Attraction is all you need: The Impact of Eye-Catching Titles on Publication Success and Citation Rates in Economics
Authors: SHEN Zekai
Date: 2024
Abstract:
The impact of title characteristics on the publication and citation outcomes of academic papers has long been a topic of extensive research. Most studies on the statistical features of titles conclude that these characteristics influence a paper’s publication and citation through the attractiveness of the title. However, no one has directly drawn this conclusion through data analysis, largely due to the high cost of conducting manual surveys. This paper utilizes emerging deep learning models and large language models, including BERT and ChatGPT fine-tuned with human data, to assess whether paper titles are eye-catching. We conduct an empirical analysis using 347,197 papers published in 321 economics journals from 2000 to 2022, as classified by the ABS Journal Guide, to explore whether papers with eye-catching titles are more likely to be published in prestigious journals and receive more citations. Our findings indicate that papers with titles considered eye-catching by BERT, which has learned human preferences, are indeed more likely to be published in top-tier journals and receive more citations. The impact of titles deemed attractive by ChatGPT on academic publication is mixed, but there is a significant positive effect on the number of citations, possibly due to ChatGPT’s broader criteria compared to human researchers.This not only reveals the impact of eye-catching titles on publication and citation outcomes but also highlights the potential role of artificial intelligence tools and deep learning in economic research as economic research agents.
Keywords:
Bert, ChatGPT, Publication, Citations, Eye-catching Title
Current Status:
Work in Progress Under the Supervision of Professor Aljoscha JANSSEN