Searcher Habits

An interesting AP story: Experts: Web searches for sex declining, e-commerce increasing that reports on some research findings published in a new book titled, Web Search: Public Searching of the Web by noted information scientists, Dr. Amanda Spink (University of Pittsburgh) and Dr. Bernard J. Jansen (Penn. State University). By the way, their web sites link to plenty of interesting material.

The AP article discusses their research finding that people are searching for e-commerce more and sex less.

"Twenty percent of all searching was sex-related back in 1997, now it's about 5 percent..."It's a little bit more in Europe, 8-10 percent, but in comparison to everything else, it's a very small percent," Spink said.

This makes sense since e-commerce has boomed in the past seven years.

What I find most interesting (and somewhat sad) is that their research shows searcher habits haven't changed much in the past seven years.

What hasn't changed much in seven years is how hard people are willing to work at searching. The answer: Not very. Spink and Jansen found that people averaged about two words per query and two queries per search session.

"The searches are taking less than five minutes and they're only looking at the first page of results," Spink said. "That's why people are wanting to get their results on the first page" of search engine results.

Spink goes on to say, "We were surprised that people weren't doing more complex searches... If you put a couple of words into the Web, you're going to get hundreds of thousands of results. I think people aren't trained very well to use the search engines."

Dr. Spink is right on the money. Now, how do we solve the problem? I wish I knew.

Some random thoughts.

It's not only one of training but also one of marketing and human nature.

For example, people can't search a vertical database or another web engine if they don't what's available and that it could save them TIME and effort in the long run. Perhaps people are searching in the wrong place but believe that one database offers it ALL. Even if this was the case, most searches don't utilize any of the advanced features web engines currently offer which can help create more precise results. I recently heard that about 98% of the searches at major web engines don't use any advanced syntax or search features.

We can clearly see with tools like Jux2 that the amount of overlap between major web databases isn't as great as people might think and that it's possible for a user to find a better answer by using more than one engine. Those of us in the library world are always reminding people that for some types of searches it's always a good idea to use more than one web engine.

Perhaps tools like Clusty/Vivisimo and other clustering tools along with federated/metasearch, answer engines, and search personalization will help. Of course, if the searcher doesn't know what they're searching for in the first place, personalized results and they're the persons past search behavior might not be very helpful. Maybe one day some search tool(s) will offer an interactive computer-based type of "reference interview."

However, the biggest issue and the one that's very difficult to change, is human nature. When it comes to research, this is what author and librarian Tom Mann calls the principle of least effort.