Partial least squares path analysis

Wayne Folta writes:

I [Folta] was looking for R packages to address a project I’m working on and stumbled onto a package called ‘plspm’. It seems to be a nice package, but the thing I wanted to pass on is the PDF that Gaston Sanchez, its author, wrote that describes PLS Path Analysis in general and shows how to use plspm in particular. It’s like a 200-page R vignette that’s really informative and fun to read. I’d recommend it to you and your readers: even if you don’t want to delve into PLS and plspm deeply, the first seven pages and the Appendix A provide a great read about a grad student, PLS Path Analysis, and the history of the field.

It’s written at a more popular level than you might like. For example, he says at one point: “A moderating effect is the fancy term that some authors use to say that there is a nosy variable M influencing the effect between an independent variable X and a dependent variable Y.” You would obviously never write anything like that [yup — AG], and most of your blog readers are pretty sophisticated.

It appears to me the PLS Path Analysis is an interesting alternative to SEM, based on partial-least-squares rather then ML. Same diagrams, similar results, similar procedures, different underlying mechanism/philosophy. And Gaston gives an interesting history of things and obviously put a lot of work into a 200+ page document and R package.

I don’t know anything about PLS path analysis but I thought I’d pass this on for the benefit of those of you who use these methods.

4 thoughts on “Partial least squares path analysis”

1. I agree that wouldn’t write “A moderating effect is the fancy term that some authors use to say that there is a nosy variable M influencing the effect between an independent variable X and a dependent variable Y” (and neither would I, for that matter) but only because (1) the word “nosy” seems wrong and (2) the grammar is odd — what’s an “effect between” one variable and another? I think you (or I) would write a very similar sentence, though…perhaps “A moderating effect is the fancy term that some authors use to say that there is a variable M that influences the relationship between an independent variable X and a dependent variable Y.” Except that would be plagiarism. But you know what I’m saying.

• If only there was some sort of baseball connection . . . then we’d get some comments!

2. i’d be willing to say it’s because, for reasons beyond my understanding, partial least squares simply hasn’t caught on a lot among the social sciences.

i’m not gonna name names but some big-name conference accepted what i would consider a pretty crappy paper from moi just because it had “partial least squares” in the title and people wanted to know what was that all about. great teaching moment…