With the convention in their rearview mirrors, President Obama enters the homestretch with a sizable lead over Mitt Romney among likely voters in New Jersey, according to the most recent survey by Fairleigh Dickinson University’s PublicMind. The poll, conducted after the Democratic convention concluded, found that 52% of likely voters say they intend to vote for President Obama, with around four-in-ten (38%) who align themselves with his opponent. The remainder say they’re supporting someone else (1%) or are unsure (9%). 

“Obviously this is good news for the president,” said Krista Jenkins, director of the poll and professor of political science at Fairleigh Dickinson University. “Even though he remains saddled with the perception among almost half of all likely New Jersey voters that the country is on the wrong track, he remains favored by double digits in the state.” 

While support for the two major party candidates is split across party lines, Obama’s lead is greater than the edge that self-identified Democrats have over Republicans among likely voters (47 vs. 37%). 

Underscoring Obama’s appeal is the percentage of respondents who say they are “very certain” about their expected choice in November. Nine-in-ten (90%) report being unlikely to change their mind, a number that is identical to that found among Romney supporters. 

The survey also provides insight into the effect, or lack thereof, that future economic trends will have on the presidential race. In an experiment, respondents were told that economists predicted either a slight increase or decrease in the unemployment rate before the election. Half were selected at random to be asked a question about unemployment increasing, while half received the question that noted the possibility of unemployment decreasing. Following this question, respondents were asked about their candidate preference.

As it turns out, the trajectory of expected unemployment makes little difference for whom someone says they support. When it’s suggested that unemployment will increase before Election Day, about half (49%) support the president. This figure increases insignificantly to 53% when told unemployment will drop before the election. A similar, but reverse effect, can be seen among Mr. Romney supporters. His support is 39% when the question indicates an increase in unemployment, and it decreases slightly to 36% in the question that references a decrease in unemployment.

“These numbers suggest that Garden State voters accept the inevitability of tough economic times, at least for the foreseeable future, and are largely unswayed by short term economic trends,” said Jenkins.

The experiment also provides insight on who voters identify as largely responsible for the high unemployment numbers. Regardless of whether unemployment was referenced as likely increasing or decreasing before Election Day, voters identify a number of usual suspects.  About three-in-ten (29%) believe President Obama and/or Democrats are to blame, and slightly fewer than a quarter (23%) identify the Bush administration and/or Republicans. The rest are convinced Congress (7%), banks and financial institutions (8%), and other political players (19%) are responsible, with the remainder simply unsure who’s to blame (14%).

Other notable findings include the persistence of the gender gap. Women remain significantly more likely than men to endorse President Obama in the election (56 versus 48%), and approve of the job he’s doing as president (56 vs. 47%). And young people are also significantly more likely to support President Obama as compared to those from older age cohorts (67 vs. 45, 52, and 52% respectfully).

“Regardless of where you look today, youth support is key,” said Jenkins. “The President will likely need the younger vote on Election Day, just as he did in 2008.  At least in New Jersey, it looks like he’s doing well with Millennials.” 

The Fairleigh Dickinson University poll of 706 likely voters statewide was conducted by telephone using both landlines and cell phones from September 6 through September 12, 2012, and has a margin of error of +/-3.8 percentage points.