Hi Brandon, you are God sent and have been a blessing during my stats course, Its a shame that I am paying $4 500 for my stats course and I have not gotten anything, everything I have learnt, I have done from your videos. Thank you for feeling in the gap. Kudos for a job well done and keep being a blessing. Cheers
You are VERY welcome my friend. This is a labor of love for me, so it never really feels like work. The real kudos goes to YOU, for committing to learning wherever that may be. Keep up the good work and never stop learning! - BF
Love your introduction to this video! So positive and healthy for students to hear. Thanks for helping calm down anyone who is struggling and encouraging them that they are capable and can get through it.
Hallo. When finding confidence intervals we are usually given the sample size and that determines that margin of error. In this case we are reversing it. We are given a margin of error and then we are solving for the sample size that will produce the margin of error we chose at the beginning. For example a rectangle, we know that Area = L x W. If we know 2 of these, we can solve for the other. Same thing here. We are just solving for a different unknown.
Hi Brandon! I am from India. I love the way you explain all the concepts from scratch. Keep making such amazing videos on statistics. Thank you very much.
I have just started to listen to your videos for the first time and already, I am sold. First off, your intonation is nowhere near as monotonous and lifeless as my stat professors. You do a good job of briefly reviewing topics which are at the base of the new topics you mention. Thank you for your videos. I do have a question though. I will put the question in another post.
If only I'd seen this before I took my Open University Data Analysis exam! The interpretation and case-study format of these videos brings the subject to life. That's what makes it stick. Highly recommended if you want bite sized nuggets of Stats know-how. Rickster
These have been an excellent supplement to my course. You explain concepts logically and clearly, which increases my understanding of these concepts. I haven't taken stat almost 20 years so these have been a life saver!
brandon, thank you for discussing the basics in detail. i just realized how amazing stats could be. i really appreciate your teaching skills. you are so dedicated and patient . thanks and more power.godbless
I don't really comment on youtube much but I just watched two of your videos and they were extreeemeellyy helpful! I'm a PhD student and I needed to refresh myself on some statistics quickly for experimental design. Thanks a lot!
In the gas problem example no. 2, we are given sample S.D (s) not population S.D(sigma), so why you have used it as population S.D. Also when 'sigma' is not known we use t distribution (i.e t score) so why you used z score (upper and lower boundary) in the formula at 20:07.
Hi Brandon, I am very fortunate that I stumbled on your video 'Statistics' while searching to estimate a sample size. You made it interesting and simple. Wish you were my instructor in the class. Now I find it interesting.
I've been using your lectures by topic as supplemental lectures and instruction for my Stats I class. You've helped so much, in particular with linear correlation. Thanks so much! I really appreciated your playlists by topic and hope you will continue to expand these and maybe also add some more application problems. You're a gifted lecturer and teacher. Thank you for your contribution to higher education and equitable access to high-quality educational resources.
Brandon, again great job! This video answer most of the question I have with regard to how to select the sample size with 95% ci and specified MoE. You explanation is really more of self explanation, which is a great way for this kind of distance/online leaning. It's Really really really helpful. I can't stop sharing your video with my friends. Thank you so much Brandon.
Thank you for this, labored over my notes ( online masters course) for days, spent one day with these videos and I am now comfortable with these topics. Thanks again
Thanks a for for these videos. My teacher, Mr. Sanjay Sane recommended this video so I was watching it and it is a wonderful way of teaching I must mention. Your teaching is really wonderful. Many good wishes. :)
Confidence intervals for variances are much more involved and that is actually the next topic I will cover. You do not need to know how to find those to do any of the problems up to this video in the playlists. :)
Brandon, you are god! I think you have done a PROFESSIONAL work and it is a jewel for education purposes. I just wanted to ask something though. In 15:08 you interpret what is our conclusion. Is it also correct to say, interpretation: to have 95% of samples contain μ (miou) in the interval of plusminus 8 from Xbar of the sample, 78 must be our sample size? i try to grasp statistical philosophy and you help tremendously with your videos! Thank you.
