Data scientists are often trained in the analysis of data. However, the goal of data science is to produce good understanding of some problem or idea and build useful models on this understanding. Because of the principle of"garbage in, garbage out," it is vital that the data scientist know how to evaluate the quality of information that comes into a data analysis. This is especially the case when data are collected specifically for some analysis (e.g., a survey).
In this course, you will learn the fundamentals of the research process—from developing a good question to designing good data collection strategies to putting results in context. Although the data scientist may often play a key part in data analysis, the entire research process must work cohesively for valid insights to be gleaned.
Developed as a language with statistical analysis and modeling in mind, R has become an essential tool for doing real-world Data Science. With this edition of Data Science Research Methods, all of the labs are done with R, while the videos are tool-agnostic. If you prefer your Data Science to be done with Python, please see Data Science Research Methods: Python Edition.
What you'll learn
After completing this course, you will be familiar with the following concepts and techniques:
Data analysis and inference
Data science research design
Experimental data analysis and modeling
To complete this course successfully, you should have:
A basic knowledge of math
Some programming experience – R is preferred.
A willingness to learn through self-paced study.
The Research Process
Planning for Analysis
Correlational and Experimental Design
Note: This syllabus is preliminary and subject to change.
Meet the instructors
Sr. Content Developer
Ben is a Sr. Content Developer for Microsoft's Learning and Readiness team, and is an analytics professional and educator with over 8 years of industry and managerial experience. Prior to joining Microsoft, Ben ran and directed multiple consulting firms, where he also held critical analytics roles in companies as diverse as Juniper Networks, Costco, and T-Mobile. He has taught Data Visualization at The University of Washington, and recently founded Seattle Pacific University's Analytics Certificate Program.
Assistant Professor of Psychology, Data Science consultant
Seattle Pacific University
Dr. Tom Carpenter is Assistant Professor of Psychology at Seattle Pacific University, and is also a Data Science consultant. His areas of expertise include personality-social psychology, research methods, and statistics. His teaching focuses on introductory and advanced research methods and statistics in psychology as well as social and personality psychology. Dr. Carpenter’s research focuses on our hypocritical human nature: our propensity to ignore our overt preferences and standards and to transgress against ourselves and others. One line of research in this area focuses on implicit bias, the impulsive thoughts that can undermine our higher reasoning. Dr. Carpenter has developed new software methods for running the Implicit Association Test (IAT) using online survey software (read more here: www.iatgen.wordpress.com). A second line of research focuses on guilt, shame, and self-forgiveness, specifically focusing on the functions of 'guilt-proneness' and 'shame-proneness' as well as associations with the general ability to forgive the self. Finally, Dr. Carpenter has conducted research related to his area of teaching (statistics education).