Since I'm avoiding compiling my own students' grades going in to their final exam, I glanced.
It's an undergraduate research project, not peer reviewed published research.
Teeny tiny sample sizes, high drop out rate.
No actual hypothesis test nor credible intervals (statistics) done, from appearances; the student seems to just look at his graphs and tell you what he thinks.
Even an undergrad should know better.
I am going to guess if actual tests were done, the standard errors would be high enough to render the whole thing uninterpretable. And that's on the big if that the study itself was well designed with no bias. including self selection. And that doesn't even get into the problem of controls, basic study design, etc.
I'm glad the undergrad learned to use physiology measurement equipment, but his analysis would not have passed my intro to stats class 100 level class, and I"m a generous grader.
K
now to finish grading my students' projects.
