“Garbage in—Garbage out” is a long-standing maxim of those who develop psychological measuring instruments, as well as some data analysts who are aware of the role of measurement in the data analysis enterprise. This maxim means that if you have poor measurements—the numbers that are used in all statistical procedures—that your resulting statistical analyses will also be of poor quality. No amount of manipulation of poor numbers will enable useful findings to result from bad measurements. The purpose of this course is to sensitize you to the issues involved in creating good psychological measurements, which then allow the use of basic and advanced statistical methods to extract meaning from numerical data. To accomplish this objective, we begin with a discussion of why we measure (answer: because people differ on every psychological characteristic) and the role of measurement in science. This will lead us to considering how, in the framework of scientific method, psychological observations are converted into numbers so they can be used in both research and applications of psychology in clinics, schools, business and industry, and elsewhere. Next we will address the two main ways that psychological measuring instruments are evaluated—reliability and validity—as well as methods for demonstrating and evaluating these criteria, followed by an excursion into methods for the construction of measuring instruments. This will include a special class of instruments that are based on special methods designed to classify people into empirically based categories, as well as the mainstream methods for measuring performance-based or cognitive variables (abilities, achievement, aptitude, skills) and self-report measures that are used to measure personality, attitudes, preferences, and other variables. Finally, we will take a brief tour of some advanced measurement methodologies that have many advantages over the classical measurement methodologies, including computerized adaptive testing to which you might have already been exposed. We’ll end the course with a discussion of methods for analyzing and correcting for item and test bias to make tests and other measuring instruments more equitable and fair for individuals from diverse groups. The intent is that when you complete this course you will know what good (and bad) measurements are, how to recognize them, and if your path through the future requires it, how to construct good measurement instruments. You should also have a good appreciation of the problems involved in psychological measurement and why psychological measurement is much more complicated (and, therefore, challenging) than measurement in other sciences.
Gopher Grades is maintained by Social Coding with data from Summer 2017 to Summer 2025 provided by the University in response to a public records request
Not affiliated with the University of Minnesota
Privacy Policy