5. THIRD STAGE SURVEY
The third stage survey applied the conjoint analysis to determine relative preferences for the limited set of critical service attributes identified at the end of the second stage survey. There have been a number of methods to measure the respondents' relative preferences in the conjoint experiments. The respondents can be asked to rate/rank the bundles to indicate their degree of preference for each bundle or choose only a single bundle from each set of service alternatives. Our conjoint experiments adopted the stated choice conjoint paradigm by presenting respondents with two service bundles
at a time and asking respondents to select the most preferred bundle between the two . The models characterizing shippers' choice behavior assumed the well-known Binary Logic formulation as follows: