The tale of uncertain choices: inclusion versus exclusion

Abstract

This article investigates the effect of perceived uncertainty on two types of screening strategies – exclusion and inclusion. Results from five studies showed that perceived uncertainty increases the preference for exclusion strategy. Using both mediation and moderation methods, we find support for accuracy consideration as the driver underlying the effect of perceived uncertainty on screening strategy. Furthermore, we demonstrated a reversal of the relationship between perceived uncertainty and screening strategy by manipulating the consideration set size. While perceived uncertainty strengthened the preference for the exclusion strategy when forming smaller consideration sets, this relationship flipped when forming larger consideration sets. We also demonstrated the effects of perceived preference uncertainty on screening strategy over and above knowledge uncertainty, choice uncertainty, as well as dispositional attitude toward uncertainty. Finally, we demonstrated the generalisability of our results by analysing verbal protocol data from the TV game show Who Wants to Be a Millionaire?

Keywords:

• Screening
• inclusion
• exclusion
• uncertainty

Acknowledgements

This research is partially based on the first author’s doctoral dissertation under the guidance of the second and third authors. The authors wish to thank Ratti Ratneshwar and the seminar participants at University of Missouri–Columbia, Saint Louis University, and the University of Central Florida for their constructive comments and Krystal Bedard and Benjamin Marshal for their research support. Vishal Bindroo’s technical assistance is greatly appreciated.

Disclosure statement

No potential conflict of interest is reported by the authors.

Notes

1 We conducted a pretest with 66 participants. They were randomly assigned to either smaller or larger choice set conditions and were then asked to rate their perceptions of how complex the task was on two 7-point scales (anchored by not complex/complex and not complicated/complicated, respectively; r = .84). The results showed that choosing an apartment from 10 available apartments was perceived as a significantly less complex task (M = 3.91) than choosing from 20 apartments (M = 4.62) (F(1, 64) = 4.31, p = .04).

2 To demonstrate the robustness of these findings, we conducted two replication studies using different study designs. Details of these follow-up studies are available on request.

3 We conducted a pretest with 44 participants randomly assigned to either the condition of high accuracy consideration or the control condition and asked to rate the importance of making an accurate decision on a 7-point scale (1 = not at all important, 7 = very important). The results showed that participants in the condition of high accuracy consideration rated that it was significantly more important for them to make an accurate decision (M = 6.30) than those in the control condition (M = 5.0) (F(1, 42) = 8.83, p = .01).

4 We also analysed the data using the OLS method and observed similar results.

5 It is plausible that when contestants were relatively uncertain, they were more likely to think aloud in general as they attempted to make the decision. To rule out this alternative explanation, we included the thoughts categorised as “others” in the model as a control variable. The results were substantively similar, with contestants higher on uncertainty preferring the exclusion strategy. The extent of other thoughts did not influence strategy preferences. Furthermore, we conducted another analysis by dividing the uncertainty measure as well as the strategy measure by the sum of thoughts and the results were also substantively similar.