How Tasks and Gender May Influence How People Google

Do men and women use search engines differently?

How might the kind of activity you are engaged in affect the way you would interact with search results?

Researchers from Cornell and Stanford University put out a paper last year that has some intriguing statistics in it – The Influence of Task and Gender on Search and Evaluation Behavior Using Google (pdf). I haven’t seen much on the Web that cites this paper, so I thought it might be worth pointing out. Here’s a snippet from the introduction:

Using eye tracking, we extend this understanding by analyzing the sequences and patterns with which users evaluate query result abstracts returned to them when using Google. We find that the query result abstracts are viewed in the order of their ranking in only about one fifth of the cases, and only an average of about three abstracts per result page are viewed at all*.

*My emphasis

Their research also seems to indicate that “gender and task significantly influence different kinds of search behaviors.”

Searching Tasks

What kinds of tasks did the study look at, and evaluate?

The researchers considered some earlier papers to see what tasks other researchers defined when it came to search, and settled on three sets of objectives as defined by Andrei Broder, which focus upon the “need behind the query.” These three are:

1. Navigational tasks, where the searcher is trying to find a particular web page, but they may not know the URL for the page.

2. Informational tasks, where the searcher is attempting to find information about a specific topic, and;

3. Transactional searches, where the searcher is attempting to perform some action, such as making an online purchase.

These three types of tasks have been cited by others in describing people’s behavior on the web.

Gender and search

There have been a few studies that look at different behavior on the web based upon gender, but none that have attempted to look at both gender and the kinds of tasks described above in the context of how people use search engines.

The Study itself

Eye tracking methods as well as log file analysis were the tools used in the study to try to see if there were different types of behavior based upon gender and task type. They initially selected 36 undergraduate students from a large North Eastern University to participate in their study, but limited their results to 23 of the subjects because they only had complete eye tracking information for those students.

Their subjects looked at more than 400 queries, and 600 Google results pages.


While an interesting study, it seems to be more of an initial foray into testing how people interact with different search results pages than something that one could draw significant conclusions from.

It was interesting to see Google acknowledged as partially funding the research.

Their conclusion notes some additional ideas and research to explore, and they state that such studies would likely be fruitful.

If you like some of the previous studies that have come out which look at Google search results with eye tracking, like one on Search’s Golden Triangle, you might find this study interesting. Especially the eye tracking scan path picture that appears on page 23 of the report.

3 thoughts on “How Tasks and Gender May Influence How People Google”

  1. As a matter of fact, gender use behaviour could be predicted from their site navigation behaviour. Genders think differently, so it tells everywhere.

    I don’t remember the exact researches done on gender site navigation behaviour, but you can probably recall a discussion “Where women go on the Net” at Cre8asite. It has quite a few interesting pieces about gender-specific information.

  2. I’m digging into some of the results a little deeper, from this study, and looking to see if there has been much followup. I haven’t seen a lot of citations of this paper yet, and I think that it’s worth more people looking over, and thinking about.

    Some of the eyetracking studies that have come out before about search don’t seem to have done much based upon different tasks and gender and other potential differences, and that seems to have been a weakness of those. So, this is good to see.

Comments are closed.