Jacques Barzun wrote "Teacher in America". He also wrote "Clio and the Doctors". He was a professor of history and like many in that discipline, was not used to averages, spreadsheets and quantification. The Clio book is about the new and possibly suspicious branch of history that used quantification, statistical analysis and measures to analyze who did what back when.
This emergence of "quants", people who use quantification and numbers for research and investigation, is similar to my appearance as a young PhD in a school of education. Numerical analysis is a major tool in scientific investigations, especially in social and human sciences. It is similar to gambling. There is a book (there is nearly always a book) called "Gambling with Truth" and that is what experimenters do. They take a group of people or fruit flies or something and smile at them. They take another group, randomly chosen from the same population as those smiled on and frown at that group. Then, they test both groups: who knows the capital of Botswana, who lives longer, who has the most descendants. If the smiled-upon have a higher average or more of them know, live or procreate, there is evidence that smiling is helpful.
Of course, in the US, there are many possible strings attached. Those who wish to become teachers are often both required and persuaded to study their majors even more deeply and not fool about with silly research. They or their parents or taxpayers or current teachers may object to testing and research. Humans sometimes say that everything that happens is God's plan and it is unwise and sacrilegious to test His plan. Some will object that such testing was not done when they were kids and look how well they turned out. Others will be deeply committed to smiling, and demand re-testing until smiling emerges as clearly superior to frowning.
The undergraduates are busy with the other demands on their time and lives so traditionally the developing teacher runs into experimental design in graduate school, if at all. However, the nearly arrived "quant" assistant prof can be of use in testing. Not experimental testing but in school tests and grading. Traditionally, after teaching something, we like to test to see if it was learned. I wound up teaching "Tests and Measurements" but only to elementary school wannabees since other campus departments gave grades to every student without any instruction in the subject so why did their students need such a course? The little book I used to teach the subject can be found here:
https://sites.google.com/site/kirbyvariety/kirby-tests-and-meas-book
It is quite true that knowledge sometimes leads to power and scholarships and prestige. Couple that with the notion that some babies grow to be wonderful and some grow to be despicable and you can wind up with competition for grades. If I limit the number of high grades, then only some of the students will win one. If I give high grades to everyone, the grades won't mean much, so what am I going to do? I used to say, after a few years of experience, that I would work to get every student to the level of the highest grade (a smiley face in grade 2 and an A in junior year of college) in the time allotted. I used to ask for some to turn their backs on me and my subject, so I could satisfy my critics who said I was too soft and undemanding but very few helped me out by justifying a couple of low grades.
By the way, general discussions of quantified analysis are nicely presented in "Everybody Lies" by Stephens-Davidowitz and in "Uncharted: Big Data as a Lens on Human Culture" by Aidan and Michel.
