The Effect of Social Media Reduction on Depression and Anxiety: A Meta-Analysis of RCTs

Due to some initial concerns about data quality, the preliminary analyses have been taken down. A task list has been provided to communicate the next steps of this project.

Design of the studies

The studies included in this meta-analysis are RCTs which randomize participants into two (or more) arms. Typically in these studies, there are two arms: 1) a social media reduction intervention and 2) a control (i.e., using social media as usual). The length of the intervention will vary between studies (typically around 1-4 weeks). The main analysis of this meta-analysis extracts mid and post-intervention effects from intent-to-treat analyses. The timeline for a typical intervention is as follows:

• Baseline assessment of depression and/or anxiety and intervention begins
• mid-intervention assessment(s) of depression or anxiety
• Intervention finishes and a post-intervention assessment is administered

Outcomes of interest are measures of depression (e.g., BDI) and/or anxiety (e.g., GAD).

Task List

• Re-calculating and double-coding effect sizes via summary statistics from studies in Ferguson’s meta.
• Developing search strategy for full systematic review beyond Ferguson’s meta-analysis.
• Obtaining raw data sets and conducting intent-to-treat analyses where possible.
• Specifying the research questions that we aim to answer.
• Constructing and designing meta-analytic models.
• Writing manuscript.