Brandon. Thank you so much for your video, which is very clear and informative. I have two questions for you: 1. How should I go about choosing a margin of error to calculate the sample size? 2. If I plan to eventually do a regression analysis on my data and will be analyzing two variables, which variable's standard deviation should I use to calculate sample size? The larger of the two? Thanks.
Hello! First and foremost there is no "easy button" for this. It will take hours of doing practice problems in your book and/or notes; preferably in a study group if that is possible. Secondly, I have not covered a few of those topics...yet. The goal of my videos has been for people like yourself to recognize the basic pattern and method for working these and then apply them in your classwork. Practice, study groups, note cards, reviewing examples in your book. Just a few ideas. You can do it!
wow! this is the very similar question out in ASQ CQE exam practice paper and I was always wondering why only a portion of the equation used to get the sample size, n. Many thx for this vid!
Hi Carolyn, you can use fisher’s formula [Z2 (1-p)p/d2], where Z= alpha (or 1.96 at the 95% CI), p= probability, d= margin of error) or Cochran’s formula [Z2pq/e2] where Z alpha (or 1.96 at the 95% CI), p= probability, e= margin of error). So in your case, you want a 95% CI (Z= 1.96), p=.20 (20% of participants volunteer in such studies), thus q= 1-p, which is .80. You did not specify your margin of error (let’s say a tolerable margin of error is at 5%, thus we will use .05 in our calculation). Using Cochran’s formula= 1.962*.20*.80/.052 = 245 Your sample size is very small Carolyn :-/ Alternatively, if you want to stick to that sample of 14, you can do the reverse to find your margin of error, which I presume would be large and you will have a weak sampling design, you would not be able to generalize your results. So, let’s take Cochran’s formula again= Z2pq/e2 So, 14 (your sample size)= 1.962 *.20*.80/e2. Square root both sides to get e= 1.962 *.20*.80/3.74= 0.16 (margin of error)
Hi Brandon. You are awesome! How lucky are your student to have a such a wonderful instructor! Would you please add some i depth videos for different statistical models, as well? like mixed and random models.
Hi SouthpawGrammar, For your problem, If we can assume normally distributed data, I get an answer of 420 specimens. We don't know sigma in your population, so we have to estimate it from the range of 53 to 680, which is 627. Divide this by 6 (since 99.7% of normally distributed data falls within 3 standard deviations of either side of the mean) to get an estimate for sigma of 104.5. Our z score for 95% is 1.96, so the answer for a 10 PPB margin of error is ([(1.96)(104.5)]/10)^2, or 420 specimens rounded up.
Thanks Brandon, very helpful! Wondering if you can help guide me to an answer on this one. I want a 95% C.I. for the difference between the means of two populations (normal distributions) to within one standard deviation, and these populations also have the same standard deviation. Taking equal-size samples from each population, what is the minimum number of samples I would need to take from each sample? I'm not sure if there is enough information here, I'm thinking since I don't have the margin of error requested that I can't solve for n. Thanks!!
Brandon thank you so much for these wonderful video of yours.. I wanted to know if you have any video for confidence interval for population proportion ?
One question here is - when calculating sample size 'n', you have taken z alpha/2, that is z value of 95% confidence. My question is why don't we take t alpha/2 as we don't know population std deviation in the problem.
I am totally amaze with your videos :) Help me a lot in order to understand some fundamental stuff about statistics :) I am just hoping that you might also be able to make some videos related to time series analysis. It would be great :)
Hey, Thanks a ton for the video. You explained how to calculate sample size where pop std dev is known. I also see people using the below formula for sample calculation. Why is it different and what is the significance? n = [(z^2 * stddev^2)/E^2] / [1 + (z^2 * stddev^2)/E^2*N] (N = Population size)
Verbally: a thumb up Thanks a lot You have paved the way thro to a tasty stat. I have a question:- from does this confidence level come? On what base do I decide my confidence level?
Choosing a confidence level is a tradeoff between Type I and Type II error. That is why 95% is used most often since it offers a balance between the two error types. I have videos on Type I and Type II error if you need more help. Thanks!
