Center for Advanced Management




Prof. Jochen Wirtz (University of Singapore)


  • Datum: 11.07.2012
    Zeit: 16:00 - 17:30
    Ort: Lehrturm, Prof.-Huber-Platz 1, Raum U 104

From switching intent to behavior: A construal level theory perspective in the context of contractual services

Measures of repurchase intent are often not good predictors of actual repurchase behaviors. A critical distinction between intention and actual switching is that the former is a response to a hypothetical situation, whereas the latter is an action that is actually carried out. Recent research on construal level theory (CLT, see Trope and Liberman 2007 for a review) shows that when an event feels psychologically far away (e.g., when the event is hypothetical), people tend to focus on “central” aspects of the event – i.e., aspects that give direct implications to the desirability of the event. According to CLT, these aspects constitute the “higher level construal” of the event. On the other hand, when an event feels psychologically near (e.g., when it is going to happen in the near future), people tend to focus on aspects that are associated with the procedures involved in carrying out the action (e.g., practicality, feasibility considerations). These aspects constitute the “lower level construal” of an event.

We first show empirical evidence that monetary incentives, compared with non-monetary switching costs, have a stronger influence on repurchase intent. Non-monetary switching costs, however, play a significantly more important role in determining actual switching behavior. We then conducted a series of experiments to demonstrate how the construal level theory explains this discrepancy. More specifically, we show that monetary incentives constitute a higher level construal, whereas non-monetary switching costs constitute a lower level construal. Moreover, based on recent findings that process simulation (i.e., asking respondents to visualize the step-by-step process of carrying out a certain action) can highlight feasibility aspects of an event and encourages a lower level construal of the event, we suggest that process simulation can be employed in surveys for reducing the inconsistency between responses to intention measures and actual switching behaviors. In one experiment, we recruited participants for an online marketing survey. Participants first registered for an easy, short survey that would take place few weeks later. A few days after the registration, they were contacted and informed that another survey that was more difficult and longer but with a higher payoff was also available to them. They were asked to indicate their intention to switch from the low-effort, low-payoff survey to the high-effort, high-payoff option. To evaluate the extent to which process simulation reduces inconsistency between intention measures and actual behavior, we also administered a “process simulation” condition, in which participants were instructed to imagine going through the survey step-by-step (i.e., to simulate the process) before they indicate their switching intention.
Finally, the online survey was open to participants, where they actually decided whether to switch or not, and completed the survey they chose. Our finding showed that process simulation lowered the intention to switch and reduced the inconsistency between switching intention and actual switching behavior. The result supported our prediction. We replicated this finding in another experiment, where participants indicated their intention to attend a paid lab study. Again, for half of the participants, they thought about the steps in the procedures before indicating their intentions. Our finding showed that process simulation made people consider the feasibility issues, such as time and effort to travel to lab, so that their indication is a better predictor of their actual behavior. Our findings suggest the possibility of an unobtrusive administration of procedure simulation in surveys that will improve the predictive power of intention measures of actual switching behavior. These findings have potential applications beyond the immediate switching behavior application to any intent-actual behavior measurement context.