Hypothesis Testing - Solving Problems With Proportions

This statistics video tutorial explains how to solve hypothesis testing problems with proportions. It explains how to calculate the sample proportion and the z-test statistic and how to compare that with the critical values in order to determine whether or not if you should accept or reject the null hypothesis at a specified confidence level or significance level.
Confidence Intervals & Margin of Error:
• How To Find The Z Scor...
Find The Z-Score Given Confidence Interval:
• How To Find The Z Scor...
Student's T-Distribution:
• Student's T Distributi...
Hypothesis Testing - Null & Alternative:
• Hypothesis Testing - N...
Type I and Type II Errors:
• How To Identify Type I...
One Tailed and Two Tailed Tests:
• One Tailed and Two Tai...
_______________________________________
Test Static For Means & Pop Proportions:
• Test Statistic For Mea...
Hypothesis Testing Problems:
• Hypothesis Testing Pro...
P-Value Method - Hypothesis Testing:
• P-Value Method For Hyp...
Hypothesis Testing - Proportions:
• Hypothesis Testing - S...
Difference of 2 Means - H Testing:
• Hypothesis Testing - D...
Cohen's D - Pooled Standard Deviation:
• How to find Cohen's D ...
Hypothesis Testing - 2 Proportions:
• Hypothesis Testing Wit...
__________________________________
Matched or Paired Samples t-Test:
• Matched or Paired Samp...
Chi Square Test:
• Chi Square Test
Chi Square Distribution Test:
• Chi Square Distributio...
Test of Independence - Chi Square Dist:
• Test of Independence U...
Final Exams and Video Playlists:
www.video-tutor.net/
Full-Length Videos and Worksheets:
/ collections

[@TheOrganicChemistryTutor]

Next Video: kzbin.info/www/bejne/i5S9qKywrLyIbdE Final Exams and Video Playlists: www.video-tutor.net/

[@Untouched77]

why do you use the mu symbol instead of p when it is a proportion?

[@Danieljordan2]

Awesome video. Nevertheless, the tutor made a mistake in the second problem: he shoudn't divide the alpha area because this is an one tail problem. The z value is 1.285 aprox.

[@michaeladdo7226]

exactly

[@Eco-zaki]

Well said! bro. At the end of the day the tutor is human❤

[@andrewshaba8677]

Yeah that's right

[@zsnowboardern9332]

i was thinking the same i thought i was hallucinating and i warched over and over again. the video basically showed that one and two tail tests are the same lol😮. hope there is a rectification somewhere or at least correct the video

[@zsnowboardern9332]

@@Eco-zakiyes human do make mistakes but without acknowledging the mistakes is a problem

[@ErtOzk]

Last question is not two side, you made an error with dividing the alpha. Passed. Thanks :D

[@IArkProject]

Another tip for anyone who reads these comments is that when finding the p-value for 0.90, it similarly is between two different values for 1.28 and 1.29. Rather than taking the average of both of these, which would still be inaccurate, you can set up a proportion. Set up a proportion where 0.89973 is to 1.28, as 0.90 is to x. x, of course, will be the value you are looking for. I set this up as .89973 over -1.28 = .90 over x. In the end you get 1.280384, and so on for a few more decimal spaces. So it is closer to 1.28. I hope that was helpful. And thank you for these videos, this is extremely helpful, and yes to err is human - all the best!

[@sirasidibe95]

paying over $2k for a stat class and almost failing but I'm actually learning and understand for FREE on KZbin.TYSM! These vids literally are saving me from failing. Ive watched other stat videos but this does it I swear!

[@paulchen355]

I am having the same experience

[@insaf_]

I know a couple people mentioned it already but in case you missed the notifications, its a one tail test. so Zc= 1.28

[@missnurse5295]

THANK YOU!! For taking your time to explain each step of the problem instead of throwing in a lot of fancy terms and a lot of blah blah l searched over 20 videos today for a simple explanation and couldn't find it until I seen your video.... Thanks a million!!

[@Jennyofthesky]

I am so greatful for these videos!!!!!!!! Thank you so much man. I love the way you carefully explain EVERYTHING you do, and I love how you always have a PERSON who is giving out H(1). It is so much clearer to determine H(1) when you say “The market manager does not believe this is true”

[@huffypuff6707]

Thank you SO much!! The amount of relief I feel when I see you have a video covering the topic I am stuck on... You've been the goat for years

[@talkingkangaroo4934]

For Q2, why do you use Z of 1.645? Isn't this a one-tailed test where the 10% rejection region is all on the right tail?

[@r.n.3709]

@Tiffany Dang yea coz alpha is just 0.1 and area to the right is 0.9

[@ErtOzk]

Answer is still true but yep its one side and we don’t need to divide alpha value.