Can you do a couple of videos pertaining the sample size calculations using Gy Pierre's sampling theory. I have browsed the entire youtube site and more and no one discusses, Ingamell's, Gy's, and Vismans sampling theorems.
Brandon, Thank you again for your videos. Others have expressed the valve better than I ever could. Question. In. this video we are estimating the population SD (especially when using the range formula). If estimating or unsure of the population SD, shouldn't we use the t value instead of the z value in the respective calculations? Thank you in advance.
Excellent video!! One question, what should I do to select a number (sample size) of surveys when qualitative data is involved? They are made to know customer perception in 6 different topics.
Hi Brandon. One quick question. In all your examples (until this point I´ve watched all videos in order) you are "assuming" that all your data are normally distributed, right? Otherwise the analysis were not possible. Shouldn´t you start by doing normality tests, before applying these analysis? Thanks
In theory yes, but my content is geared towards students who are learning the content and need a little help. In those environments normality is something that is assumed. Very few undergrad (or equivalent) courses go into that much detail.
Hello, for Example 4, using the Z distribution table, I am really confused as to why you chose, a Z score of 2.576 for 99% instead of 2.33 and 1.645 for 90% instead of 1.289? By the way, your videos are a blessing! Thank you!
Brandon I desperately need your advice. I am being tested this coming Saturday for Confidence Intervals sigma known, sigma unknown, sample size, confidence intervals for proportions and finding confidence interval for a variance and standard deviation. What do you suggest I do in terms of studying for these with your videos? Thank you.
Hi ! Thanks for the video..I could understand the difference between the margin of error and confidence interval. I can infer that the sample size doesn't depend on the size of population size i.e. for 2 desks - a and b producing 5000 policies and 2000 policies the sample size would be same for a given margin of error and confidence interval ?
Brandon. When calculating the 'desired' sample size, why do you not substitute the t-values into the equation to solve for n, when you don't know the population standard deviation. I understand that CI with same sample size will have same MOE for given population standard deviation. And, that when you aren't given the population standard deviation and have to estimate using sample standard deviation (s) that samples of same size DO NOT have same MOE. Is this the reason that you have to use Z-values when estimating n (not given population standard deviation)?
Hi thanks a lot for the videos, you are a lifesaver. I have a doubt in question 2, if it says the sample standard deviation is 0.05, then doesn't it mean s and not sigma? so wouldn't that involve using t values and not z ? how to know the difference?
A problem I am working on mentions. The types of widgets used typically fall into the second classification on the company’s pricing scale. This classification places an upper limit of 850 pounds on the mean. There is no mention of a lower limit. In determining E am I to assume 2 classifications above the mean and 2 below for a total of 4? Or am I to work with the upper limit value?
Hi Brandon. Thank you so much for the video, it was very helpful! I was just wondering it the formula used in example one can also be used if the margin of error and standard deviations are percentages rather than set values? Would you convert them to decimals? So if the the question said the margin of error was 8% and the standard deviation was 36%, would we then use 0.8 and 0.36 instead?
Brandon, quick question. When I am estimating sample size, I am using Standard Deviation from a data set. Should this data set be normally distributed? What if the data in not normally distributed. Thank you!
I'm trying to solve this problem, and just couldn't get the right answer. The Bureau of Alcohol, Tobacco, and Firearms (BATF) has been concerned about lead levels in California wines. In a previous testing of wine specimens, lead levels ranging from 53 to 680 parts per billion were recorded. How many wine specimens should be tested if the BATF wishes to estimate the true mean lead level for California wines to within 10 parts per billion with 95% confidence? Thanks
sir,,i came across with your videos,,i am looking for a sampling method applicable in my study..my study is about the use of ATM Card in our school.. the problem is not all students have an ATM..how can I get a sample size knowing that i don't know how many students are using an atm?..the only thing I know is the population of the student body..help me pls.thanks sir
Hey Brandon. I am a student doing my thesis. As a part of my thesis, i need to test electrical componants. How can i decide number of those Componants(samples) needed to test to get a confidence level of 95%? Thank you.
