Table 2 Description of the 20 experiments included in the meta-analysis.

From: Accuracy prompts are a replicable and generalizable approach for reducing the spread of misinformation

Study	Date	Sample	Platform	Topic	Accuracy prompts used	Published?
A*	September-17	847	MTurk	Politics	-Evaluation	Unpublished
B*	October-17	1158	MTurk	Politics	-Evaluation	Pennycook et al. 2021 S3²⁴
C*	November-17	1248	MTurk	Politics	-Evaluation	Pennycook et al. 2021 S4²⁴
D	March-19	1007	MTurk	Politics	-Importance -Norms -Reason -Importance+Norms+Reason	Unpublished
E	March-19	1210	MTurk	Politics	-Evaluation (10x) -Evaluation (10x) +Importance+Norms+Reason -Evaluation (10x) with feedback -Evaluation (10x) with feedback +Importance+Norms+Reason	Unpublished
F	April-19	1184	Lucid	Politics	-Evaluation (10x) with feedback +Importance+Norms+Reason -Evaluation+Norms -Importance+Norms	Unpublished
G*	May-19	1286	Lucid	Politics	-Evaluation -Importance	Pennycook et al. 2021 S5²⁴
H	September-19	2296	MTurk	Politics	-Evaluation	Unpublished
I*	March-20	855	Lucid	COVID-19	-Evaluation	Pennycook et al. 2020 S2²⁶
J	April-20	621	MTurk	Politics & COVID-19	-Evaluation	Unpublished
K	April-20	444	Lucid	Politics	-Evaluation	Unpublished
L	April-20	1192	Lucid	COVID-19	-Evaluation -Evaluation (10x) with feedback	Epstein et al. 2021 W2²⁵
M	May-20	2081	Lucid	COVID-19	-Evaluation -Tips -Norms	Epstein et al. 2021 W3²⁵
N	May-20	2778	Lucid	COVID-19	-Tips -Norms -Tips+Norms	Epstein et al. 2021 W4²⁵
O	May-20	2616	Lucid	COVID-19	-Importance -Importance+Norms	Epstein et al. 2021 W5²⁵
P	September-20	820	Lucid	COVID-19	-Evaluation- -Tips	Unpublished
Q	September-20	2010	YouGov	COVID-19	-Evaluation -Evaluation with feedback -Importance+Norms -PSA Video	Unpublished
R	September-20	2015	YouGov	Politics	-Evaluation -Evaluation with feedback -Importance+Norms -PSA Video	Guay et al. 2022⁷²
S	November-20	162	Lucid	COVID-19	-Evaluation -Tips	Unpublished
T	December-20	415	Lucid	COVID-19	-Tips	Unpublished

Note that Epstein et al. (2021) consisted of four separate waves that were collected at different times, such that there were four separate instances of randomization between control and accuracy prompt conditions. We therefore treated each of those four waves as separate experiments (L-O) for the purposes of this meta-analysis. Studies that were pre-registered are indicated with *.

Back to article page

Table 2 Description of the 20 experiments included in the meta-analysis.

Search

Quick links