Lexical sophistication and crowdfunding outcomes

Abstract

ABSTRACTIt is advised that entrepreneurs should keep the language of venture descriptions short and simple. However, knowledge about the conditions under which language in crowdfunding communications impacts investment behavior is limited. We propose that the use of sophisticated language in crowdfunding venture descriptions creates perceptions of venture distinctiveness that, in turn, increase the amount invested. We test this proposed mechanism across a controlled experiment and present evidence from 886 crowdfunding campaigns that increasing the linguistic sophistication of campaign descriptions can result in a 20% increase in the amount raised by a campaign and greater campaign success. Furthermore, we show that the impact of lexical sophistication on the amount raised is moderated by investor experience whereby experienced investors invest more in ventures with sophisticated language compared to inexperienced investors. Our work contributes to research on communication in crowdfunding by highlighting the significance of lexical sophistication in influencing funding behaviour.KEYWORDS: Crowdfundingdigital communicationentrepreneurshiplexical sophistication Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1. https://www.smallbizgenius.net/by-the-numbers/crowdfunding-stats/#gref2. We repeated the experiment with a sample of 200 undergraduate students and found a positive and significant impact of lexical sophistication on willingness to invest mediated by perceptions of distinctiveness.3. The sample included 317 successful campaigns.4. We use the LIWC software (e.g., Pennebaker et al. Citation2015) a widely used language processing software which categorizes words into predefined linguistic categories and allows for customized dictionaries to identify verbs, adverbs, adjectives, and word length. LIWC software has been used in prior studies on crowdfunding to link the use of words to psychological constructs (e.g., Parhankangas and Renko Citation2017).5. Principal-component factor analysis, which is a method of factor extraction that assumes the communalities of the observed variables are equal to 1. This means that all the variance in each variable is explained by the common factors, and there is no unique or error variance.