Are ten participants enough for evaluating information scent of web page hyperlinks?

Abstract

Information scent of hyperlinks, that is the user's assessment of semantic relevance of navigation options in a webpage, has been identified as a critical factor in Web navigation. An important question in this context is to identify the minimum number of participants required to measure reliably information scent. A two phase study was conducted in an attempt to provide an answer to this question. In the first phase, involving 101 participants, ratings produced by different size subsets of participants were compared to those of the whole set. In the second phase, the ratings of these different size subsets of participants where compared with measures of behavior of 54 participants, who performed the same information navigation tasks using a typical web browser. Results indicate that representative estimates of information scent can be obtained from 10 participants in both cases. This finding has important implications for future scent-related studies. © 2009 Springer.