How to reduce probability of type ii error Videos

• 1. Type I and Type II Errors
• 2. Type I Errors, Type II Errors, and the Power of the Test
• 3. Statistics 101: Controlling Type II Error using Sample Size
• 4. Margin of error 1 | Inferential statistics | Probability and Statistics | Khan Academy
• 5. AP Stat Ch 21 Video 2 Tests and Intervals.mp4
• 6. 20. Examples of Errors in Hypothesis Testing and Impact of Alpha
• 7. Crash Course in Statistical Error (Part II)

Type I and Type II Errors

User Comments

MrsCoachA commented on 29 Jan 2015

Hey Tony, your "answer" is based on a sample that has been taken. We take the sample to provide evidence to make a conclusion regarding the null. And even with our best made plans, sometimes samples do not reflect the population. So if we happen to get one of these weird samples, it can cause us to reject a null that is actually true, or fail to reject a null that is false. This is how we get the errors mentioned.

pavlaras83 commented on 21 Aug 2013

In the passage rate example, do you mean that the average grade of the 150 students of the sample is 88/100 or that the proportion of students that passed (in terms of either pass or fail) the final exam was 88%. Im asking that because if the latter is the case, what is the meaning of a standard deviation of 4% on that proportion? Thanks! And of course thanks for all your videos!!!

Tony Gambino commented on 21 Jan 2015

How would you make an error though? Do you mean BEFORE you run the statistical test? Because if you run the test you would have the answer so how could you have a Type I or Type II error??

Jennifer Furtado commented on 18 Nov 2015

Very good and crisp explanations of Z test, T test and Type I and Type II errors. Good work. could it possible for you to load videos on Residuals, Principal Components and Scatter plots

mzgemini0006 commented on 24 Mar 2013

I'm still confused.... but the good thing about Youtube is you can play it over and over again lol..... hopefully it will stick in my head. I wish you had a few more examples

ajcoder commented on 16 Oct 2012

Stupid question. What would make you reject it when it was the 95%. Is it anything over 95 percent?

craig listly commented on 04 Jun 2015

thank you.some examples of actual problems would help make idea clearer though.

aznxbigazn commented on 23 Nov 2011

wow very useful, love this tutorial you should be a Prof or T.A

elleteed commented on 28 Apr 2011

Helpful but you need a better microphone, too much noise!

amomi28 commented on 15 Feb 2012

Great info! Thanks for providing such a clear example.

John Cox commented on 10 Feb 2015

shouldn't your parameter be p, not mu? nice video

YY Guo commented on 07 Nov 2012

Thanks. My stat professor is a moron too.

rosewaltzer commented on 06 Nov 2012

thank you, my stats professor is a moron

Type I Errors, Type II Errors, and the Power of the Test

User Comments

Matthew Taylor commented on 12 Apr 2015

Thanks for this video. Very clear and well paced. Do you have more on power? I was hoping to find more about how power, variance and sample size interact.

jbstatistics commented on 13 Apr 2015

+Matthew Taylor I have a video that walks through the factors that affect the power (of a Z test), available here: https://www.youtube.com/watch?v=K6tado8Xcug.

uttam “mover” bhatta commented on 14 Feb 2015

this really helped ...many thanks JB.............thank u

jbstatistics commented on 14 Feb 2015

You are very welcome.

Statistics 101: Controlling Type II Error using Sample Size

User Comments

Lex Gabrees commented on 12 Dec 2015

Just a quick question ... So would we only control the Type II error if we were testing our from the perspective of the alternative hypothesis ? What I'm actually asking is : in practical situations, when would we want to control the Type II error ?

S Kool commented on 17 Oct 2015

This is the best series I have come across on hypothesis testing. Thanks a lot for all your efforts in making these videos. I have become you fan! I will surely watch all other playlists you have updated on Statistics.

Kang Emmanuel commented on 28 Jun 2015

I think controlling type 2 error is impossible since we cannot choose one point of mu alternative. This means that we ALWAYS cannot say that we accept the null hypothesis. Am I right?

