Conceptual articles may disrupt the field of marketing: Evidence from a GPT-assisted study

2023-08-28

AI Summary

Existing literature lacks systematic, quantitative assessments of the scholarly impact of conceptual research in marketing. Method: This study introduces a novel methodology integrating large language model (GPT)-driven automated literature classification with network-based citation analysis to establish the first dichotomous “conceptual–empirical” framework and compute Disruption Scores (DS) quantifying paradigmatic disruption. Contribution/Results: Conceptual papers exhibit significantly higher average citation counts and DS values than empirical papers, underscoring their pivotal role in theoretical foundation-building and paradigm innovation. This methodological advancement provides a reproducible, scalable framework for evaluating knowledge evolution within marketing and related disciplines.

Abstract

Marketing scholars have underscored the importance of conceptual articles in providing theoretical foundations and new perspectives to the field. This paper supports the argument by employing two network-based measures - the number of citations and the disruption score - and comparing them for conceptual and empirical research. With the aid of a large language model, we classify conceptual and empirical articles published in a substantial set of marketing journals. The findings reveal that conceptual research is not only more frequently cited but also has a greater disruptive impact on the field of marketing than empirical research. Our paper contributes to the understanding of how marketing articles advance knowledge through developmental approaches.

Problem

• Measures disruptiveness of conceptual marketing papers
• Compares citation and disruption scores for research types
• Assesses impact of conceptual vs empirical marketing articles

Innovation

• Used network-based citation and disruption measures
• Applied large language model for article classification
• Compared conceptual and empirical research impact