Measuring large-scale emotion aggregates through social media text

class: center, middle, inverse, title-slide

# Measuring large-scale emotion aggregates through social media text
### David Garcia Graz University of Technology and CSH Vienna
### 29.07.2021, IC2S2 Slides vailable at: dgarcia.eu/IC2S2-2021

---

layout: true
<div class="my-footer">David Garcia - Computational Social Science Lab - TU Graz + CSH Vienna</div>

---

## The Computational Social Science Lab Austria
<center>
<img src="figures/Lab-July2021.png" width="925" />

---

## The Hype Cycle of Computational Social Science
<center>
<img src="figures/Hype1.svg" width="900" />

---
## The Hype Cycle of Computational Social Science
<center>
<img src="figures/Hype2.svg" width="900" />
---
## The Hype Cycle of Computational Social Science
<center>
<img src="figures/Hype3.svg" width="900" />
---

---

layout: true
<div class="my-footer">
<a href=https://xxx> Social media emotion macroscopes reflect emotional experiences in society at large. David Garcia, Max Pellert, Jana Lasser, Hannah Metzler. https://arxiv.org/abs/2107.13236 (2021)</a></div>

---

# Social Media Macroscopes of Emotions

.pull-left[
<img src="figures/Macy.jpg" width="1100" />

<a href="https://science.sciencemag.org/content/333/6051/1878/"> Diurnal and seasonal mood vary with work, sleep, and daylength across diverse cultures. Golder & Macy, Science (2011) </a>

]

.pull-right[
<img src="figures/hedonometer.png" width="1100" />

<a href="https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0026752"> Temporal patterns of happiness and information in a global social network: Hedonometrics and Twitter. Dodds et al. PLoS One (2011) </a>

]

---

# Social Media Macroscopes of Emotions
.pull-left[
<img src="figures/Paris.png" width="500" />
 <a href="https://journals.sagepub.com/doi/full/10.1177/0956797619831964"> Collective Emotions and Social Resilience in the Digital Traces After a Terrorist Attack. Garcia & Rime, Psychological Science (2019) </a>

]

.pull-right[
<img src="figures/COVID.png" width="1100" />

<a href="https://psyarxiv.com/qejxv"> Collective Emotions During the COVID-19 Outbreak. Metzler et al. Psyarxiv (2021) </a>

]

---

## Limits of Social Media Data to Study Emotion

[Estimating geographic subjective well-being from Twitter: A comparison of dictionary and data-driven language methods. Jaidka et al. PNAS (2020)](https://www.pnas.org/content/117/19/10165.short)

---

# Validating a UK emotion macroscope
1. Two years of weekly representative UK emotion survey by YouGov
2. UK Twitter data for the same period: 1.5 Billion tweets (without RT)
3. Text analysis: dictionary-based (LIWC) and supervised (RoBERTa)
4. Gender-rescaled time series of emotional expression
5. Pre-registered hypotheses with prediction period from Nov 2020

---

# Sadness in Twitter and YouGov

- Similar results with dictionary-based and supervised (r~0.65)
---

# Anxiety in Twitter and YouGov

- Better results with dictionary-based method and with gender rescaling
- Results robust to autocorrelation and heteroskedasticity

---

# Joy in Twitter and YouGov

- Substantially better results with supervised method than dictionary-based
---

# Exploring 12 emotional states
.pull-left[

- Time series of number sentences like "I am [emotion]" on Twitter

- Weak correlations happen for infrequent emotions in text

- Comparison: US weekly pre-election polls correlate with 0.66

- Arxiv preprint at https://arxiv.org/abs/2107.13236

]

.pull-right[
<img src="figures/Figure2.svg" width="700" />
]

---

layout: true
<div class="my-footer">
Validating daily social media macroscopes of emotions. Max Pellert, Hannah Metzler, Michael Matzenberger, David Garcia. Working Paper (2021)</div>

---

## Study 2: Validating an Austrian macroscope

.pull-left[

- 20-day emotion survey in derstandard.at (N=268,128)
- Daily frequency, 3-day windows

- Text from Der Standard forum (N=452,013)

- Austrian tweets (N=515,187) filtered as UK macroscope

- Compared dictionary-based (LIWC) and supervised model (GS)

]
.pull-right[
<img src="figures/DS1.svg" width="800" />
]

---

## Twitter sentiment and Der Standard survey

---

# Correlations with new COVID-19 cases

.pull-left[
<img src="figures/DS31.svg" width="600" />
]
.pull-right[
<img src="figures/DS32.svg" width="600" />
]

- Do correlations attenuate due to additional social media measurement error?
- Survey emotion correlation with new cases as strong as Twitter sentiment
- Errors sources might be different: Need for conceptual validations

---

# Social Sensing Emotions

3rd person and tweet emotion: +74.85% in anx. +62.12% in sad +34.97% in pos.

---

---

---

---

# Thanks for listening!
.pull-left[
<img src="figures/Logo_CSS.png" width="750" style="display: block; margin: auto;" />

New English-speaking **masters programme** at Uni Graz and TU Graz
]
.pull-right[
<center>
🚨🚨**We are hiring!**🚨🚨
</center>
- [PhD at CSH Vienna on emotional misinformation](https://www.csh.ac.at/wp-content/uploads/2021/07/Emomis_PhD_CSH.pdf)

- [Postdoc at University of Bristol on text as data about honesty](https://www.bristol.ac.uk/jobs/find/details/?nPostingId=100334&nPostingTargetId=240694&id=Q50FK026203F3VBQBV7V77V83&LG=UK&mask=newuobext)

- Two upcoming postdoc positions at the CSS Lab at TU Graz

]

**More at: [www.dgarcia.eu](https://dgarcia.eu) and [@dgarcia_eu](https://twitter.com/dgarcia_eu)**

Slides:  [www.dgarcia.eu/IC2S2-2021](http://www.dgarcia.eu/IC2S2-2021)