Trust and influence study on social media applications for observing potential new measurements and metrics.
Abstract
Social media has become a vital component of modern communication allowing people to share ideas, connect with others, and in many cases, influence human behavior around the world. Social media platforms rely on metrics such as likes, shares, followers and other social signals to measure influence and trustworthiness of users and content. However, these traditional metrics do not always provide an accurate representation of realworld influence and trustworthiness. In fact, they have tended to be designed for purposes that lean towards advertising-oriented metrics and until very recently, there has been very little transparency into the measurement algorithms. Additionally, there is very little measurement from within the social media ecosystem in the way of connecting social media online behavior with actual offline behavior. At the time of this writing, Twitter has recently open sourced portions of its recommendation algorithm, which will be taken into account as part of this study, but the vast majority of social media influence metrics are designed to promote increased user attention to the platforms, therefore serving the commercial needs of the social media companies. As such, there is a need to explore new metrics and measurements to evaluate trust and influence on social media platforms to study what types of elements affect trust and possibly drive behavior vs. what drives views and likes. Our basic research study posits that there is a difference between actually influencing behaviors and the more passive consumption of social media content. The former being a powerful artifact of trust, the later being extremely wide-spread, but perhaps less influential in shifting actual outcomes. It is well-known that a very small percentage of social media users across all platforms account for the vast majority of content creation that is deemed to be the highest reach, highest engagement content. This does not necessarily correlate to content that affects behavior other than passive engagement.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 06, 2024
- Source ID
- FA95502310518
Entities
People
- James Thompson
Organizations
- Air Force Office of Scientific Research
- United States Air Force