Hi Brandon, you are God sent and have been a blessing during my stats course, Its a shame that I am paying $4 500 for my stats course and I have not gotten anything, everything I have learnt, I have done from your videos. Thank you for feeling in the gap. Kudos for a job well done and keep being a blessing. Cheers
You are VERY welcome my friend. This is a labor of love for me, so it never really feels like work. The real kudos goes to YOU, for committing to learning wherever that may be. Keep up the good work and never stop learning! - BF
Love your introduction to this video! So positive and healthy for students to hear. Thanks for helping calm down anyone who is struggling and encouraging them that they are capable and can get through it.
Hallo. When finding confidence intervals we are usually given the sample size and that determines that margin of error. In this case we are reversing it. We are given a margin of error and then we are solving for the sample size that will produce the margin of error we chose at the beginning. For example a rectangle, we know that Area = L x W. If we know 2 of these, we can solve for the other. Same thing here. We are just solving for a different unknown.
Hi Brandon! I am from India. I love the way you explain all the concepts from scratch. Keep making such amazing videos on statistics. Thank you very much.
This video saved my life. Well, at least my GPA.
Thank you so much. I looked at so many videos before yours. But you helped me see the math.
I have just started to listen to your videos for the first time and already, I am sold. First off, your intonation is nowhere near as monotonous and lifeless as my stat professors. You do a good job of briefly reviewing topics which are at the base of the new topics you mention. Thank you for your videos. I do have a question though. I will put the question in another post.
If only I'd seen this before I took my Open University Data Analysis exam! The interpretation and case-study format of these videos brings the subject to life. That's what makes it stick. Highly recommended if you want bite sized nuggets of Stats know-how.
Rickster
These have been an excellent supplement to my course. You explain concepts logically and clearly, which increases my understanding of these concepts. I haven't taken stat almost 20 years so these have been a life saver!
Thanks so much! Glad you find them helpful. Keep on learning!
brandon, thank you for discussing the basics in detail. i just realized how amazing stats could be. i really appreciate your teaching skills. you are so dedicated and patient . thanks and more power.godbless
I don't really comment on youtube much but I just watched two of your videos and they were extreeemeellyy helpful! I'm a PhD student and I needed to refresh myself on some statistics quickly for experimental design. Thanks a lot!
In the gas problem example no. 2, we are given sample S.D (s) not population S.D(sigma), so why you have used it as population S.D. Also when 'sigma' is not known we use t distribution (i.e t score) so why you used z score (upper and lower boundary) in the formula at 20:07.
I had the same question. Did you find the answer?
Same question
Hi Brandon, I am very fortunate that I stumbled on your video 'Statistics' while searching to estimate a sample size. You made it interesting and simple. Wish you were my instructor in the class. Now I find it interesting.
' just started watching your vids. They are super helpful in expaiining what''s going on. Thanks and keep up the great work!
I've been using your lectures by topic as supplemental lectures and instruction for my Stats I class. You've helped so much, in particular with linear correlation. Thanks so much! I really appreciated your playlists by topic and hope you will continue to expand these and maybe also add some more application problems. You're a gifted lecturer and teacher. Thank you for your contribution to higher education and equitable access to high-quality educational resources.
Brandon, again great job! This video answer most of the question I have with regard to how to select the sample size with 95% ci and specified MoE. You explanation is really more of self explanation, which is a great way for this kind of distance/online leaning. It's Really really really helpful. I can't stop sharing your video with my friends. Thank you so much Brandon.
Thank you for this, labored over my notes ( online masters course) for days, spent one day with these videos and I am now comfortable with these topics. Thanks again
Absolutely perfect explanation for a question that has been taunting me for weeks, thank you very much.
Excellent explanation focusing on the concept and with enough of examples to support it. Thank you so much for the effort.
Excellent video. I wish I had this instructor in my graduate class.
