SUMMER SCHOOLS

This three-day course provides a guide to experimental economics researchers on best practices for translating an idea for an experimental design to an implementable experiment, whether in the laboratory or online. In contrast to courses which focus on software packages (such as oTree or z-Tree courses), on this course you will learn how to create a program that is efficient, reliable, and which enables collaboration, adaptability, and reproducibility. In the first part of the course, we will implement two sets of commonly seen experiments using oTree. We will introduce a conceptual framework for implementing experimental designs, which includes defining the different states of an experiment, and developing storyboards and wireframes for the information screens, input screens, and calculations/output screens. We will then demonstrate how to translate the designs to software, including topics such as defining groups and matchings, sequential decision-making, multiple decision-making periods, randomization devices, and systematically recording variables for logging and reproducibility. In the second part of the course, we will provide a brief introduction to key technologies used in producing professional-looking applications, including webpage generation via templates, CSS and Bootstrap, and JavaScript and jQuery. There will also be opportunities for discussions centered on specific aspects of experimental designs based on individual participants’ needs. Throughout, we will emphasise the use of git and GitHub for version control and collaboration, and management of Python virtual environments for reproducibility and portability. This is therefore not just another oTree course! By the end of the course, you will acquire skills which are transferrable to other platforms or technologies.

Lecturers:

• Ted Turocy (University of East Anglia)
• Boon Han Koh (University of Exeter)

Fee:  125 EUR

Capacity: 25 participants

Registration

This course aims to provide a critical view on the “rules of a game named science” and introduces remedies to the manifold issues jeopardizing the credibility of scientific results: power calculations, confirmatory research (pre-registration), and open and transparent research practices (see, e.g., Wagenmakers et al., 2012; Miguel et al., 2014; Munafo et al., 2017). In addition, the course covers the fundamentals of meta-analysis, a method used to systematically combine results from multiple studies to assess consistency, resolve conflicting findings, and improve the precision of effect size estimation (e.g., Stanley, 2001; Borenstein et al., 2009; Irsova et al., 2024).
Upon successful completion of the course, students are expected to:
(i) understand why and how common (mal)practices in scholarly research translate into low levels of replicability and low reliability;
(ii) understand the virtue of confirmatory research and transparent research practices regarding the credibility of research findings;
(iii) be able to undertake a priori power calculations and sensitivity analysis, and to devise pre-analysis plans for research projects;
(iv) be able to critically assess scientific projects and results with respect to malpractices, research integrity, and ethical aspects;
(v) understand the relative advantages of conducting meta-analyses as compared with narrative literature reviews;
(vi) understand the principles of meta-analysis, including such steps as study selection, choice of summary statistics, moderators, and models;
(vii) be able to conduct a meta-analysis using Stata, including such steps as heterogeneity assessment, subgroup analysis, meta-regression, and bias detection and mitigation;
(viii) be able to critically evaluate published meta-analyses, identifying methodological strengths or weaknesses.

Lecturers:

• Felix Holzmeister (University of Innsbruck)
• Matteo M. Marini (University of Bologna)

Fee: 125 EUR

Capacity: 25 participants

Registration

Registration