To calculate these indexes, the Bureau of Transportation Statistics used ASQP data that airlines report monthly on both scheduled and actual flight times (based on gate-departure and gate-arrival). The data cover all domestic nonstop flight segments flown by U.S. carriers with at least 1 percent of passenger revenues in the previous year. The ATTVI is measured using standardized deviations of actual air travel time from its average value. Deviations are weighted to increase the variability index more for extreme deviations than for small deviations. Both the ATTVI and the ATTI are designed to control for changes in carriers, routes, and time of day in order to improve comparability over time. Airports included in the analysis are those that ranked in the top 50 (by passenger enplanements of all large certified carriers) for at least one year between 1990 and 2004. Analysis of the time-of-day for departure is based on four periods: morning offpeak (before 9 a.m.); mid-day peak (between 9 a.m. and 3 p.m.); evening peak (between 3 p.m. and 9 p.m.); and evening offpeak period (after 9 p.m.). Analysis of flight distance is based on three categories: 500 miles and less; 501 to 1,000 miles; and more than 1,000 miles.