VehiclesFashionRecipesBlogsHuntTravelsSportFunHandmadeITEducation
Mini-Games
x

x
zakruti.com » Knowledge, science, education » Crash Course
Controlled Experiments: Crash Course Statistics #9

Controlled Experiments: Crash Course Statistics #9

FBTwitterReddit

video description

Rating: 4.0; Vote: 1
We may be living IN a simulation (according to Elon Musk and many others, but that doesn't mean we don't need to perform simulations ourselves. Today, we're going to talk about good experimental design and how we can create controlled experiments to minimize bias when collecting data. We'll also talk about single and double blind studies, randomized block design, and how placebos work
Date: 2022-04-04

Comments and reviews: 10


To be clear: that's not what placebo effects are. Sham treatments don't cause anyone to feel better, though they may bias subjects toward -reporting- that they feel better. You correctly note that the cut may have healed without Neosporin. So if a cut heals with a placebo ointment, -placebo effect- is a catch-all category accounting for natural progression, regression to the mean, effect of treatment on reporting bias, etc. ; but the placebo isn't actually responsible for any healing or pain reduction.
reply

I'm not an expert, but from what I know multiple studies don't want to use placebos because of ethical concerns. Instead, the control group is given the old, standard treatment while the experimental group received the experimental treatment. It would have probably been nice if ethical concerns like that would have been addressed in the video. Despite this point of criticism the video was still pretty well made.
reply

-Elon Musk, along with actually intelligent people he read and the repeated from, argue that. -
lol, can we stop quoting people on things they didn't come up with and don't have an understanding of just because they happen to be be wealthy please? There's no reason to treat Musk as if he as a clue what he's talking about outside of perhaps specific parts of engineering.

reply

I have a question: when we compare two variables, and we do linear regression, we tend to see a line that can be described as y = mx+b
But I was wondering, do we ever see variables that are connected with an exponential correlation? The reason I ask is because so much of statistics seems to revolve around linear correlations of variables.

reply

Climate change is dependent on the Grand solar Minimum coming and the current Magnetic weakening and Pole shift happening at the same time more than certain kinds of pollution IMHO. I mean what are the odds the environment will follow the same rules it followed the last great die offs? 100% or 100%?
reply

This is by far the best Statistics course i've ever taken! Instructor is great, jokes are funny, and i love how the personality comes out! Great work here, team! Even though my background is in applied mathematics, statistics, and data science. this course has been the best stats class i've taken!
reply

I'll reverse that question next time you see a prescription you got from the doctor for Oxycontin ask yourself how was it tested on how many people could be addicted to this stuff and what's in the chemical makeup to make it so addictive more so than sugar? Stay curious
reply

Ich feel you skipt the problem of sampel size. I need a large enough group of people in the experiment. If I take 30 people, splitt them into 3 random groups, chances are these groups are quite dissimilar. This alone might invalidate the experiment.
reply

You have made statistics really fun, especially because you pick up several examples from different fields thus some actual real life applications, that doesn't happen really in classrooms! Which actually should be the way to go about things.
reply

The Philly sugar tax didn't encourage people to buy less, it encourage people to buy out of the city limits instead, that and it propped up a small black market.
reply
Add a review, comment






Other channel videos