Wilfrid Laurier University
Using Our Past to Predict Our Future: Applying Reference Class Forecasting to Debias Individual Project Completion Predictions
Abstract
dc:descriptionPeople often predict that they will finish projects sooner than they actually do, i.e., exhibit the planning fallacy (e.g., Buehler et al., 2010). This bias has important consequences for everyday life, including failure to meet deadlines, taking on too many projects, and increased stress. Several solutions have been proposed, including interventions which ask individuals to take an “outside view” (e.g., Kahneman & Lovallo, 1993), such as using information from past completion times to make predictions for a current project (e.g., Buehler et al., 1994). In this work, we take a novel approach to helping individuals use past project information: recalling past completion times in reference to predictions (i.e., “how much later did I finish compared to my original expectation?”) as opposed to deadlines (i.e., how close to the deadline did I finish?”) and reference class forecasting (RCF; Lovallo and Kahnemann, 2003). In Study 1 and 2 (N = 322), we asked participants to report their planning fallacy beliefs, i.e., how many days after or before their predictions they believed they finished past projects. Although people on average reported finishing projects slightly later than predicted, awareness of this bias did not lead to less optimistic predictions for a current project. In Study 3-6 (N = 1,425), we instructed participants to recall relevant past project completion times using RCF, a technique that has been successful in reducing the planning fallacy for large-scale infrastructure projects (e.g., Flyvbjerg, 2008; Flyvbjerg et al., 2009), but has not been tested in individual, personal projects. Although our results were not completely consistent, we found evidence that RCF led to less optimistically biased completion predictions in three of the four studies. Overall, our work suggests that RCF, especially with past completion times recalled in reference to predictions, is a promising strategy for helping people make more accurate completion predictions for their individual personal projects.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy (PhD)
- Discipline thesis:degree_discipline
- Psychology
- Grantor dc:publisher
- Wilfrid Laurier University
- Year dc:date.available
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Friesdorf, Rebecca
- Contributors dc:contributor
-
- Dr. Roger Buehler
Subjects
dc:subject × 6Rights
dc:rights- Statement dc:rights
-
- 2 Publicly accessible
- Language dc:language
- en
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://scholars.wlu.ca/etd/2245