Use "predictor" instead of "independent variable" in regression models
Wisdom from Frank Harrell, author of "Regression Modeling Strategies"
I recently wrote an article to discourage using the term “independent variable” in regression models, and I provided a strong reason for doing so: The word “independent” has a specific meaning in probability theory, and I want to avoid any confusion with that concept.
Some people passionately1 disagreed with me. I found none of their arguments to be persuasive, and I sense that some of them viewed my claim as my own unique perspective. Actually, many statisticians agree with me about this, including one notable professor.
Frank Harrell also discourages using the term “independent variable” in regression models. He is a Professor of Biostatistics at Vanderbilt University, and he is the author of “Regression Modeling Strategies”. Many people use this textbook as the definitive source for learning regression analysis.
The following quotation comes from one of Dr. Harrell’s answers on “Cross Validated”, a discussion forum about statistics under the “Stack Exchange” family.
In many ways, "independent variable" is an unfortunate choice. The variables need not be independent of each other, and of course need not be independent of the dependent variable Y. In teaching and in my book Regression Modeling Strategies I use the word predictor.
The original question came from someone who was understandably confused about the term “independent variables” in regression analysis - due to exactly the reason that I mentioned about probabilistic independence. There are others who agree with me and Dr. Harrell in that discussion, and I encourage you to read it entirely.
If you are reading my posts for the first time: I'm Eric Cai, a statistician based in Toronto, Canada. I write about statistics, communication, and career development for professionals in data & analytics. Subscribe to get my articles delivered to your inbox at 9:30 AM Eastern time on Monday to Friday.
At least 2 people disagreed with me in the comments under my LinkedIn post, then admitted that they actually had not read my rationale beforehand. Nonetheless, their rhetoric was insulting and intellectually dishonest. I do NOT tolerate this type of behaviour in the comments under my writing.