Yahoo announced in June that it would switch from the Inktomi search engine to Google for many Internet user queries. Despite the implications of the new report, a Google spokesman denied that the business deal caused Google to deliberately boost the rankings of Yahoo directory pages above others.
The author of the study, Eric Rumsey, is a medical librarian at the University of Iowa's Hardin Library for the Health Sciences.
Rumsey publishes a hand-selected index to reliable medical information sites on the Web, known as the Hardin Meta Directory, or Hardin MD.
Rumsey's study examined Google's searches on 22 medical topics--ranging from heart disease to pregnancy--over a nine-month period beginning in November 1999.
"In November, Google not only found the same high-quality directories that we have found, but often it even ranked them in the same relative order that we did," the study said.
But just before the Google-Yahoo alliance was announced, Rumsey said, "It was obvious that Yahoo was dominating the Google rankings."
Last November, searches on medical topics such as "cardiology," "gynecology" and "obstetrics" resulted in Yahoo pages being ranked by Google between 50th and 100th from the top, according to a chart in the study.
But by August, the same topics and similar searches at Google resulted in Yahoo directory pages being ranked by Google in the first, second or third position.
Rumsey cites SciCentral, a science news and links site, as an example of a high-quality directory of medical information that has lost visibility at Google.
SciCentral consistently appeared in Google's top 100 hits in medical-related searches last year, but now it's not even in the first 500, Rumsey said.
"Not only did Yahoo rise, they've got six or seven (hits) in the top 100," he said.
Google spokesman David Krane said they didn't do anything to give preferential treatment to any customer, but he did offer two reasons why the Yahoo pages have risen.
One reason is the sheer number of pages in the database. According to Krane, Google indexed only 135 million Web pages in November 1999, when the study began. By June 2000, just before the study's final sample, Google's database had grown to more than 1 billion pages, including many more Yahoo pages.
The other is based on the nature of Google's search engine, which ranks sites by the number of links it gets from other sites. The ranking formula, which is proprietary, favors links from "authoritative" or well-respected sites.
"Our PageRank technology recognizes Yahoo as one of the most important domains on the Web," Krane said. "Since this is the case, it's only natural that Yahoo Web pages would show up near the top."
Whether Google's high rankings for sites like Yahoo are an unintended side effect of its expansion or a deliberate change in Google's formula, consumers may wonder which search engines they can trust.
"Since Google started increasing its index size, its results haven't been as good as I've felt in the past," said Danny Sullivan, editor of Search Engine Watch, which independently evaluates search services.
Sullivan hastened to add, however, "They've still been much better than other search engines."
Greg Notess of Search Engine Showdown said his own independent rankings show potential problems with sheer index size. For example, he said, Google doesn't identify as many unique sites as other search engines that index the same number of pages, such as Fast Search.
So which engine should consumers choose for their Web searches? While Notess recommends Yahoo for beginners, he says experienced surfers should use two or three different search engines, including Fast Search and AltaVista, instead of simply going ga-ga over Google.
Consumer advocate Brian Livingston appears at News.com every Friday. Do you know of a problem affecting consumers? Send info to tips@BrianLivingston.com. He'll send you a book of high-tech secrets free if you're the first to submit a tip he prints.