Date of Award

Spring 2015

Degree Type


Degree Name

Master of Arts (MA)



First Advisor

Sorin Matei

Committee Chair

Sorin Matei

Committee Member 1

Seungyoon Lee

Committee Member 2

Glenn Sparks


The goal of this thesis is to develop and begin to test a modified knowledge gap model that builds upon the assumptions of the knowledge gap hypothesis, incorporates findings of recent digital divide research and accounts for the unique affordances of new media. This thesis draws upon information behavior and social capital theory and builds off past findings on knowledge and usage gaps in order to develop and explain a knowledge gap model for a new media setting. The traditional knowledge gap hypothesis explains that people of a high socioeconomic status (SES) gain more knowledge from exposure to media messages than people of a low SES, resulting in increasing "gaps" in knowledge after media exposure. This paper argues that differences in the types of ways people use new media (usage) and differences in the connections available to them through new media (social capital) mediate SES's effect on knowledge formation, and that the features of new media like social networking sites (SNSs) can maintain or even facilitate these differences. Thus, usage and social capital variables must be added to the traditional knowledge gap model to make it useful for a new media setting. Particularly, if SES also predicts a usage gap in the use of SNSs to accrue information and social capital, it might predict an accelerated knowledge gap phenomenon in a SNS setting. This thesis tests the accelerated knowledge gap model in a Facebook setting, using multiple regression and mediation analyses to test its hypotheses. The results support a potential causal connection between SES, usage, social capital and knowledge gain. Interestingly, while SES marginally predicted knowledge gain in one model, its effect on knowledge gain was suppressed, because people of a higher SES were less likely to use Facebook for informational purposes. Also, the data revealed a surprising finding that one of the most significant usage gaps may be explained by sex instead of SES, as women are more likely to use Facebook for interaction purposes. This could also inform potential social capital gaps, since interaction usage motivations were significant predictors of both bridging and bonding social capital.

Included in

Communication Commons