The Use of Indicators for Unobservable Product Qualities: Inferences Based on Consumer Sorting

Matthew G. Nagler, Fredi Kronenberg, Edward J. Kennelly, Bei Jiang, Chunhui MaInternational Journal of Marketing Studies 2012 · DOI

Abstract

We propose a method for measuring the conjectural errors that inexpert consumers make relative to experts in using observable product characteristics as surrogate indicators of a valued unobservable characteristic. Observations on the unobservable characteristic, available to the researcher but not consumers, are used to divide the data into high- and low-quality subsamples. Separate hedonic estimation on the subsamples enables measurement of the relative valuations and conjectures of experts and non-experts with respect to indicators under the assumption that consumers sort across quality grades based on their appraisal expertise. The method is demonstrated using a small sample of SKU-level data on the dietary supplement black cohosh. Our exploratory findings on this sample suggest that, relative to experts, inexpert consumers underestimate the value of most observable characteristics as indicators of black cohosh chemical authenticity; however they overweight therapeutic claims on the product label as a negative indicator of authenticity.

Introduction

In economics, "credence goods" present a unique challenge: their true quality remains hidden from consumers even after purchase and use. To navigate this uncertainty, consumers rely on "surrogate indicators"—observable traits like price, brand, or specific label claims—to estimate a product's value. However, information is rarely distributed evenly. While experts may correctly interpret these cues, novices often make "conjectural errors." This paper introduces a novel econometric method to mathematically extract these errors by leveraging the phenomenon of consumer sorting.

Methodology: Consumer Sorting and Hedonic Regression

The core of the proposed method is the Consumer Sorting Assumption. It posits that expert consumers, possessing superior appraisal skills, will disproportionately select high-quality (authentic) goods, while inexpert consumers will gravitate toward lower-quality (adulterated) alternatives.

The research process involves four primary steps:

  1. Establish Ground Truth: The researcher measures the hidden quality variable (e.g., chemical purity) that is invisible to the consumer.
  2. Split-Sample Hedonic Regression: The market data is divided into "authentic" and "non-authentic" subsamples. Separate regressions are run to determine how experts (authentic buyers) versus novices (non-authentic buyers) financially value different label claims.
  3. Relative Valuation Analysis: The model calculates $\mu_k$, representing the relative valuation and conjecture of experts versus non-experts.
Relative valuation and conjecture signs based on mu_k

Error Taxonomy

By comparing consumer valuations against the actual predictive power of label indicators, the authors categorize novice mistakes into a specific taxonomy:

  • Overshoot: The consumer correctly identifies the indicator's direction but overestimates its magnitude.
  • Undershoot: The consumer correctly identifies the direction but underestimates the magnitude.
  • Fallacy: The consumer misinterprets the indicator entirely (e.g., viewing a negative cue as positive).
Taxonomy of consumer errors based on indicator significance and sign consistency

Case Study: The Black Cohosh Market

The method was demonstrated using SKU-level data for black cohosh, a popular dietary supplement. The researchers chemically analyzed 38 distinct products (55 price observations) using high-performance liquid chromatography and mass spectrometry to determine chemical authenticity.

Key Findings

  • Authenticity and Predictive Power: 66% of products were chemically authentic. The 14 observable label characteristics collectively explained approximately 60% of the variance in actual authenticity, suggesting that a knowledgeable consumer could successfully navigate the market.
  • Misleading Reassurances: Label terms seemingly designed to build trust—such as "Certified" or "Guaranteed"—were actually significant negative indicators of chemical authenticity.
  • Novice Biases: For most indicators, inexpert consumers "undershot," suggesting a general unawareness of subtle quality cues. A notable exception was therapeutic claims (e.g., "ends hot flashes"). Inexpert consumers correctly identified these as negative indicators but "overshot," penalizing the value of products with these claims much more heavily than experts did.

Conclusion

This exploratory study demonstrates that market sorting can serve as a mathematical proxy for cognitive expertise, bypassing the need for complex consumer surveys. While limited by a small sample size and the use of retail list prices, the framework offers a powerful tool for understanding information asymmetry in credence good markets. Future work may apply this methodology to larger scanner datasets or other high-stakes markets, such as illegal drugs, to inform public health interventions.

Summary updated

Abstract

We propose a method for measuring the conjectural errors that inexpert consumers make relative to experts in using observable product characteristics as surrogate indicators of a valued unobservable characteristic. Observations on the unobservable characteristic, available to the researcher but not consumers, are used to divide the data into high- and low-quality subsamples. Separate hedonic estimation on the subsamples enables measurement of the relative valuations and conjectures of experts and non-experts with respect to indicators under the assumption that consumers sort across quality grades based on their appraisal expertise. The method is demonstrated using a small sample of SKU-level data on the dietary supplement black cohosh. Our exploratory findings on this sample suggest that, relative to experts, inexpert consumers underestimate the value of most observable characteristics as indicators of black cohosh chemical authenticity; however they overweight therapeutic claims on the product label as a negative indicator of authenticity.
BibTeX
@article{nagler2012use,
  author = {Matthew G. Nagler and Fredi Kronenberg and Edward J. Kennelly and Bei Jiang and Chunhui Ma},
  title = {{The Use of Indicators for Unobservable Product Qualities: Inferences Based on Consumer Sorting}},
  journal = {International Journal of Marketing Studies},
  year = {2012},
  doi = {10.5539/ijms.v4n6p19},
}
Figure