Work

Increasing sample size compensates for data problems in segmentation studies

Year: 2015

Type: article

Abstract: Survey data frequently serve as the basis for market segmentation studies. Survey data, however, are prone to a range of biases. Little is known about the effects of such biases on the quality of data-driven market segmentation solutions. This study uses artificial data sets of known structure to study the effects of data problems on segment recovery. Some of the data problems under study are partially under the control of market research companies, some are outside their control. Results indicate that (1) insufficient sample sizes lead to suboptimal segmentation solutions; (2) biases in survey data have a strong negative effect on segment recovery; (3) increasing the sample size can compensate for some biases; (4) the effect of sample size increase on segment recovery demonstrates decreasing marginal returns; and—for highly detrimental biases—(5) improvement in segment recovery at high sample size levels occurs only if additional data is free of bias.

Source: Journal of Business Research

DOI: 10.1016/j.jbusres.2015.09.004

Authors: Sara Dolnicar, Bettina Grün, Friedrich Leisch

Institutions: The University of Queensland, Johannes Kepler University of Linz, BOKU University

Language: English

Topic: Customer churn and segmentation

Subfield: Marketing

Field: Business, Management and Accounting

Domain: Social Sciences

SDG: Reduced inequalities

Open Access status: hybrid

Estimated APC paid: $1,739