Thanks a for for these videos. My teacher, Mr. Sanjay Sane recommended this video so I was watching it and it is a wonderful way of teaching I must mention. Your teaching is really wonderful. Many good wishes. :)
I am sorry I didn't see your sessions earlier. I was so lost. It made things so much clearer. Well done.
Thanks
You explain with simple way. Loved it
Confidence intervals for variances are much more involved and that is actually the next topic I will cover. You do not need to know how to find those to do any of the problems up to this video in the playlists. :)
Brandon, you are god! I think you have done a PROFESSIONAL work and it is a jewel for education purposes. I just wanted to ask something though. In 15:08 you interpret what is our conclusion. Is it also correct to say, interpretation: to have 95% of samples contain μ (miou) in the interval of plusminus 8 from Xbar of the sample, 78 must be our sample size? i try to grasp statistical philosophy and you help tremendously with your videos! Thank you.
Just wanted to let you know you literally saved my life today. :)
Thank You Brandon, So informative, slow in explanation and super easy to follow. good work, God bless.
Nobel Prize for Best Educator!!!!
Brandon.
Thank you so much for your video, which is very clear and informative. I have two questions for you:
1. How should I go about choosing a margin of error to calculate the sample size?
2. If I plan to eventually do a regression analysis on my data and will be analyzing two variables, which variable's standard deviation should I use to calculate sample size? The larger of the two?
Thanks.
Hello! First and foremost there is no "easy button" for this. It will take hours of doing practice problems in your book and/or notes; preferably in a study group if that is possible. Secondly, I have not covered a few of those topics...yet. The goal of my videos has been for people like yourself to recognize the basic pattern and method for working these and then apply them in your classwork. Practice, study groups, note cards, reviewing examples in your book. Just a few ideas. You can do it!
wow! this is the very similar question out in ASQ CQE exam practice paper and I was always wondering why only a portion of the equation used to get the sample size, n. Many thx for this vid!
Hi Carolyn, you can use fisher’s formula [Z2 (1-p)p/d2], where Z= alpha (or 1.96 at the 95% CI), p= probability, d= margin of error) or Cochran’s formula [Z2pq/e2] where Z alpha (or 1.96 at the 95% CI), p= probability, e= margin of error).
So in your case, you want a 95% CI (Z= 1.96), p=.20 (20% of participants volunteer in such studies), thus q= 1-p, which is .80. You did not specify your margin of error (let’s say a tolerable margin of error is at 5%, thus we will use .05 in our calculation).
Using Cochran’s formula= 1.962*.20*.80/.052 = 245
Your sample size is very small Carolyn :-/
Alternatively, if you want to stick to that sample of 14, you can do the reverse to find your margin of error, which I presume would be large and you will have a weak sampling design, you would not be able to generalize your results.
So, let’s take Cochran’s formula again= Z2pq/e2
So, 14 (your sample size)= 1.962 *.20*.80/e2. Square root both sides to get e= 1.962 *.20*.80/3.74= 0.16 (margin of error)
Hi Brandon. You are awesome! How lucky are your student to have a such a wonderful instructor! Would you please add some i depth videos for different statistical models, as well? like mixed and random models.
Thank you very much! You just helped me through my Operations Management presentation. More power bro!
Wow, can't believe I took so long to find you. You make it all sound so easy! Thank you!
I was already getting so mad I wasn't understanding this, this video saved me lol
Thanks for adding such an amazing presentation really liked it
Gives a good understanding of factors interplay with sample size.
You are really good. I can understand it easily. Many thanks
thnkxx u started with imparting some motivation...
came across your video today... good one!
Great Videos and a lot of motivation throughout a very fascinating topic. Thanks for that!
Hi SouthpawGrammar,
For your problem, If we can assume normally distributed data, I get an answer of 420 specimens. We don't know sigma in your population, so we have to estimate it from the range of 53 to 680, which is 627. Divide this by 6 (since 99.7% of normally distributed data falls within 3 standard deviations of either side of the mean) to get an estimate for sigma of 104.5. Our z score for 95% is 1.96, so the answer for a 10 PPB margin of error is ([(1.96)(104.5)]/10)^2, or 420 specimens rounded up.