Jiong Wang commented on 06 Jun 2013

thank you so much for sharing this, Brandon, you can never imagine how a group of chinese students are benefiting from your video, this is alot more helpful than our text book here !!

Antony Gelberg commented on 28 Jun 2015

"By aligning on critical values, aligned on different means which cannot happen". Not quite getting the impact of that statement - any tips?

Muhammad Akhtar commented on 12 Sep 2013

Sir imaging, thanks for sharing all this useful information to us. your style of communication is awesome,

Genma Saotome commented on 11 Nov 2015

OMG.. this is awesome... :)

Hans de Cocq commented on 07 Sep 2013

Thanks for sharing!!!!!!

Margin of error 1 | Inferential statistics | Probability and Statistics | Khan Academy

User Comments

Panagiotis Goulas commented on 21 Oct 2013

The fact that you replace σ with s seems risky to me. For example when you study normal or not normal population and you don't know σ you use the same notation to estimate σ through s but you never use an arithmetic value for s (later in this occasion you go to your t scores and get it done). It's like saying I got a sample and I found this standard deviation so I am gonna use it to solve my problem. With the same logic I found the sample mean so since it estimates the population mean

parvanoff1 commented on 08 Nov 2010

Great like always Sal, i was just thinking about what if you post under downloads of the khanacademy home page just the screenshots from Paint of all yours videos. In that way i think we could have a quick preview of all the Info in that lecture and we would not always need to play the whole video when just refreshing the memory ;) . I hope you get it what i mean. You are just great!!! Thanks a lot !

delcapslock100 commented on 17 Aug 2013

Actually he qualifies that statement by saying he is "reasonably confident" that there is a 95% chance that the true mean is in that interval, and he makes the point over and over that this is really a best guess, because we don't know the true standard deviation of the sampling distribution.

Norwana commented on 02 Jun 2012

actually your wording is a bit incorrect, you don't say there is a 95% chance that it is in that intervall, you say that you are confident in 95 of 100 times that it is in that intervall - it has nothing to do with chance in a confidence intervall!

MixTheMeister commented on 08 May 2011

This question has popped up several times when I've been watching these videos. Why can you say that μx = μ = p I agree that μx would approximate μ = p depentent on the size of the n. Wouldn't rather be μx ≈ μ = p?

MixTheMeister commented on 09 May 2011

@BreakerByte Thanks, I've actually viewed the entire playlist. The answer dawned on me basically as soon as I answered the question, but thanks anyway. I need to get through the probability playlist as well though

seeknaylover422 commented on 15 Dec 2015

At 5:30, why is the sample variance divided by (n-1) instead of n? In the most recent videos, the standard error for the sample mean was simply divided by n. Is Bernoulli different? Why is it (n-1?

oxblood commented on 02 Dec 2012

He didn't calculate the values but the margin of error is the the difference between the left bound and the right bound divided by two.

SteinN 'd commented on 21 Aug 2013

Why the variance of the sample is divided by 100 - 1? And if so, in what of those videos can I find this?

Celovis commented on 09 May 2011

@MixTheMeister check out the whole playlist on statistics, he explains this in a previous video.

Celovis commented on 09 May 2011

Am I the only one that thinks that donating money to KhanAcademy would be a better spent tithe?

MyGoldenMonkey commented on 15 Apr 2015

Thanks alot Khan Academy! My stats teachers shit and you just saved my life.

Satak commented on 02 Sep 2012

It's actually .2476 because the next digit after 5 is 7 and you roundup that.

AP Stat Ch 21 Video 2 Tests and Intervals.mp4

User Comments

xUnderTheGreyx commented on 05 Mar 2013

Thank you so much! I always watch your videos days before we get to these topics! Love you a lot! :)

20. Examples of Errors in Hypothesis Testing and Impact of Alpha

Crash Course in Statistical Error (Part II)

User Comments

smicha7 commented on 23 Nov 2010

At 1:54 and 3:43 the video hangs. Good examples but turn off the music :)