Table 1.

Overview of preregistered research questions and hypotheses

Research question	Hypothesis
Q1. Is the classic scientific consensus message effective compared to a control message?	H1a–e (main effects: control versus classic consensus).
	Compared to the control condition, participants in the classic consensus condition:
	(a) perceive a higher scientific consensus that human-caused climate change is happening (controlling for pre-intervention perceptions of the reality consensus),
	(b) believe more in the reality of climate change,
	(c) believe more in the human causation of climate change,
	(d) worry more about climate change and
	(e) support public action on climate change more.
Q2. Is the updated scientific consensus message effective compared to a control message?	H2a–g (main effects: control versus updated consensus).
	Compared to the control condition, participants in the updated consensus condition:
	(a) perceive a higher scientific consensus that human-caused climate change is happening (controlling for pre-intervention perceptions of the reality consensus),
	(b) perceive higher scientific agreement that climate change is a crisis (controlling for pre-intervention perceptions of the crisis agreement),
	(c) believe more in the reality of climate change,
	(d) believe more in the human causation of climate change,
	(e) believe more that climate change constitutes a crisis,
	(f) worry more about climate change and
	(g) support public action on climate change more.
Q3. Is the updated scientific consensus message more effective than the classic consensus message?	H3a–c (main effects: classic consensus versus updated consensus).
	Compared to the classic consensus condition, participants in the updated consensus condition:
	(a) believe more that climate change constitutes a crisis,
	(b) worry more about climate change and
	(c) support public action on climate change more.
Q4. Does the effectiveness of the classic consensus message vary by subgroup?	H4a–c (interaction effects: control versus classic consensus).
	Controlling for pre-intervention perceptions of the reality consensus, the effect of the classic consensus versus control condition on reality consensus perceptions is moderated by:
	(a) message familiarity, such that the message is more effective for those who report lower familiarity with the classic consensus statement and
	(b) trust in climate scientists, such that the message is more effective for those who report greater trust in climate scientists.
	Controlling for pre-intervention perceptions of the reality consensus, the effect of the classic consensus versus control condition on reality consensus perceptions is not moderated by:
	(c) political ideology.
Q5. Does the effectiveness of the updated consensus message vary by subgroup?	We planned the following exploratory analyses:
	1. Is the effect of the updated versus control condition on reality consensus perceptions moderated by:
	2. (a) message familiarity (of the classic consensus message),
	3. (b) trust in climate scientists and
	4. (c) political ideology, controlling for pre-intervention perceptions of the reality consensus?
	5. Is the effect of the updated versus control condition on crisis agreement perceptions moderated by:
	6. (a) message familiarity (of the classic and the updated consensus message),
	7. (b) trust in climate scientists and
	8. (c) political ideology, controlling for pre-intervention perceptions of the crisis agreement?
Q6. Does the effectiveness of both interventions vary by country?	We planned to explore if the effectiveness of the interventions varies by country.