Scott McConnell, associate director of LifeWay Research, said LifeWay Research decided to ask this question after European headlines used the phrase "dangerous religion" to describe results drawn from a 2008 study across 21 European countries that found an "overwhelming majority" of people believe immigration from predominantly Muslim countries poses a threat to Europeans' traditional way of life.

"It appears that Protestant pastors in America are overwhelmingly willing to use that phrase and cite Islam as a 'dangerous religion,'" McConnell said.

Additionally, a study by the Pew Research Center found that 38 percent of all Americans say Islam is more likely to encourage violence than other religions. However, studies also indicate a need for interaction. For example, data from the Gallup Muslim-West Dialogue Index shows that when given the option of labeling greater interaction between Muslim and Western worlds a threat or a benefit, 70 percent of Americans call it a benefit.

"It's important to note," Stetzer pointed out, "our survey asked whether pastors viewed Islam as 'dangerous,' but that does not necessarily mean 'violent.' 'Dangerous' can be defined in a variety of ways, including from the perspective of spiritual influence. Regardless of the definition, the numbers tell us that Protestant pastors are concerned."

The LifeWay Research study found six statistically significant differences in the belief about Islam statement among pastors:

* Mainline denomination pastors are less likely than evangelicals to say Islam is "a dangerous religion." While 77 percent of evangelical pastors either somewhat or strongly agree Islam is dangerous, only 44 percent of mainline pastors feel the same way, and 38 percent strongly disagree.

* More educated pastors are less likely to agree than those with less education. While 64 percent of pastors with a bachelor's degree or less strongly agree Islam is dangerous, only 37 percent with a master's degree or more feel the same way, and 25 percent of those strongly disagree.

* The majority of pastors affiliated with the Democratic Party are more likely to strongly disagree than Republicans or Independents. While 61 percent of Republicans and 40 percent of Independents strongly agree Islam is dangerous, only 16 percent of Democrats feel the same way, and 52 percent of Democrats strongly disagree.

* Older pastors are more likely to strongly agree than any other age group. While overall agreement differs little by age, 58 percent of pastors age 65 and older strongly agree about the danger of Islam, contrasted with 42 percent of pastors ages 50-64, and 44 percent of pastors younger than 50.

* Rural and smaller city pastors are more likely to agree than pastors in large cities and suburbs. A full 51 percent of rural pastors and 47 percent of small-city pastors agree that Islam is dangerous, while 37 percent of suburban pastors and 39 percent of large-city pastors feel the same way.

* Politically conservative pastors stood in starkest contrast with politically moderate and liberal pastors. Among very conservative pastors, 78 percent strongly agree about the danger of Islam and 55 percent of conservative pastors feel the same way, contrasted with 69 percent of liberal or very liberal pastors and 38 percent of moderates who strongly disagree.

The Pew study, conducted in August 2009, asked more than 2,000 adults in the United States whether Islam is more likely to encourage violence than other faiths. While 38 percent say yes, views on the subject have fluctuated in recent years. Similar Pew studies found 25 percent answered yes in 2002, 36 percent in 2005 and 45 percent in 2007.

The Gallup study was commissioned for the World Economic Forum, and released as "Islam and the West: Annual Report on the State of Dialogue."

-----The LifeWay Research project, entitled "Protestant Pastors' Reaction to a Statement on Islam," was conducted by telephone in October 2008. The random sample of 1,002 senior pastors was weighted to reflect geographic distribution and the sample provided a 95 percent confidence that sampling error did not exceed ± 3.1 percent, although margins of error were higher in subgroups.