=20 ---------------------------------------------------------------------- This article was sent to you by someone who found it on SF Gate. The original article can be found on SFGate.com here: http://www.sfgate.com/cgi-bin/article.cgi?file=3D/chronicle/archive/2002/05= /05/TR2006617.DTL ---------------------------------------------------------------------- Sunday, May 5, 2002 (SF Chronicle) Annual airline quality report offers ratings that don't fly Ed Perkins With due press coverage, the professors at the University of Nebraska at Omaha and Wichita State University have issued their annual Airline Quality Ratings for 2001, and overall scores for the industry improved a bit. Among individual airlines, the winner was Alaska, with now-defunct TWA at the bottom of the list. The composite AQR scores measure how well airlines did - collectively and individually - in avoiding problems. But I'd never base my choice on those scores: They exaggerate small differences among problem areas, and they omit several vital elements of the total quality picture. The two universities have been compiling their AQR scores for more than a decade. The original scoring included some 20 separate factors, many of which had nothing at all to do with the "quality" of an airline as experienced by its customers. Now the index is based on only four critical factors: On-time performance, denied boardings (bumpings), mishandled baggage and numbers of complaints received by the Department of Transportation's consumer office. Each received roughly equal weighting in the composite score. In descending order, the ranking for 2001 was Alaska, US Airways, Northwest, Southwest, Delta, American, America West, Continental, United, American Eagle and TWA. Among the more important lines AQR did not score were AirTran, American Trans Air, Frontier, newcomer Jet Blue, and perennial "best airline" winner Midwest Express. Given that its four criteria are all important to travelers, why do I disagree with AQR? Two reasons: First: Three of its four rating elements are based on the numbers of failures rather than the numbers of satisfied travelers. Since the numbers of failures are very small compared to numbers of travelers, scoring on failures grossly inflates minor performance differences. Here's an example: On average, the airlines bumped about eight out of every 100,000 boarding passengers, with a high of 18 for TWA and a low of about four for America West and American. According to scoring by failure, then, TWA was more than four times worse than America West or American. But if you look at success rates, the differences were tiny: TWA did not bump 99.982 percent of its travelers, compared with 99. 996 percent on America West and American. That difference is so small that it =1Fwouldn't even show up as a blip on a typical chart. Similarly, scores based on complaints (an average of two per 100,000 travelers) and mishandled bags (one out of 200) inflate minor differences. Only the on-time arrival measures are based on success rates. Second: AQR entirely omits what to me are significantly more important "quality" factors. As you must surely be aware by now, I believe that seating is by far the most important element of airline quality, and AQR =1Fdoesn't include seating measures (although it could). Food service is important to some travelers, and the number of available frequent flyer seats is critical to many others. And to lots of consumers, easy availability of low fares is surely the most important of all quality issues. To me, three domestic airlines stand head and shoulders above the others: -- Midwest Express, for offering near-first-class seating and food servi= ce at coach prices. -- Among the big lines, American's extra room in coach puts it far ahead of the pack. -- Southwest, for its widespread availability of low fares - at the last minute as well as in advance. If Midwest Express, American and Southwest flew everywhere you wanted to go, you'd have no need for any of the other lines. Unfortunately, the AQR scores would never lead you to that conclusion. E-mail Ed Perkins at eperkins@mind.net=20 ---------------------------------------------------------------------- Copyright 2002 SF Chronicle