Another close US election

IN AN earlier piece (“Predicting the US election,” 9/8/2012), I pointed out that, since the President of the United States is not elected by the voters directly but by the electoral college votes (ECVs) of the 50 states and the District of Columbia (DC), the way to predict the outcome is to use the 51 statewide polls, not the national polls.

The 22 states that went consistently Republican—colored red in the TV maps—in the three past presidential elections were: Alabama, Alaska, Arizona, Arkansas, Georgia, Idaho, Kansas, Kentucky, Louisiana, Mississippi, Missouri, Montana, Nebraska, North Dakota, Oklahoma, South Carolina, South Dakota, Tennessee, Texas, Utah, West Virginia, and Wyoming.  These states, with 180 ECVs, all appear safe for Mitt Romney, according to the statewide polls.

The 19 states, counting DC as a “state,” that went consistently Democratic, or “blue,” in the three past presidential elections were: California, Connecticut, Delaware, DC, Hawaii, Illinois, Maine, Maryland, Massachusetts, Michigan, Minnesota, New Jersey, New York, Oregon, Pennsylvania, Rhode Island, Vermont, Washington, and Wisconsin. These 19 states, with 242 ECVs, all appear safe for Barack Obama now.

Michigan and Wisconsin were called tossups two months ago.  But Obama got +7 in Michigan on Oct. 6-8, in a Detroit News/WDIV poll conducted by the Glengariff Group (R), with error margin +/- 4.  He also got +5 in Wisconsin on Oct. 24-25, in a state poll by Grove Insight for Project New America and USAction (D), with error +/- 4.4, and got +6 on Oct. 15-17, as polled by NBC News/Wall Street Journal/Marist Poll, with error +/- 3.1. (My source for state polls is the subscription section of PollingReport.com.)

The remaining 10 states, namely Colorado, Florida, Indiana, Iowa, Nevada, New Hampshire, New Mexico, North Carolina, Ohio, and Virginia, with a total of 116 ECVs, all went blue in 2008, giving Obama an easy victory over John McCain.  In 2012, however, at least one of them, namely Indiana, looks likely to swing Republican.

As of Oct. 29, the Huffington Post (elections.huffingtonpost.com) scored the race as Obama 277 (of which 217 strong), Romney 191 (all strong), with only five states, namely Florida, North Carolina, Virginia, Colorado, and New Hampshire, as tossups.  That would put Obama already past the 270 minimum he needs to win a second term as president.

Apparently, Huffington assigned Ohio (18 ECVs), Iowa (6), Nevada (6) and New Mexico (5) to Obama, and Indiana (11) as going to Romney, to which I would agree, except for Ohio and Nevada.  With Ohio and Nevada as tossups, I would subtract 24 ECVs from Huffington’s score, resulting in Obama 253, Romney 191, with 94 ECVs from seven states where the outcome is still statistically uncertain.  This implies that Obama still needs to win 17 more ECVs from the seven states.

In Ohio, Obama was +4 on Oct. 23-25 in a CNN/ORC Poll, with error +/- 3.5, but was only +2 on Oct. 23-25, as polled by American Research Group, error +/- 4.

In Nevada, Obama was +3 on Oct. 23-24 in an NBC News/Wall Street Journal/Marist Poll, error +/- 3, and was +2 on Oct. 19-22 in a poll by American Research Group, error +/- 4.

In Florida (29 ECVs), Obama was at +2 on Oct. 23-24, in a poll by Grove Insight for Project New America and USAction (D, for Democratic pollster), error +/- 4.4; and in a dead heat on Oct. 18-21, in a poll by The Mellman Group for Americans United for Change (D), error +/- 3.4.

In North Carolina (15 ECVs), Obama was at +3 on Oct. 23-24, in a poll by Grove Insight, error +/- 4.4; and at +1 on Oct. 20-21, in a Civitas Institute poll conducted by National Research (R, for Republican pollster), error +/- 4.

Virginia (13 ECVs) has mixed results. Obama was at +4 on Oct. 22-26, in a Washington Post Poll, error +/- 3.5; but Romney was at +2 on Oct. 23-24, in a Fox News Poll conducted by Anderson Robbins Research (D) and Shaw & Company Research (R), error +/- 3.

In Colorado (9 ECVs), Obama was +3 on Oct. 23-24, in a poll by Keating Research/Onsight Public Affairs (D), error +/- 4.4; and in a dead heat on Oct. 23-24 in a NBC News/Wall Street Journal/Marist Poll, error +/- 2.9.

In New Hampshire (4 ECVs), Obama was +3 on Oct. 24-25, in a poll by Grove Insight for Project New America and USAction (D), error +/- 4.4.

Since Obama has the advantage in at least six of the seven battleground states, I think he is the slight favorite to win.  The reason he cannot yet be judged as a statistical winner is because the statewide sample sizes are not large enough to warrant smaller error margins.  To reduce the error margin to only +/- 2 percent, for instance, would require a sample size of about 2,500 voters per state.

Fil-Ams are mostly for Obama.  In the last few days, there have been some misrepresentations in the media of the 2012 Survey of Asian Americans by the Pew Research Center (www.pewsocialtrends.org) as allegedly finding that Filipino-Americans are mostly pro-Republican.

In fact, the Pew Report, “The rise of Asian Americans,” cites 43 percent of Filipinos in the United States as identifying themselves as Democrat or leaning Democrat, versus 40 percent identifying themselves as Republican or leaning Republican.  As a whole, US Asians are 50 percent Democrat or leaning Democrat, versus 28 percent Republican or leaning Republican.

*  *  *

Contact SWS: www.sws.org.ph or mahar.mangahas@sws.org.ph.