Temporal patterns of happiness and information in a global social network: Hedonometrics and Twitter
📄 Original studyPlain English Summary
Ever wondered which day of the week is happiest? These researchers built a giant mood ring for the internet — a "hedonometer" measuring how happy people are based on the words they tweet. They rated over 10,000 English words on a happiness scale, then unleashed it on 46 billion words from 4.6 billion tweets. The results are delightful: Saturday is the happiest day, Tuesday the gloomiest, and people are cheeriest around 5-6 am — early birds really do get the emotional worm. Christmas tops the happiness charts while grim events drag the mood down. The really interesting twist? This tool later caught the eye of psi researchers, who used it to test whether collective Twitter sentiment might shift before big events — a kind of mass "gut feeling" in social media data.
Research Notes
Not a psi paper per se — mainstream computational social science establishing the labMT hedonometer methodology for large-scale sentiment analysis. Included in the library because this tool was later adopted by Radin et al. (2023) to test for presentiment effects in collective Twitter sentiment. Relevant as methodological background for social media approaches to studying collective consciousness and anomalous temporal patterns in aggregate human behavior.
A tunable, real-time, text-based 'hedonometer' was constructed using 10,222 English words rated for happiness (1-9 scale) by 50 Amazon Mechanical Turk evaluators each, validated against the established ANEW word set (Spearman r_s = 0.944, p < 10^-10). Applied to approximately 46 billion words from 4.6 billion tweets by over 63 million users across 33 months, the instrument revealed robust temporal patterns: a weekly cycle with Saturday happiest (h_avg ~ 6.06) and Tuesday least happy (h_avg ~ 6.03), a daily cycle peaking at 5-6 am (h_avg ~ 6.12), and sensitivity to major events (Christmas consistently happiest; Osama Bin Laden's death lowest overall). Happiness and information content (Simpson lexical size) were found to be statistically independent (r_s = -0.038, p ~ 0.71).
Links
Related Papers
Companion
- Stock Returns and the Mind: An Unlikely Result that Could Change Our Understanding of Consciousness — Holmberg, Ulf (2020)
- Effects of Mass Consciousness: Changes in Random Data During Global Events — Nelson, R.D (2011)
- Searching for Global Consciousness: A Seventeen Year Exploration — Bancel, Peter A (2017)
More in Methodology
Paranormal belief, conspiracy endorsement, and positive wellbeing: a network analysis
Planning Falsifiable Confirmatory Research
Addressing Researcher Fraud: Retrospective, Real-Time, and Preventive Strategies — Including Legal Points and Data Management That Prevents Fraud
Quantum Aspects of the Brain-Mind Relationship: A Hypothesis with Supporting Evidence
Paranormal beliefs and cognitive function: A systematic review and assessment of study quality across four decades of research
📋 Cite this paper
Dodds, Peter Sheridan, Harris, Kameron Decker, Kloumann, Isabel M, Bliss, Catherine A, Danforth, Christopher M (2011). Temporal patterns of happiness and information in a global social network: Hedonometrics and Twitter. PLoS ONE. https://doi.org/10.1371/journal.pone.0026752
@article{dodds_2011_hedonometrics,
title = {Temporal patterns of happiness and information in a global social network: Hedonometrics and Twitter},
author = {Dodds, Peter Sheridan and Harris, Kameron Decker and Kloumann, Isabel M and Bliss, Catherine A and Danforth, Christopher M},
year = {2011},
journal = {PLoS ONE},
doi = {10.1371/journal.pone.0026752},
}