# The effect of wording on message propagation: Topic- and author-controlled natural experiments on Twitter

arXiv: https://arxiv.org/abs/1405.1438 · HTML: https://mlsanity.com/paper/1405.1438/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 May 2014.

## Built on by (most cited first, 21 of 21)

- [Can Large Language Models Transform Computational Social Science?](https://mlsanity.com/paper/2305.03514/) (2023)
- [Hypothesis Generation with Large Language Models](https://mlsanity.com/paper/2404.04326/) (2024)
- [A Causal Lens for Controllable Text Generation](https://mlsanity.com/paper/2201.09119/) (2022)
- [On the Diversity and Limits of Human Explanations](https://mlsanity.com/paper/2106.11988/) (2021)
- [Language Models Trained on Media Diets Can Predict Public Opinion](https://mlsanity.com/paper/2303.16779/) (2023)
- [On Behalf of the Stakeholders: Trends in NLP Model Interpretability in the Era of LLMs](https://mlsanity.com/paper/2407.19200/) (2024)
- [Measuring and Improving Persuasiveness of Large Language Models](https://mlsanity.com/paper/2410.02653/) (2024)
- [Causal Estimation for Text Data with (Apparent) Overlap Violations](https://mlsanity.com/paper/2210.00079/) (2022)
- [Decision-Focused Summarization](https://mlsanity.com/paper/2109.06896/) (2021)
- [A Survey of Deep Causal Models and Their Industrial Applications](https://mlsanity.com/paper/2209.08860/) (2022)
- [How Metaphors Impact Political Discourse: A Large-Scale Topic-Agnostic Study Using Neural Metaphor Detection](https://mlsanity.com/paper/2104.03928/) (2021)
- [Bayesian Topic Regression for Causal Inference](https://mlsanity.com/paper/2109.05317/) (2021)
- [Modeling Language Usage and Listener Engagement in Podcasts](https://mlsanity.com/paper/2106.06605/) (2021)
- [LLM-Driven Treatment Effect Estimation Under Inference Time Text Confounding](https://mlsanity.com/paper/2507.02843/) (2025)
- [IDIAPers @ Causal News Corpus 2022: Extracting Cause-Effect-Signal Triplets via Pre-trained Autoregressive Language Model](https://mlsanity.com/paper/2209.03891/) (2022)
- [The Branch Not Taken: Predicting Branching in Online Conversations](https://mlsanity.com/paper/2404.13613/) (2024)
- [Generator-Guided Crowd Reaction Assessment](https://mlsanity.com/paper/2403.09702/) (2024)
- [The Influence of Text Variation on User Engagement in Cross-Platform Content Sharing](https://mlsanity.com/paper/2505.03769/) (2025)
- [What Are Research Hypotheses?](https://mlsanity.com/paper/2509.00185/) (2025)
- [Predicting Post Virality with Temporal Cross-Attention over Trend Signals](https://mlsanity.com/paper/2605.02358/) (2026)
- [Paying for Too Many Tokens? Valid and Cost-Efficient Multimodal LLM Annotation with Simple Heuristics](https://mlsanity.com/paper/2610.00809/) (2026)

## Abstract (arXiv)

Consider a person trying to spread an important message on a social network. He/she can spend hours trying to craft the message. Does it actually matter? While there has been extensive prior work looking into predicting popularity of social-media content, the effect of wording per se has rarely been studied since it is often confounded with the popularity of the author and the topic. To control for these confounding factors, we take advantage of the surprising fact that there are many pairs of tweets containing the same url and written by the same user but employing different wording. Given such pairs, we ask: which version attracts more retweets? This turns out to be a more difficult task than predicting popular topics. Still, humans can answer this question better than chance (but far from perfectly), and the computational methods we develop can do better than both an average human and a strong competing method trained on non-controlled data.