The Generalizability of Survey Experiments*

By Kevin J. Mullinix, Thomas J. Leeper, James N. Druckman, J. Freese. Journal of Experimental Political Science, 2015

1,042 citations recorded by the source corpus.

Subfields: Safety Research, Sociology and Political Science, Statistical and Nonlinear Physics

View Paper: https://www.semanticscholar.org/paper/341194e0da4bf086f9efbf5dbc84268916ec650d

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety Research

In Sociology and Political Science

In Statistical and Nonlinear Physics

Citations in the corpus

Generalizing from Survey Experiments Conducted on Mechanical Turk: A Replication Approach (2018, methodology): “None of the seven I replicated were in the set of TESS studies replicated Mullinix et al. (2015).”
Advantages of Matched Over Unmatched Opt-in Samples for Studying Criminal Justice Attitudes: A Research Note (2020, methodology): “Recent research suggests that observational findings from such samples often fail to generalize (Thompson & Pickett, 2019), even though experimental results tend to be externally valid (Coppock, 2019; Mullinix et al., 2015).”
Believing and sharing misinformation, fact-checks, and accurate information on social media: The role of anxiety during COVID-19 (2021, background): “Although MTurk samples differ from nationally representative samples, recent work has found that they do not lead to false negatives and positives, nor to inaccurate effect sizes, and that they are more representative than in-person convenience samples (Berinsky et al., 2012; Mullinix et al., 2016).”
The Long-lasting Effects of Newspaper Op-Eds on Public Opinion (2018, result): “While we are sensitive to such concerns, we also note that recent replication efforts (Coppock, 2018; Mullinix et al., 2015) have shown that experimental estimates obtained on Mechanical Turk correspond very closely with those obtained on nationally representative samples.”
No Harm in Checking: Using Factual Manipulation Checks to Assess Attentiveness in Experiments (2018, methodology): “In our first study, we replicated a framing experiment featured in Mullinix et al. (2015). The original experiment featured two conditions regarding a student loan forgive-”
The Implied Truth Effect: Attaching Warnings to a Subset of Fake News Headlines Increases Perceived Accuracy of Headlines Without Warnings (2020, neutral): “Mechanical Turk (Horton et al. 2011), although not nationally representative, has been shown to be a reliable resource for research on political ideology (Krupnikov and Levine 2014, Mullinix et al. 2015, Coppock 2018).”
Does Counter-Attitudinal Information Cause Backlash? Results from Three Large Survey Experiments (2018, neutral): “While we share these concerns, we also note that the evidence to date indicates that estimates obtained from MTurk samples match national samples well (Coppock 2017; Mullinix et al. 2015).”
The shape of and solutions to the MTurk quality crisis (2020, background): “In fact, MTurk samples generally provide higher quality data than student samples, community samples, and even some high-quality national samples (Hauser and Schwarz, 2015; Mullinix et al ., 2015; Thomas and Clifford, 2017; Anson, 2018).”
Improving the External Validity of Conjoint Analysis: The Essential Role of Profile Distribution (2021, neutral): “We note that our focus is on the external validity of conjoint profiles , and is distinct from another important issue of representativeness of respondents in survey experiments (see, e.g., Mutz, 2011; Mullinix et al. , 2015; Coppock, Leeper, and Mullinix, 2018; Miratrix et al. , 2018).”
Recruiting large online samples in the United States and India: Facebook, Mechanical Turk, and Qualtrics (2018, result): “In two out of three experiments in the U.S., we were able to replicate a pattern of heterogeneous treatment effects by party ID using each of our samples (see also Mullinix et al., 2015, for similar results using MTurk).”
Conducting Clinical Research Using Crowdsourced Convenience Samples. (2016, background): “These have been largely corroborated by high-powered batteries of experiments administered to both MTurk and non-MTurk samples (Klein et al. 2014, Mullinix et al. 2014), with the caveat that phenomena that are moderated by demographic characteristics upon which MTurk differs from representative…”