Observing many researchers using the same data and hypothesis reveals a hidden universe of uncertainty.

This study explores how researchers' analytical choices affect the reliability of scientific findings. Most discussions of reliability problems in science focus on systematic biases. We broaden the lens to emphasize the idiosyncrasy of conscious and unconscious decisions that researchers make during data analysis. We coordinated 161 researchers in 73 research teams and observed their research decisions as they used the same data to independently test the same prominent social science hypothesis: that greater immigration reduces support for social policies among the public. In this typical case of social science research, research teams reported both widely diverging numerical findings and substantive conclusions despite identical start conditions. Researchers' expertise, prior beliefs, and expectations barely predict the wide variation in research outcomes. More than 95% of the total variance in numerical results remains unexplained even after qualitative coding of all identifiable decisions in each team's workflow. This reveals a universe of uncertainty that remains hidden when considering a single study in isolation. The idiosyncratic nature of how researchers' results and conclusions varied is a previously underappreciated explanation for why many scientific hypotheses remain contested. These results call for greater epistemic humility and clarity in reporting scientific findings.

No generalizable effect of income inequality on public support for governmental redistribution among rich democracies 1987-2010.

We revisit a longstanding hypothesis that the public become more supportive of redistributive policy as income inequality rises. Previous tests of this hypothesis using various forms of general least squares regressions are inconclusive. We suggest improvements and alternatives to these tests. Using the World Inequality Data and International Social Survey Program we analyze 91 surveys in 18 countries. We incorporate three alternative measures of income inequality, including a measure of liberalization as a known cause of increases in income inequality. We also employ two alternative test formats that arguably reflect the data generating model better than a least squares regression. The first is vector-autoregression aiming to account for path dependency of public opinion and income inequality, and the endogeneity between them. Next is qualitative comparative analysis to capture sets of conditions that collectively should have led to inequality having an impact on public opinion. Finally, we run our regression models separately for low and high socio-economic strata. In all tests we find no measurable impact of income inequality on support for redistribution. From a macro-perspective we argue that this suggests ruling out a general effect that exists across space and time, and focusing instead on theory to explain why there should not be a general effect. Some arguments suggest the public are normatively opposed to what sounds like 'handouts'. We therefore discuss model specification via theory, but also Type II errors, statistical power and the limitations of our conclusions.

The replication crisis has led to positive structural, procedural, and community changes.

The emergence of large-scale replication projects yielding successful rates substantially lower than expected caused the behavioural, cognitive, and social sciences to experience a so-called 'replication crisis'. In this Perspective, we reframe this 'crisis' through the lens of a credibility revolution, focusing on positive structural, procedural and community-driven changes. Second, we outline a path to expand ongoing advances and improvements. The credibility revolution has been an impetus to several substantive changes which will have a positive, long-term impact on our research environment.