yego.me
💡 Stop wasting time. Read Youtube instead of watch. Download Chrome Extension

Comparing P value to significance level for test involving difference of proportions | Khan Academy


2m read
·Nov 11, 2024

A veterinarian is studying a certain disease that seems to be affecting male cats more than female cats. They obtain a random sample of records from 500 cats. They find 24 of the 259 male cats have the disease, while 14 of 241 female cats have the disease.

The veterinarian uses these results to test their null hypothesis that the true proportion is the same amongst the male and female cats versus the alternative hypothesis that the proportion of males who get the disease is actually higher than the proportion of females. The test statistic was z is equal to 1.46, and the p-value was approximately 0.07. That's useful; they've done a lot of work for us.

Assume that all conditions for inference were met. At the alpha equals 0.10 level of significance, is there sufficient evidence to conclude that a larger proportion of male cats have the disease?

So pause this video and see if you can answer that.

All right, so remember what we do with the significance test. We assume our null hypothesis, and then given that assumption, we look at our data and we calculate a p-value. We say, "Hey, what is the probability of getting the data that we did, assuming our null hypothesis is true?" If that is lower than our significance level, then we reject the null hypothesis because we say, "Hey, assuming the null hypothesis, we got a pretty unlikely event."

We reject it, which would suggest the alternative. Here, our p-value is 0.07, which is indeed less than our alpha, which is less than our significance level—this thing right over here, 0.10. Because of that, we would reject the null hypothesis.

Rejecting the null hypothesis would suggest that there is sufficient evidence to conclude that a larger proportion of male cats have the disease. So we could say it suggests the alternative hypothesis.

On Khan Academy, there are some multiple choice questions like this, and the choice that you would pick might be something like, "Because our p-value is less than our significance level, we would reject the null hypothesis," and that would suggest the alternative.

More Articles

View All
I'm moving back to California. I'm done.
Hey guys, so this is definitely not the video I was planning to make, and it’s certainly not a video I want to make, but I think I owe you complete transparency to tell you what happened and why I’m back in California. First of all, you probably already …
Lasting Lessons from Charlie Munger.
Charlie Munga: businessman, investor, mathematician, meteorologist, developer, lawyer, husband, father, and philanthropist. But most importantly, Charlie was a thinker. As much as he’s known for his investment track record and decades-long partnership wit…
How Facebook is Stealing Billions of Views
A few days ago, Facebook proudly announced that they’d achieved eight billion video views a day. That’s really impressive, until you learn that in the first quarter of 2015, seven hundred and twenty-five of the 1000 most-viewed videos on Facebook were sto…
Comparing proportionality constants
We’re told that cars A, B, and C are traveling at constant speeds, and they say select the car that travels the fastest. We have these three scenarios here, so I encourage you to pause this video and try to figure out which of these three cars is travelin…
Warren Buffett: How to Invest for 2023
So 2022 was a rough year for investors, and people are worried about what’s ahead. That’s not a secret. The US stock market has been down over 20 percent, and this only tells part of the story. There are many stocks that were formerly high flyers that are…
Batten Down | Life Below Zero
Like we’re stuck at home late. Red-flag! I know for three days I should go get firewood, and we should go get a couple days’ worth of something to eat here: caribou or a few ducks. The Hailstone family spends their summer living in Kowalik, away from the…