We leverage a large-scale incentivized survey eliciting behaviors from (almost) an entire university student population, a representative sample of the U.S. population, and Amazon Mechanical Turk (MTurk) to address concerns about the external validity of experiments with student participants....

We introduce DOSE - Dynamically Optimized Sequential Experimentation - and use it to estimate individual-level loss aversion in a representative sample of the U.S. population (N = 2;000). DOSE elicitations are more accurate, more stable across time, and faster to administer than standard methods....