Calculating Power and P(Type II error) (A One-Tailed Z Test Example)
HTML-код
- Опубликовано: 7 мар 2012
- I have an updated and improved video on calculating power and beta available at: • Calculating Power and ...
I work through an example of calculating the probability of a Type I error, the probability of a Type II error, and the power of a one-tailed test.
You're welcome Ben! Good luck on your test!
was struggling to calculate beta and power of a right tailed paired t test. Got some relief by watching your video. Thanks.
You are very welcome!
Thanks! I'm glad you found this helpful.
all ur videos r awesome, its simple n lucid...... u earn a fan,,,, hats off to u.
I have a stats test tomorrow and was totally stuck on this concept. You managed to explain it in a way that was easy for me to understand. Thanks!
Those are areas under the standard normal curve. They can be found using software or a standard normal table.
Thanks for helping out my review session for ap stats 12 years later
I built 'em to last :) I'm glad to be of help!
I actually love your voice! haha also you just clarified hours worth of lectures in under 10 minutes!
How would effect size be affected if you used an alpha of .01 instead of .05?
Went to lecture for 3 hours, walked out with no clue. Saw this video and I really believe this video worth more than 150 dollars considering the tuition fee we're paying to the university breaks down to approximately 50 dollars per hour in a lecture.
this might be the easiest to understand stats video I have ever seen in my life
That's just a made up decision rule. A premise of the question is that we have decided to reject the null hypothesis if the test statistic is at least 2.4. Under that given decision rule, I calculate alpha, beta, and power. In my updated version of this video I start with a given significance level and go from there.
I was pretty clear up until the numbers of .008 and .945 came in, would you mind explaining where you got those from? Thanks
Thanks! I'm glad to help.
Thanks! I'm glad you liked it. I'm happy to add to all the free resources available out there.
You're welcome!
hi, in the example you use a z statistic but n
@Oliver Lester: Once you are given the distribution, you don’t use T even if they sample size is way less than 30.
This was so helpful thank you!
i still dont get where 2.4 came from, on the very first screen...
Him: "Get the logic down"
Me: "There is ZERO logic in statistics"
how did you get the .345? Magic?
Where did you get the 0.008 for the Type 1 error?
@Nyla McGee.He wants the true mean to be Uo=400 but this won’t happen when you found out that the 16 samples have a mean of 406 or when z>= 2.4. So 406 becomes your true sample mean which does not reflect your population mean of 400 assuming your population mean of 400 was actually the true mean or the true average of all the samples.So what is the probability of making this error.when you standardize 406 to (406-400)/(10/4). But p(z>2.4)=1-p(z
Where in the world are you getting 2.4 from ._. im so confused
I have an updated and improved version of this video available at ruclips.net/video/BJZpx7Mdde4/видео.html.
"Why are we using that rule? Well, we just are." I picked a value to illustrate the calculations. In this video I calculate the power given the rejection region is Z >= 2.4. I could have chosen a different z value, but I chose 2.4. If, say, we were using a significance level of 0.05 and an upper tailed test, the value would be 1.645 instead of 2.4.
amazing video sir ji
Type I error is actually p-value
YAY, it makes sense :)
Very confusing video
Justin Bieber!
You're welcome!