Research on the Influence Mechanisms of the Affective and Cognitive Self-Esteem.

Most prior studies examined the direct relation between social media usage and loneliness. This study tries to reveal the covert mechanisms involved in how different types of SMU affect older adults' loneliness, which has rarely been an object of attention in the prior literature. A partial least squares structural equation modeling (PLS-SEM) method was used to analyze the data collected from 466 older adults in a field study. This research divided self-esteem into two dimensions: affective self-esteem (AE) and cognitive self-esteem (CE). The study found that changes in CE only stemmed from functional SMU (FSMU), rather than active SMU (ASMU) and passive SMU (PSMU). ASMU and PSMU had a significant effect on FSMU. CE had a significant effect on loneliness. Objective social isolation (OSI) had a positive relationship with loneliness. Moreover, PSMU, FSMU, and CE had a significant effect on ASMU, CE, and AE, respectively. For older adults, the feeling of connecting with others was more valuable than acquiring specific outcomes. The mediation test results showed that FSMU could play a completely mediating role in the relationship between ASMU and PSMU, as well as that between ASMU and CE. CE significantly mediated the relationship between FSMU and loneliness. Finally, the total effect sizes of ASMU and PSMU on FSMU were significant, and those of CE on older adults' loneliness and AE were significant, while the total effect of AE on older adults' loneliness was non-significant. AE moderated the relationship between PSMU and OSI, so PSMU was related to higher OSI only for users experiencing high AE. These findings offer a guide for the use of social media to conduct future loneliness interventions for older adults.