Standing out in a crowd of victim entrepreneurs: How entrepreneurs’ language-based cues of personality traits affect public support

ABSTRACT

Catastrophic events challenge the resilience of society and require entrepreneurs to act proactively. Government COVID-19 responses forced thousands of businesses to close, resulting in a staggering loss of revenue for small businesses. Many small business entrepreneurs turned to crowdfunding to make public funding appeals. Through the lens of the identifiable victim effect, we examine how donations to affected businesses are related to language-based cues of personality traits embedded in appeals. Using the IBM Watson Personality Insights algorithm, we assess charitable appeals for language-based cues that convey entrepreneurs’ Big Five personality traits. We test our model using 6,803 donation-based campaigns between March and May 2020. We further tested how crisis salience influenced prosocial behavior, discovering that donation effects were increased for appeals that highlighted the pandemic’s impact on the business. Our results suggest that language-based cues of personality traits have significant associations with public support when embedded in charitable appeals.

KEYWORDS:

• Personality traits
• COVID-19
• identifiable victim effect
• crisis response
• crowdfunding

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ This is not a requirement for our study given that our focus is on the written narrative of charitable appeals and aggregate public support for entrepreneurs, however, it does eliminate a potential source of noise. In a regional or lesser-known global crisis, donors who learn of the crisis for the first time through the charitable appeal may have bifurcated reactions, with some remaining indifferent and turning away from suffering and others helping with great zeal given the newness to them of knowledge of suffering. In contrast, a crisis which affected not only essentially all small businesses, but also any and every potential donor to at least some degree eliminates this source of noise.

² Practical interpretations of each statistically significant term are calculated using the following formula: (exp(coefficient) – 1) × 100.