[Paper Review] It's all in the (Sub-)title? Expanding Signal Evaluation in Crowdfunding Research

Constantin von Selasinsky, Andrew Isaak|arXiv (Cornell University)|Oct 27, 2020

FinTech, Crowdfunding, Digital Finance4 citations

TL;DR

This study enhances crowdfunding success prediction by integrating video subtitles as textual signals alongside project titles and descriptions in technology projects on Kickstarter. Using a sample of 1,049 projects, it demonstrates that incorporating subtitle data significantly increases model explanatory power, advancing understanding of how entrepreneurs signal credibility and project quality to backers.

ABSTRACT

Research on crowdfunding success that incorporates CATA (computer-aided text analysis) is quickly advancing to the big leagues (e.g., Parhankangas and Renko, 2017; Anglin et al., 2018; Moss et al., 2018) and is often theoretically based on information asymmetry, social capital, signaling or a combination thereof. Yet, current papers that explore crowdfunding success criteria fail to take advantage of the full breadth of signals available and only very few such papers examine technology projects. In this paper, we compare and contrast the strength of the entrepreneur's textual success signals to project backers within this category. Based on a random sample of 1,049 technology projects collected from Kickstarter, we evaluate textual information not only from project titles and descriptions but also from video subtitles. We find that incorporating subtitle information increases the variance explained by the respective models and therefore their predictive capability for funding success. By expanding the information landscape, our work advances the field and paves the way for more fine-grained studies of success signals in crowdfunding and therefore for an improved understanding of investor decision-making in the crowd.

Motivation & Objective

• To investigate whether video subtitles provide additional predictive signals for crowdfunding success beyond titles and descriptions.
• To address the gap in crowdfunding research that underutilizes textual signals, especially in technology project categories.
• To improve understanding of investor decision-making by expanding the information landscape available for signal evaluation.
• To advance methodological approaches in CATA (computer-aided text analysis) by incorporating multimodal textual data from video subtitles.

Proposed method

• A random sample of 1,049 technology projects was collected from Kickstarter’s public database.
• Textual data were extracted from three sources: project titles, project descriptions, and video subtitles.
• Natural language processing techniques were applied to analyze textual content and extract meaningful signals related to project quality and credibility.
• Multivariate regression models were used to assess the predictive power of each textual component on funding success.
• Model performance was evaluated by measuring the proportion of variance in funding outcomes explained by each signal source.
• The study compared models with and without subtitle data to isolate their incremental contribution to predictive accuracy.

Experimental results

Research questions

• RQ1 Do video subtitles contain unique textual signals that predict crowdfunding success beyond titles and descriptions?
• RQ2 How does the inclusion of subtitle data affect the explanatory power of predictive models for project funding outcomes?
• RQ3 Are textual signals from subtitles more informative for technology projects than for other project categories?
• RQ4 To what extent do different textual components (title, description, subtitles) contribute to explaining variance in funding success?

Key findings

• Incorporating video subtitles significantly increases the variance explained by predictive models of crowdfunding success.
• The addition of subtitle data improves model performance beyond what is achievable using only titles and descriptions.
• Textual signals from subtitles carry distinct and valuable information that is not fully captured by other textual components.
• The study confirms that subtitles serve as a meaningful source of credibility and project quality signaling in technology crowdfunding campaigns.
• The results suggest that video subtitles are underutilized in current CATA-based crowdfunding research and represent a promising data source for future studies.
• The findings support the theoretical framework of signaling theory by demonstrating that richer textual signals lead to better prediction of backer behavior.