[@estherkang4593]

@Tiffany DangSame, someone explain this!

[@marijatosic217]

Exactly. I think he just got a bit confused and didn't review it. There is a video where he explains this correctly. :)

[@masterlaughter4924]

I agree!

[@nurulbalqis2647]

In the 2nd question, the alpha does not have to be divided into 2 right? the whole alpha is on the right tailed side

[@mindoII]

you doing more than my teachers u deserve my tuition fee 🤧 edit: got a good grade because of u

[@danielmorrison5293]

I second that!!!

[@abdoali5271]

I also have an exam Wish to say the same!

[@sylviogiraldes]

Great example. Objective and complete. Thanks for sharing.

[@HappinessYata]

This is so well explained. I was struggling a lot with that.

[@hananyisehak6693]

Perfectly explained, Thank you!

[@richardl.4374]

Why did you use 1.645 in the second question when our alpha is only 1 tailed? Isnt the critical value supposed to be 1.285

[@valendioah1350]

all of your video are easy to understand, with a good example and explanation, its very help me to study, thxx

[@okellofrancis2218]

I'm grateful for the video it really helped me to understand hypothesis testing

[@COD10410]

In the second question, shouldn't the critical z-value be 1.282? Because we don't have to divide the alpha

[@richardl.4374]

Same question in mind

[@Memesfromthestreets]

He did it wrong

[@y.se-won6711]

You're the absolute best, thank you so much!

[@iremiposiajayi122]

To get the standard error, why do we use the proportion in the null rather than the sample proportion? Why can’t the sample proportion estimate the population proportion (like with means)?

[@URANUSguys]

Im taking a re-test for the statistics course this sem, even if i pass it i can only have a D grade, if only i had seen these videos on time i could have gotten atleast a b+. Last sem i barely passed the prob course, which you have also covered in this playlist. I barely passed the course!!😭. i wish i had seen these videos before!!

[@Ankit-hs9nb]

Thank you so much!

[@omaimaelshamy6680]

Thank you so so so much , I have understood this lesson very well 👍👍👍👍👍👍😎

[@undergraduateMath]

Thanks alot OCT.. but how can we find the critical point without using tables?

[@alisyed156]

so much helpful thank you so much man great work

[@bars1628]

thanks for video series

[@PicoOp]

your videos are really great

[@dimitriysevryukov1629]

Thank you very much !

[@primrosemwendwa801]

Thank you!

[@devkumar9889]

Thank you so much

[@thatomofolo452]

Thank You Mr J 🙋👍👍

[@lol_gmz]

Thank you so much😁

[@lsacy8347]

why are we using for Q1 the variance of h0 instead of the sample proportion?

[@isabelgoh566]

Why need to draw a two-tail test when it one tail? Then alpha 0.05 + 0.9 in order to get Z score?

[@karlbautista1960]

isnt it suppose to be left tailed test coz it says less than 60%? and if more than is right tailed test?

[@russelacierto3108]

Thank you so much sirrr

[@malaik9]

I understand the logic of you distributing the 10% into 2 but are there conditions one can use the whole 10% significance interval for a 1 tail t test or is it just always that one must divide whatever SL they get

[@RoseLK]

Thank you so much

[@Dark_matician]

I have an exam on this today. I hope this helps me.

[@srachellem]

thank you

[@nonono8368]

Why does the table he's using cantains different values than the ones on the web?

[@shagurtham]

can you please perform one example with t-test when n < 30

[@borisszeto2381]

Thank you for your teaching video, you are the best professor on youtube who saved my GPA

[@claudiamesaaparicio8517]

the critical value is 1.28 because is one side test, we do not need to devide alpha. Two side test will be 1.645

[@josemariaalipio449]

1.28 critical value is correct. It is indeed a one-sided test.

[@travelwithsamuel9220]

how did you get that 0.025?

[@aneelailyas5973]

this is one tail, upper tail, and for Z alpha (10%) for one side value will be 1.28 is not it?

[@dericktembo1466]

Awesome explanation, but shouldn’t the alternative hypothesis be # 0.60? Because the sales manager didn’t specify if the percentage is greater or less.

[@m-cs22]

The alternative hypothesis is always inverse of the null hypothesis. so if the null hypothesis states "less than", the alternative hypothesis is "more than"

[@user-rc7ik7xj1p]

brother, the null hypothesis should = 60 in Q2 because null hypothesis is an equality not an interval + it's a lower tail test meaning that p

[@viktor2978]

Are we always using the z value because n > 30? Or is it always the z value because it's different for proportions?

[@shadowfire04]

because n > 30, from what i understand. you would use t-distribution if n < 30 and the population std. dev was not known.

