zlacker

To learn and demonstrate understanding of statistical concepts, no calculator is required. In fact calculating and graphing by hand are great for learning. To bridge the gap to the real world, a computer, however crappy, with Excel installed, however outdated, is infinitely better. (Not that I endorse Excel, it’s just the most common tool among the general public.)

I happen to be a physicist too and while I’m not an experimentalist, I’ve been through plenty of experimental training, and have participated in real world data analysis projects. Never once have I seen any physicist doing any statistics with a graphing calculator (I did see a few when I taught undergrads mostly from other departments, so there’s that).

>>oefrha+(OP)
Perhaps you can learn without a calculator, but these timed statistics tests do not function without one. Do you really expect people to do repetitive operations on even n=10 datasets when they only have an hour? You can’t use Excel (because it’s on more capable PC that you can use to cheat).

At the end of the day, if you want to remove the calculator from the statistics classroom you probably also have to remove the standardized test.

>>oefrha+(OP)
Yes, Excel is superior, but part of the beauty of the graphing calculator is the limited feature-set. It works well for classrooms and especially test-taking environments.

>>pietro+54
Maybe they should do as when I learned and reduce the size of the data set.

>>pietro+54
I’ve been through tests with statistical problems where calculators are forbidden, so this is completely false.

>>gtk40+y5
Where calculators are absolutely not necessary, doesn’t help understanding and more of a waste of time.

>>pjmlp+R7
Actually, even n=10 is really nothing, not being able to do that kind of calculations by hand reasonably quickly is more of a reflection of terrible basics, which isn’t surprising in American high schools.

>>pietro+54
Take a look at some UK tests for example as they don't use graphing calculators, just scientific ones. I can't say my education was worse for it.

It does cost more to mark those tests than pure multiple choice though.

>>pietro+54
You can always make problems with steps that involve "easy" numbers. My experience with high school math was that if you wrote 1.414 when the answer was sqrt(2), you got the problem wrong. So I am not sure what the calculators added, really.

>>jrockw+qI
If the answer is ugly, always try squaring it or dividing by pi to see if you get something that looks rational.

>>edflsa+XK
Or use a calculator with some sort of CAS.

>>oefrha+M8
For a stats class, even n=10 is tedious and absolute overkill. N=3 or 4 is entirely sufficient to prove the student understands the process.

>>edflsa+XK
Yeah, it's amusing how often that works. I remember taking the amateur radio exam which involves some path around impedance and power. The answers were always in the form of 0.5, 1, 1.414, 2. It's always 1.414 (or 0.707, its close cousin).

>>jrockw+R41
The ham radio exam is a joke (at least for the technician class in the U.S.). Just a bunch of multiple choice questions from a public question pool. I literally went through the pool twice before my exam and got a perfect score, although I hardly knew how to install and operate radios. (I just needed the license to be able to remotely operate a radio telescope.)

>>oefrha+p81
If I recall correctly there is no math until you get to Extra. But I agree that the questions are trivial and the multiple-choice format makes it even more trivial.

I am honestly shocked that there are any operators that aren't Extra class.