Experiments

Campaign ad persuasion experiments

Hewitt, Broockman, Coppock et al. · APSR 2024 · Swayable

617 real campaign ads, each tested against a control group: measured persuasion effects, hand-coded features for every ad, and 607,450 de-identified responses.

persuasionrandomizedvideo ads20182020

Rows previewed609.8K

Files1 · 269.3 MB

LicenseCC0

Original source

What it can tell you

Which coded ad features went with larger measured persuasion

How much true effects vary from ad to ad

How persuasion differs by party and gender of the viewer

What it can't

The ads themselves, their transcripts, or campaign names

Paid engagement metrics like views or shares

regression_ates · 1.6K tagging_2020 · 494 tagging_2018 · 148 studies · 146 responses · 607.5K

study_id	term	estimate	std.error	statistic	p.value	conf.low	conf.high	df	outcome	outcome_name	content_id
X495cbb4fe018	content_idX968edf32f1b4	8.7698	8.9343	0.9816	0.3307	-9.1453	26.685	53.6246	favorability	Favorability	X968edf32f1b4
X495cbb4fe018	content_idX05df5365013f	17.7614	8.9006	1.9955	0.0513	-0.1057	35.6286	51.1925	favorability	Favorability	X05df5365013f
X495cbb4fe018	content_idXdbe1138cec23	6.9841	9.3359	0.7481	0.4581	-11.7867	25.755	48.0308	favorability	Favorability	Xdbe1138cec23
X495cbb4fe018	content_idX3b9cca90c36e	1.9841	9.0066	0.2203	0.8266	-16.1245	20.0927	48.0308	favorability	Favorability	X3b9cca90c36e
X3c2f80c20258	content_idX412d6c0a63fd	-3.8193	1.9485	-1.9601	0.0513	-7.6602	0.0216	212.3638	favorability	Favorability	X412d6c0a63fd
X0c2b064a71c5	content_idX22f692d2c703	5.9076	0.9584	6.1642	0	3.9797	7.8356	47.0871	favorability	Favorability	X22f692d2c703
X0c2b064a71c5	content_idX031f89e21d56	6.4343	0.8997	7.1519	0	4.6239	8.2447	46.4762	favorability	Favorability	X031f89e21d56
X5a248af8c0f1	content_idXa0ca2f6dd0e3	3.2646	1.7736	1.8406	0.0675	-0.2381	6.7672	160.3937	favorability	Favorability	Xa0ca2f6dd0e3

Schema · 13 columns

study_idstring

termstring

estimatefloat64

std.errorfloat64

statisticfloat64

p.valuefloat64

conf.lowfloat64

conf.highfloat64

dffloat64

outcomestring

outcome_namestring

content_idstring

n_obsInt32

Files

swayable/replication_archive.zip269.3 MB

Used in PAX2

Scoring · What controlled tests found Script research priors

Cite

Hewitt, L., Broockman, D., Coppock, A., Tappin, B., Slezak, J., Coffman, V., Lubin, N., & Hamidian, M. (2024). How Experiments Help Campaigns Persuade Voters. American Political Science Review.

Library

Files