In statistics, samples are drawn from a population in a data-generating process (DGP). Standard errors measure the uncertainty in estimates of population parameters. In science, evidence is generated to test hypotheses in an evidence-generating process (EGP). We claim that EGP variation across researchers adds uncertainty—nonstandard errors (NSEs). We study NSEs by letting 164 teams test the same hypotheses on the same data. NSEs turn out to be sizable, but smaller for more reproducible or higher rated research. Adding peer-review stages reduces NSEs. We further find that this type of uncertainty is underestimated by participants.

Promoting green product consumption is one important element in building a sustainable society. Yet green products are usually more costly. In times of high inflation, not only budget constraints but also the fear that prices will continue to rise might dampen green product consumption and, hence, limit the effectiveness of exerted efforts to promote sustainable behaviors. To test this suggestion, we conducted a Germany-wide survey with almost 1,200 respondents, followed by a framed field experiment (N=500) to confirm causality. In the survey, respondents’ stated “green” purchasing behavior is, as to be expected, positively correlated with concerns about climate change. It is also negatively correlated with concerns about future inflation and energy costs, but after controlling for observable characteristics such as income and educational level only the correlation with concerns about future prices remains significant. This result is driven by individuals with below-median environmental attitude. In the framed field experiment, we use the priming method to manipulate the saliency of inflation concerns. Whereas sizably relaxing the budget constraint (i.e., by 50 percent) has no impact on the share of organic products in participants’ baskets, the priming significantly decreases the share of organic products for individuals with below-median environmental attitude, similar to the survey data.