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

Conclusion for a two sample t test using a P value


2m read
·Nov 10, 2024

We're told a sociologist studying fertility in Argentina and Bolivia wanted to test if there was a difference in the average number of babies women in each country have. The sociologist obtained a random sample of women from each country. Here are the results of their test.

So they take a sample of 75 women in Argentina, and these women had a mean of 2.4 babies each, with a standard deviation of 1.5. Then the standard error of the mean was 0.17. Then they calculated similar statistics for Bolivia.

Then they give us the t-test for the means being different. We were able to calculate these statistics, and they say assume that all conditions for inference have been met at the alpha equals 0.05 level of significance. Is there sufficient evidence to conclude that there is a difference in the average number of babies women in each country have?

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

All right, now let's work through this together. So this is classic hypothesis testing right over here, where your null hypothesis is actually going to be that your means are the same—so that the mean in Argentina is equal to the mean in Bolivia.

And then your alternative hypothesis is that your means are different. What you do is you say, all right, if we assume the null hypothesis, what is the probability that we would have gotten means this far apart? That's what our p-value tells us. We have a 0.31 probability, or 31 percent probability, of getting means this far apart.

Now, if your probability, assuming the null hypothesis, is below your level of significance, your alpha right over here, then you would say all right, that seems like such a low probability. I'll reject the null hypothesis, which suggests the alternative hypothesis.

But in this situation here, if we compare our p to our alpha, we see that our p-value is for sure greater than our alpha. So in this situation, I mean, you could see it right over here: 0.31 is for sure greater than 0.05.

So in this situation, we cannot reject the null hypothesis. Cannot reject our null hypothesis, and so there is not sufficient evidence to conclude that there is a difference in the average number of babies women in each country have.

More Articles

View All
Worked example: Determining the effect of temperature on thermodynamic favorability | Khan Academy
Let’s do a worked example where we calculate the standard change in free energy, ΔG⁰, for a chemical reaction. For our reaction, let’s look at the synthesis of ammonia gas from nitrogen gas and hydrogen gas at 25 degrees Celsius. ΔH⁰ for this reaction is…
The Times When Paranoia Fueled Technological Advancement
We’re here to announce our pills. Yes, brain pills. Yes, make you smart. Dalton plus Michel pills.com brain pills. Yes, smart guy brain pills. They will protect you from overb believing in conspiracy. [Music] All right, this is Dalton plus Michael, and t…
Implanting Memories | Breakthrough
My work focuses on finding individual memories in the brain and actually turning them on or off. We had a series of projects where we started off by asking really simply: can we go in and can we just find a memory in the brain? Can we isolate a memory in …
My Thoughts On Paying Higher Taxes | Kamala Harris Tax Plan
So first of all, let me just say this: initially I was not planning to make a video on this topic because, one, I really dislike involving politics on the channel; two, I don’t want anything I say to be taken out of context; and three, I just don’t know h…
How to Talk to Aliens
[Michael] Where is everyone? We have been listening for messages from outer space for more than half a century, and so far… silence. Why? Are we truly alone in the universe? Or is everyone else acting like us and just doing a lot of listening? Maybe we ne…
Influence of political events on ideology | AP US Government & Politics | Khan Academy
In discussing political socialization, we’ve talked a lot about factors that go into how people develop their opinions on government and politics. Your family, your friends, your demographic characteristics like your race or gender, they all contribute to…