spend bil­lions of dol­lars to make sure there are no sur­prises. Fi­nally, re­mem­ber the points sys­tem. For a can­di­date to pull off an up­set in the Amer­i­can sys­tem the na­tional polls and lit­er­ally dozens upon dozens of state polls would all have to be wrong and likely all wrong in the same di­rec­tion. That kind of sys­tem­atic er­ror is highly un­likely.

The full an­swer to all polling will be re­vealed on Tues­day. Ei­ther Clin­ton or Trump will be the next pres­i­dent of the United States.

Con­stant ob­sess­ing over polls won’t make the days go by any faster but they may pro­vide you with some re­lief, or anx­i­ety de­pend­ing on which can­di­date you sup­port.

Newspapers in English

Newspapers from Fiji

© PressReader. All rights reserved.