[@HellenofTroy897]

Why did u split the the Alpha in question 2?

[@Coat8678]

I need to know why I’m got 1.28 for critical z

[@underageguru]

Q2 you made an error. It’s a 2 tailed test so the critical region on the right side should include 10% area

[@pran8466]

You just save a life

[@cypriankibet1205]

Take down question 2 part where you divide by 2 its alpha, no need to divide by 2 when its a one tailed test. Otherwise, am enjoying the series Sir. Hours to my exams hahaha!!!

[@cartoon.cartoon6757]

The best

[@HeckaS]

Like others have mentioned in question 2: Hnull should be an equality.

[@carlotabarrosferreira3117]

meu amigo, amo te com tudo o que tenho

[@DeepakYadav-dk1dh]

I'm confused...i saw one of your video for z test...the Z test formula there was different than what it is here....there Z = ((sample mean - population mean )/ (population standard deviation * root of sample size))

[@argha-qi5hf]

There are two formulas. Both are related and can be derived from the other.

[@user-vu9uw5jy1r]

Amazing Video. However, I think you made a mistake in second problem. As it is one tail test(right tail), hence you should not divide alpha(significance level region) region and z critical value should be 1.282, not 1.645.The couple of Guys have already mentioned it and i guess Correction should be made in video.

[@aeriiful]

i can't do this anymore

[@CRESCENTP-cm8nj]

I woke up at 12:00AM today

[@SkyrimInMyPants]

HOW IS EVERYONE GETTING 1.28? PLEASE EXPLAIN!!!

[@AliAkbar-bv7zp]

please check the previous video of this playlist 😃

[@CRESCENTP-cm8nj]

3:00

[@CRESCENTP-cm8nj]

12:00

[@subhamlaha7943]

Sir , With due respect , the video is awesome , but in question it was given 10 % significance level , (i.e. 90 % Confidence interval).However , you have solved using 95 % Confidence interval. Please rectify it if possible. Otherwise , the video is awesome.

[@hanifayyash2336]

i'm glad someone noticed this too

[@moindalvs]

@2:19 I was it was high on potenuse

[@indatawetrust101]

Why is the first problem not solved using binomial distribution?

[@indatawetrust101]

Normal approximation for a binomial distribution just clicked. :D

[@user-sd5dq8wb4j]

Why did the marketing manager go to town XYZ with a survey? Because he wanted to know if everyone there was "cell-fish" with their phones! (a) Null hypothesis: The percentage of cell phone owners is 70%. Alternative hypothesis: The percentage of cell phone owners is different from 70%. (b) Let's crunch the numbers and see if he's about to make a "call" that changes the marketing game!

[@martinavila7673]

I love us much

[@emmettdonnelly4260]

LEGEND

[@m-cs22]

you made a mistake for Q2. Shouldn't divide the alpha value, since it's a right-tailed test

[@bnnybn]

Why is proportion represented by mu intead of p? Time-stamp: 1min

[@undergraduateMath]

I guess he's just used to the mu symbol.. lol.

[@bnnybn]

@@undergraduateMath No, I think he is using these symbols incorrectly. Mean vs proportion - continuous variables vs proportions (can also be found in forms of binary variables represented by 0s and 1s).

[@lethalllex]

none of my assignments for hypothesis training are worded like this

[@suptydewan10]

67

[@lanavi732]

Since it's a one tailed test, why did we have to divide the alpha by 2? Edit: okay glad to see I'm not alone in the comments hahaha

[@andredejager3637]

2nd half of video is wrong, please edit or remove, thanks

[@hetavvakani7025]

I guess you made a mistake while calculating the z-value in the second problem. It should be 1.29

[@jaggyd2862]

8:13 Dont mind me

[@akirmarmay3142]

1 minus 0.60 is 40?!! what

[@sohamdata]

Significance Level = 100 - Confidence Level In the question, it's the S.L. that's given that's why, he has used C.L. = 90 % also, when he did Ar to the left = 0.90 + 0.05, it doesn't mean that alpha is halved. 0.90+0.05 = 0.95 is the same as the usual formula: Ar to the left = (1+C.L.)/2 = 0.95 just visualize (1/2) + (C.L./2) on the normal dist. (50% of the area + 50% of C.L. is needed to get the area left to the right bound)

[@NaifAlqahtani]

this doesnt make sense. The C.V should be 1.28 not 1.645. The shaded area for 1.645 (the rejection region) has an area of 0.05 or a S.L of 5%. The question wants a S.L of 10% which means the shaded area (rejection region) should be 0.1. using the table the C.V that give area to the right = 0.1 is C.V appx. = 1.28. NOT 1.645

[@abdulazizalharbi8066]

Thank you so much