Thanks Brandon, very helpful! Wondering if you can help guide me to an answer on this one. I want a 95% C.I. for the difference between the means of two populations (normal distributions) to within one standard deviation, and these populations also have the same standard deviation. Taking equal-size samples from each population, what is the minimum number of samples I would need to take from each sample? I'm not sure if there is enough information here, I'm thinking since I don't have the margin of error requested that I can't solve for n. Thanks!!
Hello Brandon, you teach really well
I purposely listen to the intro because I want Brandon to tell me I'm smart even though I'm not. :(
Beautiful video, Brandon. Thanks for posting it.
Excellent video. It was very helpful. Thank you very much.
Great job, topic was presented clearly, concisely, and logically. Thank you.
Thank you very much for your videos.
God absolutely blesses you!
Vangchue Lee, from Kobe University, Japan.
Brandon thank you so much for these wonderful video of yours..
I wanted to know if you have any video for confidence interval for population proportion ?
One question here is - when calculating sample size 'n', you have taken z alpha/2, that is z value of 95% confidence. My question is why don't we take t alpha/2 as we don't know population std deviation in the problem.
To use t distribution table, we should know the sample size.. So this problem can not be solved.
this is a super presentation. it help me a lot believe me.
i learnt a lot from this video
Thankyou for the video,its absolutely clear
I am totally amaze with your videos :)
Help me a lot in order to understand some fundamental stuff about statistics :)
I am just hoping that you might also be able to make some videos related to time series analysis. It would be great :)
Thank you for the video! May I know how you got 4 in the denominator of the planning value? :)
Yes please, I was wondering that. Why range/4?
thanks a lot .. you are god to me ! Love from India
God bless you, Brandon.
in the 2nd example , sample standard deviation is taken as sigma instead of population standard deviation
Awesome video! Thank you!
Hey, Thanks a ton for the video. You explained how to calculate sample size where pop std dev is known. I also see people using the below formula for sample calculation. Why is it different and what is the significance?
n = [(z^2 * stddev^2)/E^2] / [1 + (z^2 * stddev^2)/E^2*N]
(N = Population size)
Verbally: a thumb up
Thanks a lot
You have paved the way thro to a tasty stat.
I have a question:- from does this confidence level come? On what base do I decide my confidence level?
Choosing a confidence level is a tradeoff between Type I and Type II error. That is why 95% is used most often since it offers a balance between the two error types. I have videos on Type I and Type II error if you need more help. Thanks!
Can you do a couple of videos pertaining the sample size calculations using Gy Pierre's sampling theory. I have browsed the entire youtube site and more and no one discusses, Ingamell's, Gy's, and Vismans sampling theorems.
Brandon, Thank you again for your videos. Others have expressed the valve better than I ever could. Question. In. this video we are estimating the population SD (especially when using the range formula). If estimating or unsure of the population SD, shouldn't we use the t value instead of the z value in the respective calculations? Thank you in advance.
thanks a lot Brandon ,so helpful
Professor do you have a cd kit with all your videos in order of learning?
Excellent video!! One question, what should I do to select a number (sample size) of surveys when qualitative data is involved? They are made to know customer perception in 6 different topics.
Bless u Brandon. U've made my day
fantastic discussion
Thanks Brandon It's really helpful. I just want to ask you if there are ranges of accepted margin of error that scholars use frequently. Thanks a lot!
Can't make it better. Thank you!
Hi Brandon. One quick question. In all your examples (until this point I´ve watched all videos in order) you are "assuming" that all your data are normally distributed, right? Otherwise the analysis were not possible. Shouldn´t you start by doing normality tests, before applying these analysis? Thanks
In theory yes, but my content is geared towards students who are learning the content and need a little help. In those environments normality is something that is assumed. Very few undergrad (or equivalent) courses go into that much detail.
Hello, for Example 4, using the Z distribution table, I am really confused as to why you chose, a Z score of 2.576 for 99% instead of 2.33 and 1.645 for 90% instead of 1.289? By the way, your videos are a blessing! Thank you!
