Too Positive to Be Tweeted? An Experimental Investigation of Emotional Expression on Twitter and Instagram

Figures & data

Table 1. Pre-registered hypotheses for experiments 1a and 1b.

Pre-registered hypotheses	Experiment 1a	Experiment 1b
H1. When the publication is negative, Twitter is chosen more often than Instagram.	✔	✔
H2. When the publication is positive, Instagram is chosen more often than Twitter.	✔	✔
H3. The position of the image has no impact on the choice of Twitter or Instagram.	✔	✔
H4. The association between the publication’s emotional valence (positive or negative) and the choice of a SNS (Twitter or Instagram) is mediated by (injunctive) norms of emotional expression on Twitter and Instagram.	✔	✔
H5. Having an Instagram and a Twitter account has no impact on the choice of Twitter or Instagram.		✔
H6. The association between the publication’s emotional valence (positive or negative) and the choice of a SNS (Twitter or Instagram) is not mediated by the frequencies of Instagram and Twitter use.		✔

The hypothesis 4 for experiment 1a focused on injunctive norms, while the hypothesis 4 for experiment 1b did not specify the type of norm.

Figure 1. Effect of message valence on Instagram or Twitter choice (experiment 1a).

Figure 2. Descriptive and injunctive norms for Twitter and Instagram (experiment 1a).

Figure 3. Effect of message valence on Instagram or Twitter choice (experiment 1b).

Table 2. Pre-registered hypotheses for experiments 2a and 2b.

Pre-registered hypotheses	Experiment 2a	Experiment 2b
H1. When the publication is negative, Twitter is chosen more often than Instagram.	✔	✔
H2. When the publication is positive, Instagram is chosen more often than Twitter.	✔
H3. The norms of emotional valence expression are more positive for Instagram than for Twitter.	✔
H4. When the publication is negative and the arousal is low, Instagram is chosen more often than Twitter.		✔
H5. When the publication is negative and the arousal is high, Twitter is chosen more often than Instagram.		✔

Figure 4. Effect of message valence and message arousal on SNS choice (experiment 2a).

Figure 5. Effect of message valence and message arousal on SNS choice (experiment 2b).

Figure 6. Diagram of the thematic analysis for Instagram (experiment 2b).

Note. The term “ref.” is the abbreviation of references and refers to the number of text fragments that were coded for each theme.

Figure 7. Diagram of the thematic analysis for Twitter (experiment 2b).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author, upon reasonable request https://doi.org/10.17605/OSF.IO/VMKXC and https://doi.org/10.17605/OSF.IO/4B8SD for experiment 1a, https://doi.org/10.17605/OSF.IO/Z2B4T for experiment 1b, https://doi.org/10.17605/OSF.IO/8P5MS for experiment 2a, and https://doi.org/10.17605/OSF.IO/BPK83 for experiment 2b.