[We’re pleased to welcome Eunseong Cho of Kwangwoon University in the Republic of Korea. Professor Cho recently collaborated with Seonghoon Kim of Chungbuk National University on their paper “Cronbach’s Coefficient Alpha: Well Known but Poorly Understood” from Organizational Research Methods.]
- What inspired you to be interested in this topic?
It was due to my ignorance in an ordinary sense and my knowledge in a Confucian sense. I never dreamed of writing this kind of methodology paper two years ago because I was not a quant wizard or statistics expert. As a professor who regularly taught research methods in my school, I thought that I knew coefficient alpha because I had the commonsense knowledge about it. For example, most textbooks explained that it was first developed by Cronbach and its value was between zero and unity. Two years ago, I accidentally recognized that I had little knowledge of it. Confucius said, knowledge is recognizing that you know what you know, and recognizing that you do not know what you do not know. In a Confucian sense, it was the moment when I got the real knowledge because I got to know the extent of my ignorance.
After writing an initial draft, I invited Prof. Seonghoon Kim to this project. That the author is a layman can have the merit of writing an understandable article in a layman’s eyes. This study also needed an expert author because it should be technically sophisticated in an expert’ eyes to pass the peer review. He was the right person to do the job. He wrote several articles on the related topics, which were published in respected journals such as Psychometrika. He transformed my rough draft into much developed one. I am also grateful to the Associate Editor and the two anonymous ORM reviewers. Their insightful ideas helped us include many value-added components into this study.
- Were there findings that were surprising to you?
Several things surprised us. First, we were surprised to find that many textbooks on research methods provided incorrect explanation on the coefficient alpha. Second, we were surprised to find that there were few comprehensive and easily comprehensible studies on this topic. A non-expert reader should read dozens of related papers to fully understand this topic. A majority of previous studies required a considerable level of mathematical background knowledge beyond and above what typical social scientists had. Surprise is a good thing in science. A successful study is to find a surprising thing, which is rare by definition. Only what surprised the authors can surprise the readers. We therefore interpreted such surprises as an opportunity for a new study that disproves common misconceptions about coefficient alpha in a comprehensive and accessible manner.
- How do you see this study influencing future research and/or practice?
This study can influence many studies in many disciplines in many ways. Numerous studies in diverse areas of social sciences report reliability information. Coefficient alpha, usually called “Cronbach’s alpha,” is an automatic and unconditional choice in most studies. This study suggests that many things such as what to call, how to use, and even whether to use it should change. For example, this study suggests that coefficient alpha should be used as a reliability estimator only if the data are unidimensional and tau-equivalent. We do not claim that we are the originators of these ideas. The contribution of this study is to provide them in a more comprehensive and accessible way. Occasionally, people continue to use a more familiar alternative (e.g., the QWERTY keyboard layout) instead of switching to more effective one (e.g., the Dvorak keyboard layout). The key point is to lower the switching cost of typical users. Without such a study, one of the least effective method will continue to enjoy its decades-long monopoly status in the reliability coefficient market.
You can read “Cronbach’s Coefficient Alpha: Well Known but Poorly Understood” from Organizational Research Methods for free by clicking here. Want to know about all the latest research from Organizational Research Methods? Click here to sign up for e-alerts!
Eunseong Cho is an associate professor of marketing, College of Business Administration, Kwangwoon University, the Republic of Korea. He received his PhD from Korea Advanced Institute of Science and Technology. His current research interests include exploratory factor analysis, construct validity, gift-giving behavior, and determinants of article influence.
Seonghoon Kim is an associate professor of education, College of Education, Chungbuk National University, the Republic of Korea. He received his PhD from the University of Iowa. His primary research interests include educational measurement and statistics.