The Bureau of Transportation Statistics found that many Americans have different Thanksgiving plans than they did a year ago (Figure 1 & 2). In a new BTS survey, about 18 percent of adults in the U.S., or 36 million people, said they plan to travel this Thanksgiving. That's down from the 29 percent, or 58 million people, who said they traveled last Thanksgiving.
More than half the people, about 30 million, who traveled last year said they do not plan to travel this year. Almost 10 million who did not travel last year plan to travel this year, a net loss of about 20 million Thanksgiving travelers.
About one-fourth of U.S. adults, about 51 million people, had Thanksgiving holiday travel plans before the terrorist attacks of September 11. About 18 percent of the prospective travelers or about 9 million people changed their plans because of the events of September 11 and subsequent events.
About a half million of those changing their minds currently plan to travel for Thanksgiving while 8 million do not currently plan to travel for the holiday.
About 15 percent of travelers, or 5.4 million people, plan to fly commercially (Figure 3). More than four out of five travelers will use no other form of transportation than a personal vehicle.
The BTS survey found a sharp difference between gender groups concerning Thanksgiving travel. About 62 percent of the 12 million who changed their Thanksgiving holiday travel plans because of the events of September 11 were female even though females make up only 52 percent of the population.
Data presented in this issue of OmniStats are from the February 2002 BTS Omnibus Household Survey. Data for the survey were gathered from approximately 1000 randomly selected telephone households. The response rate for the February 2002 survey was 46.2 percent. The survey has a margin of error of plus or minus two percent. The data are weighted to allow inferences about the noninstitutionalized population, aged 18 years or older who are currently living in the United States. Because the data come from a sample, they are subject to both sampling and nonsampling error. Sampling error may affect the reliability of an estimate. Nonsampling error can affect both the reliability and accuracy of an estimate and is more difficult to measure. The most general categories of nonsampling error are coverage error, nonresponse error, response error, processing error, and estimation error. For additional information about the survey, please contact Liz Grossman, firstname.lastname@example.org. Press contact: Dave Smallen, email@example.com.