Visual Representations in Organizational Instagram Photos and the Public’s Responses: Focusing on Nonprofit Organizations

Yunhwan Kim, Siyeon Jang

Abstract


This study aimed to explore what were visually represented in nonprofit organizations’ (NPOs’) Instagram photos and how the features of the photos were related to the public’s responses. The contents of the photos were examined using online artificial intelligence services. NPOs’ Instagram photos and accounts were clustered discretely, and the resulting clusters were compared in terms of the photo features at content and pixel levels. The public’s responses were correlated with and predicted from the photo features. The results showed that photos of people made up the largest share of NPOs’ Instagram photos. Three photo clusters and three account clusters were detected and found to be different in terms of their content- and pixel-level characteristics. A part of photo features was significantly associated with the public’s responses, and engagement was predicted from the photo features with an acceptable level of accuracy whereas comment sentiment was not.

 


Keywords


visual representation, Instagram, nonprofit organization, clustering, engagement, comment sentiment

Full Text:

PDF