Here's a magisterial handbook chapter, by Muriel Niederle
Experiments: Why, How, and A Users Guide for Producers as well as Consumers by Muriel Niederle
NBER Working Paper 33630, DOI 10.3386/w33630, March 2025A
Abstract:"This chapter is intended as an introduction to laboratory experiments, when to use, how to evaluate them, why they matter and what are the pitfalls when designing them. I hope that users as well as consumers will find Sections that broaden their views. I start with when an economist might want to run an experiment. I then discuss basic lessons when designing experiments. I introduce a language to start a systematic description of tools we have when designing experiments to show the importance or role of a new model or force in explaining behavior. The penultimate chapter provides an advanced toolkit for running experiments. I end this chapter with my views on pre-registration, pre-analysis plans and the need for replications, robustness tests and extensions. "
...
"While I hope to convey general lessons, I will make them more accessible and understandable by providing specific examples. These will often, though not always, come from my own papers, or from economists whose work I know exceptionally well (mostly my advisors, students or coauthors). While this may seem self-serving, the main reason is that for those experiments I know– rather than have to infer– why authors made certain choices. And one aspect of experiments that will become obvious almost immediately, is that they require the researcher to make a lot of decisions. This chapter therefore is in no way a literature survey, nor really a highlight of amazing papers. It rather showcases papers whose history I am exceptionally familiar with. I will also not provide negative examples, but rather present potential pitfalls, with one exception in Section 3.1 "
Posts
