When color studies aren't real

I'm always amused when I read about certain "studies" paid for by companies to support some marketing project or advertising campaign. While there are most certainly credible research studies conducted on the efficacy of colors and how they relate to user/user experience, I must caution you. There are many many so-called "studies" out there that have absolutely no credibility. A sure sign is when a blanket statement is made, encompassing huge demographics and situations.

For example, I was at a lecture the other week, listening to an eco-friendly company rep talk about using green products for building. While he had a captive audience of color consultants in the room, he asked our opinion about a specific paint color chosen for a doctor's office.
image source
The architect of the project had supported his color choice by declaring that "studies had shown" that this particularly dreadful shade of diarrhea green was calming and soothing to patients. I wish I had taken a picture of this color sample- it was SO awful!

Then you have this article on US News and World Report about how car colors reveals the psyche of the driver. This "study" comes from CNW Marketing Research where they evidently asked nearly 1,900 Americans about their attitudes toward their own lives at several points over the course of a year. They also asked each participant the color of the car they drive most often, which allowed the researchers to develop a kind of color-confidence index.
Pixar images source
I am immediately skeptical... A color-confidence index? Obviously, automotive companies invest a Ton of money on market research to reach their target audience. But honestly, this seems just a tad bit formulaic. Maybe it works for caricatures, but we're more complex than that.
According to CNW, here's what the color of a car says about the person who bought it:
They were also able to calculate the "moodiness" of drivers—how widely their confidence varied from one extreme to the other, in the course of a year.
What do they mean by "confidence"? See what I mean? Recipes involving colors just don't work. There's no validity to that. What if a subject chose their particular car, not based on a color, but instead based on availability of that model? My car is silver gray. Probably one of the last colors I would have personally chosen, had I had options.

Well, that's my take on this, in any case. Anyone share my sentiments?