Ten weeks before the November elections, American adults are sullen if not surly, a national poll conducted by Allstate Insurance and the National Journal has found.

Results from the companies' sixth Heartland Monitor Poll, released Friday, "paints a portrait of a populace increasingly convinced that the country is on the wrong track, and unlikely to right itself soon," the companies said.

The pall over the public is a result of personal circumstance and that of family and friends. There also is a palpable sentiment that the public wants something done about the lack of jobs, even if it means tolerating further government spending and/or action.

"They continue to be less confident in our economy, institutions and leadership. As a result, they will spend less and save more," said Thomas J. Wilson, Allstate chairman, president and CEO.

"This survey captures the continued coalescing of a prickly 'back- to-basics' streak among many Americans," said Ron Brownstein, editorial director of National Journal Group. "They increasingly believe that individual financial irresponsibility is a key factor in the economic mess and must be corrected for the economy to recover. There is also a strengthening chord of 'America First' sentiment, showing itself in greater anxiety about outsourcing and increased support for nationalist economic policies. Combined with a decreasing faith in major government and business institutions, unease about the prospects for the next generation, and deep divisions over the proper role for government in the economy, these attitudes explain why American politics could remain volatile long beyond the November election."

Among the key findings of the survey:

* 62% now say the country is on the wrong track, up from 54% in April. 40% now believe the economy will worsen over the next 12 months, up from 27% in April.

* The President's job approval rating is 46%, down two points since April, and 10 points since last July. The gap between the President and Congressional Republicans on who is more trusted for economic solutions has narrowed to five points - 42% for the President versus 37% for Republicans in Congress. The President held an advantage of seven points in April, eight in January, and 21 in September 2009.

* 53% say their personal experiences, and those of family and friends, have the most influence on how they gauge the economy, compared with 43% who consider statistics such as the unemployment rate, the GDP, or the stock market. A total of 21% say they are looking for work, and 70% say a close friend or family member have been laid off since the beginning of the recession in late 2007.

* 52% say they support tax incentives for business to hire more workers, even if it increases the deficit. 61% say government should play an active role in the economy, though more than half of those express skepticism that they can trust government to do it effectively. A total of 68% support either government support for critical industries (32%) or protectionist measures like tariffs and penalties for outsourcing (36%), versus 23% who support a pure free trade policy.

* Americans believe individual responsibility is important to help the U.S. economy recover from the recession. Choosing among individual actions, 83% say reducing personal debt is extremely or very important, 81% say the same about financial literacy, 73% identify saving more, and 72% cite additional education and training as extremely or very important to help the U.S. economy recover from the recession. Among government actions, 80% say it's extremely or very important to make it easier for small businesses to start and grow, 74% point to helping workers get more education and training, and 70% prioritize reductions in outsourcing. Among possible business actions, the highest-rated is to shift more operations from overseas into the U.S. (66%), followed by working with the government to develop new economic growth ideas (58%).

The survey was conducted via telephone between August 27 and 30 among 1,201 adults 18 or older. A sample of 1,200 is normally considered the minimum to obtain a statistically valid national survey. The margin of error is +/-2.8%.