Construction and validation of an overconfidence scale in investment decisions

Abstract

Abstract Existing studies that directly measure the three types of overconfidence—overprecision, overplacement, and overestimation—are largely exploratory, highlighting the need for further confirmatory research to establish robust overconfidence measures. Moreover, there is no comprehensive instrument for analyzing these types of overconfidence in the context of investment decisions. To address these gaps, the present study aims to develop and validate a scale for assessing overconfidence in investment decisions, considering its three forms. The research involved 380 Brazilian participants and employed confirmatory factor analysis and standard multivariate validation methods. The scale, based on a theoretical framework of the three types of overconfidence, consisted of 30 questions (10 for each type) designed to measure the degree of overconfidence in investment decision-making. The scale demonstrated acceptable factor loadings and internal and external validity, indicating its suitability for evaluating overconfidence in investment decisions. The standardized factor loadings showed p values < 0.01 for all factors. The average variance extracted was above 0.5, and the composite reliability exceeded 0.7 for all three constructs analyzed. These results support the convergent validity of the proposed measurement model. In this way, the developed scale fills a gap in the literature by providing a valid tool to assess overconfidence in investment decision-making, which can be practically applied by financial market participants, such as investors, fund managers, and financial professionals. This instrument enables more informed and personalized investment decisions, mitigates the risks of bias in decision-making, and contributes to a more efficient and resilient financial market.

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 — 7 citations exist upstream.

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