2-151 - Step by Step: Improving best practices for research with children

Session Type: Paper Symposium

Integrative Statement

Given the recent “replication crisis” (Nelson, Simmons, & Simonsohn, 2018), psychology researchers are increasingly interested in improving their research practice. However, researchers working with children face particular challenges in how to do so in the context of our data and workflows. At the International Conference on Infant Studies (2018) we began a conversation about best practices in infant research, which was met with great enthusiasm (over 100 Tweets and retweets), but it has left open important questions for implementing these best practices in everyday developmental research. Here, we continue the conversation on improving developmental research.

Paper 1 goes beyond describing open science practices to providing principles and tips for implementing these practices in a developmental researchers’ day-to-day workflow. Paper 2 tackles the topic of pilot studies, delineating where they can improve research design, and where they can have undesirable outcomes (e.g., Type-I error inflation). Papers 3 and 4 focus on practical recommendations for studies using looking-time measures, which is a common measure in infant laboratory studies. Paper 3 explores sources of variability in looking-time studies and gives practical recommendations for increasing statistical power. Paper 4 discusses statistical transformations of looking time data (i.e., log transformations) that avoid Type-I error inflation and increase statistical power. The talk will also give guidance for interpreting transformed data.

Together, this session will provide hands-on advice for improving research practice, and will generate lively debate about when and how to embrace proposed changes to the field.

Sub Unit

• Methods, History, Theory

Chairs

• Esther Schott, Concordia University

• Krista Byers-Heinlein, Concordia University

Individual Presentations

• Making open science work for you: Practical tips for incorporating open science practices into your everyday workflow - Presenting Author: Jessica Elizabeth Kosie, University of Oregon

• Planning your pilot study: What to do and what to avoid - Presenting Author: Esther Schott, Concordia University; Non-Presenting Author: Anne Scheel, Eindhoven University of Technology; Non-Presenting Author: Mijke Rhemtulla, University of California, Davis; Non-Presenting Author: Krista Byers-Heinlein, Concordia University

• Infant looking-time data: Effects of variability, number of trials, and sample size on statistical power - Presenting Author: Michaela Carol DeBolt, University of California, Davis; Non-Presenting Author: Mijke Rhemtulla, University of California, Davis; Non-Presenting Author: Lisa Oakes, University of California, Davis

• The benefits and practical guidelines to log-transforming looking time data - Presenting Author: Angeline Sin Mei Tsui, University of Ottawa; Non-Presenting Author: Patricia Brosseau-Liard, University of Ottawa