Debiasing Using Decomposition: The Case of Memory‐Based Credit Card Expense Estimates

Abstract

This article develops a conceptual framework to examine how consumers incorporate memory‐based and context‐based cues in estimating past and future credit card expenses. Focusing on memory‐based information, in this framework we suggest that past credit card expenses may be recalled as holistic totals or episodic individual expenses depending on the accessibility of each form of expense. We tested the conceptual framework with 3 studies. Study 1 showed that the recency of past expenses and frequency of credit card usage moderated the extent to which past expenses were used in estimating future expenses. Study 1 also showed that self‐awareness of estimation bias affected consumers’ estimates of future expenses. Study 2 showed that a decomposition strategy (unbundling a total into subcategories), which serves to make episodic individual expenses more accessible, was effective in reducing sampling‐based bias. Study 3 provided a validity check and showed that decomposition cues reduced estimation errors and affected estimates of future expense. In this article, we demonstrate the manner in which memory‐based information is used to make estimates, and we also demonstrate the effectiveness of decomposition as a debiasing technique for past and future estimates.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 52 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.