Abstract
This study aims to investigate the association between trust in institutions and the reasons for sharing unverified information on social media. Specifically, this study explores the role of perceived self-efficacy in detecting misinformation and the motivation to authenticate information in online contexts. We draw on a sample of 2600 respondents, mainly Generation Z and Millennials (ages between 15 and 30). The findings show a blinding side of trust, revealing a positive association between trust in institutions on social media and reasons for sharing unverified information. Trust in institutions is positively associated with perceived self-efficacy in detecting misinformation. We suggest that the positive correlation between trust in institutions and perceived self-efficacy in detecting misinformation implies an overconfidence effect – i.e., individuals may overestimate their ability to assess information based on their belief that a source (institution) is trustworthy. This arguably represents a tendency to divert attention away from the accuracy of the information and explains the positive indirect association between trust and the likelihood of sharing unverified content. Moreover, trust is negatively associated with individuals' motivation to authenticate information, suggesting that individuals may rely on information utility rather than engage in critical thinking and verification. This study contributes to understanding the spread of misinformation on social media by highlighting the role of trust in institutions and its association with individuals' reasons for sharing unverified information. It also emphasizes the importance of perceived self-efficacy in detecting misinformation and the motivation to authenticate information as mediating mechanisms.
| Original language | English |
|---|---|
| Article number | 107992 |
| Pages (from-to) | 1-9 |
| Number of pages | 9 |
| Journal | Computers in Human Behavior |
| Volume | 150 |
| Early online date | 17 Oct 2023 |
| DOIs | |
| Publication status | Published - Jan 2024 |
Bibliographical note
Funding Information:Structural equation modeling using IBM AMOS 25 supported the analysis of our hypothesized model. First, we conducted a confirmatory factor analysis to examine the reliability and validity of the measurement instrument. Subsequently, we examined the structural regression model to examine our hypotheses. The models were estimated using a maximum likelihood estimator. Model parameters were obtained through bootstrapping, extracting 5000 bootstrap samples. Model fit was evaluated through various model fit indices. First, we report the chi-square/df ratio; values below 5 are typically considered to indicate good model fit (Hu & Bentler, 1999). However, this measure is known to be sensitive to sample size. In addition, we examined two incremental and two absolute fit indices: Tucker-Lewis Index (TLI), Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). For the TLI and CFI values greater than 0.90 indicate good model fit. For the RMSEA and SRMR, values less than 0.08 indicate a good model fit (Hair et al., 2010; Hu & Bentler, 1999).
Publisher Copyright:
© 2023 The Author(s)
Funding
Structural equation modeling using IBM AMOS 25 supported the analysis of our hypothesized model. First, we conducted a confirmatory factor analysis to examine the reliability and validity of the measurement instrument. Subsequently, we examined the structural regression model to examine our hypotheses. The models were estimated using a maximum likelihood estimator. Model parameters were obtained through bootstrapping, extracting 5000 bootstrap samples. Model fit was evaluated through various model fit indices. First, we report the chi-square/df ratio; values below 5 are typically considered to indicate good model fit (Hu & Bentler, 1999). However, this measure is known to be sensitive to sample size. In addition, we examined two incremental and two absolute fit indices: Tucker-Lewis Index (TLI), Comparative Fit Index (CFI), Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMR). For the TLI and CFI values greater than 0.90 indicate good model fit. For the RMSEA and SRMR, values less than 0.08 indicate a good model fit (Hair et al., 2010; Hu & Bentler, 1999).
| Funders | Funder number |
|---|---|
| Comparative Fit Index | |
| RMSEA |
Keywords
- Generation Z
- Millennials
- Misinformation
- Motivation for authentication
- Reasons for sharing unverified information
- Self-efficacy
- Social media
- Trust in institutions