Brandon I desperately need your advice. I am being tested this coming Saturday for Confidence Intervals sigma known, sigma unknown, sample size, confidence intervals for proportions and finding confidence interval for a variance and standard deviation. What do you suggest I do in terms of studying for these with your videos? Thank you.
Brandon. Little confused with this video. Do you have a video for confidence intervals for variances and standard deviations.
better than my lecturer haha thank you so much
Hi Brandon. Example 4: Confident retirement: Why are the n =153, n = 217, n = 374 Instead of n = 152, n = 216, n = 373 ? df?
Thank you so much, helped me a lot!
Thank you so much, your video was great. How can I see all the videos that you have posted?
Thank you for your videos!!!
I think it would be great if at minute 31:27 you explain how you got the numbers that are in red.
Hello! I do that at the 28:00 mark where we have the three confidence intervals listed and we walk trough each one. Hope that helps!
Hi ! Thanks for the video..I could understand the difference between the margin of error and confidence interval. I can infer that the sample size doesn't depend on the size of population size i.e. for 2 desks - a and b producing 5000 policies and 2000 policies the sample size would be same for a given margin of error and confidence interval ?
Brandon. When calculating the 'desired' sample size, why do you not substitute the t-values into the equation to solve for n, when you don't know the population standard deviation. I understand that CI with same sample size will have same MOE for given population standard deviation. And, that when you aren't given the population standard deviation and have to estimate using sample standard deviation (s) that samples of same size DO NOT have same MOE. Is this the reason that you have to use Z-values when estimating n (not given population standard deviation)?
very interest for learning
Hi thanks a lot for the videos, you are a lifesaver. I have a doubt in question 2, if it says the sample standard deviation is 0.05, then doesn't it mean s and not sigma? so wouldn't that involve using t values and not z ? how to know the difference?
A problem I am working on mentions. The types of widgets used typically fall into the second classification on the company’s pricing scale. This classification places an upper limit of 850 pounds on the mean. There is no mention of a lower limit. In determining E am I to assume 2 classifications above the mean and 2 below for a total of 4? Or am I to work with the upper limit value?
Great Video
Thank you
Hi Brandon. Thank you so much for the video, it was very helpful! I was just wondering it the formula used in example one can also be used if the margin of error and standard deviations are percentages rather than set values? Would you convert them to decimals? So if the the question said the margin of error was 8% and the standard deviation was 36%, would we then use 0.8 and 0.36 instead?
Thank you Brandon
In Gas example problem, why was the sample standard deviation used as population standard deviation?
You are the best! Thank u
Great job!!
Hi. May I know the author of the formula of the sample calculation that you used? I find it more researcher friendly. Thank you.
Brandon, quick question. When I am estimating sample size, I am using Standard Deviation from a data set. Should this data set be normally distributed? What if the data in not normally distributed. Thank you!
I'm trying to solve this problem, and just couldn't get the right answer.
The Bureau of Alcohol, Tobacco, and Firearms (BATF) has been concerned about lead levels in California wines. In a previous testing of wine specimens, lead levels ranging from 53 to 680 parts per billion were recorded. How many wine specimens should be tested if the BATF wishes to estimate the true mean lead level for California wines to within 10 parts per billion with 95% confidence?
Thanks
Why aren’t we using t distribution since we are estimating the pop standard deviation?
sir,,i came across with your videos,,i am looking for a sampling method applicable in my study..my study is about the use of ATM Card in our school.. the problem is not all students have an ATM..how can I get a sample size knowing that i don't know how many students are using an atm?..the only thing I know is the population of the student body..help me pls.thanks sir
Simply brilliant, thanks a ton.
In Example 2, should E be a % of the average ($3.55)? Thus, if we are looking for an margin of error of $0.03, then should E = 0.03/3.55 =0.845?
Hey Brandon. I am a student doing my thesis. As a part of my thesis, i need to test electrical componants. How can i decide number of those Componants(samples) needed to test to get a confidence level of 95%? Thank you.