Table 2. Experimental social science replicated with volunteer subjects on DLABSS.

Replicated Study	Dependent Variable	N	MTurk	N	Original	N
Tversky and Kahneman (1981)	Risk acceptance	539	✔	450	students	307
Rasinski (1989)	Support for gov’t spending	788	✔	329	GSS	1,470
Kam and Simas (2010)	Policy acceptance	752	✔	699	KN	752
Tomz (2007)	Audience costs	495			KN	1,127
Hainmueller and Hiscox (2010)	Immigration attitudes	736	✔*	833	KN	1,601
Gadarian and Albertson (2014)	Information seeking	668	✔*	736	KN	384

The first three studies were replicated by Berinsky, Huber and Lenz (2012) using paid MTurk subjects. “Original” column indicates the original sample, of which we are aware, other than MTurk or DLABSS, on which the study was carried out and is not an exhaustive list of replications. The first N column is the number of subjects on DLABSS, the second N is the number of subjects on MTurk, and the third N is the number of subjects on the original platform by the original researchers. A * next to the ✔ for MTurk indicates that we carried out the MTurk replication ourselves.