Comparison of the Emotional Effects of Facebook versus Instagram Deactivation

For Facebook, the study reported an estimated improvement of 0.060 standard deviations in the emotional state index among users in the deactivation group compared to the control group. This represents a statistically significant enhancement (p < 0.001), indicating that users who deactivated Facebook experienced measurable improvements in overall well-being based on the composite measure of how often they felt happy, less depressed, and less anxious. In contrast, Instagram deactivation led to a 0.041 standard deviation improvement in the emotional state index relative to the control group. This effect was statistically significant at the p = 0.016 level when considered in isolation; however, after adjusting for multiple hypothesis tests (with the full set of political outcomes also being considered), the result did not meet the pre-registered significance threshold (q-value = 0.139). Thus, while both platforms showed improvements, the effect was stronger and more robust for Facebook.^[1]

The study also examined the three individual components of emotional state. In the Facebook experiment, deactivation improved self-reported happiness by 0.064 standard deviations, reduced the negative feeling of depression (re-signed as depressed × (−1)) by 0.039 standard deviations, and reduced anxiety (re-signed as anxious × (−1)) by 0.028 standard deviations. The Instagram experiment, while similarly oriented, produced smaller point estimates with improvements of 0.044, 0.026, and 0.024 standard deviations respectively for happiness, depression, and anxiety. These findings indicate that the magnitude of the beneficial effects is consistently higher for Facebook deactivation across all measured emotional states.^[1]

The analysis further highlighted interesting subgroup differences between the two platforms. For Facebook, the results suggested that the beneficial effects on emotional state were primarily driven by users over the age of 35. In contrast, the effects observed for Instagram deactivation were most pronounced among young women under the age of 25, with an estimated improvement of 0.111 standard deviations for this subgroup. Other subgroup comparisons (including those based on baseline usage, baseline emotional state, and baseline political engagement) did not reveal significant differences, indicating that the overall improvements were uniformly distributed outside of the noted age and gender interactions. This suggests that while both platforms exert a positive influence when deactivated, the characteristics of the affected population differ between Facebook and Instagram.^[1]

The study provided several benchmarks that help contextualize the size of the improvements. The effects of Facebook deactivation, for instance, were equivalent to moving a median user from the 50th percentile to the 52.4th percentile in the distribution of the emotional state index, while Instagram’s effect moved users to the 51.6th percentile. Moreover, when translated back into the original survey response units, Facebook deactivation shifted responses by values equivalent to about 3.8 percent of people indicating higher frequencies of feeling happy. In comparative terms, the emotional state improvements from deactivation were roughly 15 to 22 percent as large as those reported by an average psychological intervention, such as those cited in the psychological interventions meta-analysis. This comparison suggests that the magnitude of the improvements, while positive, is modest relative to standard psychological treatments.^[1]

In summary, the study reveals that deactivating social media accounts in the weeks leading up to an election can have a positive impact on users’ emotional state. Facebook deactivation produced a larger and more statistically robust improvement (0.060 standard deviations) compared to Instagram deactivation (0.041 standard deviations), which became non-significant after multiple testing adjustments. Furthermore, subgroup analyses indicate that the benefits for Facebook are more pronounced among older users, whereas Instagram deactivation appears to benefit young women more significantly. Overall, these findings contribute to the ongoing debate concerning the psychological impacts of social media and offer evidence that targeted periods of deactivation may lead to modest improvements in emotional well-being.^[1]