Influencing recommendation algorithms to reduce the spread of unreliable news by encouraging humans to fact-check articles, in a field experiment.

Society often relies on social algorithms that adapt to human behavior. Yet scientists struggle to generalize the combined behavior of mutually-adapting humans and algorithms. This scientific challenge is a governance problem when algorithms amplify human responses to falsehoods. Could attempts to influence humans have second-order effects on algorithms? Using a large-scale field experiment, I test if influencing readers to fact-check unreliable sources causes news aggregation algorithms to promote or lessen the visibility of those sources. Interventions encouraged readers to fact-check articles or fact-check and provide votes to the algorithm. Across 1104 discussions, these encouragements increased human fact-checking and reduced vote scores on average. The fact-checking condition also caused the algorithm to reduce the promotion of articles over time by as much as -25 rank positions on average, enough to remove an article from the front page. Overall, this study offers a path for the science of human-algorithm behavior by experimentally demonstrating how influencing collective human behavior can also influence algorithm behavior.

The Upworthy Research Archive, a time series of 32,487 experiments in U.S. media.

The pursuit of audience attention online has led organizations to conduct thousands of behavioral experiments each year in media, politics, activism, and digital technology. One pioneer of A/B tests was Upworthy.com, a U.S. media publisher that conducted a randomized trial for every article they published. Each experiment tested variations in a headline and image "package," recording how many randomly-assigned viewers selected each variation. While none of these tests were designed to answer scientific questions, scientists can advance knowledge by meta-analyzing and data-mining the tens of thousands of experiments Upworthy conducted. This archive records the stimuli and outcome for every A/B test fielded by Upworthy between January 24, 2013 and April 30, 2015. In total, the archive includes 32,487 experiments, 150,817 experiment arms, and 538,272,878 participant assignments. The open access dataset is organized to support exploratory and confirmatory research, as well as meta-scientific research on ways that scientists make use of the archive.

Preventing harassment and increasing group participation through social norms in 2,190 online science discussions.

Theories of human behavior suggest that people's decisions to join a group and their subsequent behavior are influenced by perceptions of what is socially normative. In online discussions, where unruly, harassing behavior is common, displaying community rules could reduce concerns about harassment that prevent people from joining while also influencing the behavior of those who do participate. An experiment tested these theories by randomizing announcements of community rules to large-scale online conversations in a science-discussion community with 13 million subscribers. Compared with discussions with no mention of community expectations, displaying the rules increased newcomer rule compliance by >8 percentage points and increased the participation rate of newcomers in discussions by 70% on average. Making community norms visible prevented unruly and harassing conversations by influencing how people behaved within the conversation and also by influencing who chose to join.

Caveat emptor, computational social science: Large-scale missing data in a widely-published Reddit corpus.

As researchers use computational methods to study complex social behaviors at scale, the validity of this computational social science depends on the integrity of the data. On July 2, 2015, Jason Baumgartner published a dataset advertised to include "every publicly available Reddit comment" which was quickly shared on Bittorrent and the Internet Archive. This data quickly became the basis of many academic papers on topics including machine learning, social behavior, politics, breaking news, and hate speech. We have discovered substantial gaps and limitations in this dataset which may contribute to bias in the findings of that research. In this paper, we document the dataset, substantial missing observations in the dataset, and the risks to research validity from those gaps. In summary, we identify strong risks to research that considers user histories or network analysis, moderate risks to research that compares counts of participation, and lesser risk to machine learning research that avoids making representative claims about behavior and participation on Reddit.

A pragmatist philosophy of psychological science and its implications for replication.

A pragmatist philosophy of psychological science offers to the direct replication debate concrete recommendations and novel benefits that are not discussed in Zwaan et al. This philosophy guides our work as field experimentalists interested in behavioral measurement. Furthermore, all psychologists can relate to its ultimate aim set out by William James: to study mental processes that provide explanations for why people behave as they do in the world.