Thanks for the videos. But just a quick note: you should have mentioned that we use the P bar when we do not know the standard deviation and the mean for the whole population, which is the university's students. Thanks again for your helpful videos :)
Pls, explain the standard eror = 0.0922 at 5:08 where it come from, what formula did you use? I used formula √p(1-p)/√n but the result did not fit with your calculation.
Hi Brandon, your video was very helpful. But I am still confused about how the idea of proportion is related to the binomial distribution. Specifically, how population proportion, p, and the probability, p (from binomial distribution) are related? Are these two the same?
the only issue about this topic is using the term "sample" and "sample size". You can use the term subgroup for sample while each subgroup has a sample size.
Hey dear Mr. Foltz, It would be nice if you make videos about Partei and Semi partei correlation and also about Phi coefficients, Y coefficient ( gamma) , discordant und concordance for ordinal data correlation . Thank you so much. You are the best @
11:49 This is the part where I'm confused. A few videos ago you mentioned the "Standard err of the mean" is inversely proportional to the sample size, so why is "Std Err Mean" decreasing when sample size stays constant here?
Ok never mind I think I get it now. There's another layer of sampling being introduced, so collecting more samples in the 1st layer is equivalent to increasing the sample size in layer 2: 1st layer: Sampling distribution of the sample proportion of size n=30 (with 25, 100, ... samples). Standard error only dependant on sample size, not number of samples. 2nd layer: Sampling distribution of the mean in layer 1. Our n now is the 25, 100, .. samples in layer 1 so it now has an impact on Standard Error of Mean of this layer.
Greetings. Have you figured out the solution to this question? I have the same one. Using the standard deviation formula, which Mr. Brandon introduced in Statistics PL03 - Descriptive Statistics II.
Regarding the software calculations it seems that the Standard Error of mean (0.0144) is calculated using n as 25 rather than 30. This would mean that n is the number of samples rather than sample size. Please clarify which is correct. Thanks
In 13:59, why are you comparing the Mean to the Theoretically calculated Std Error? Shouldn't the Std Deviation (generated by the software) be compared with the theoretical value?
Hi Brandon, At 11:49, you mentioned the number of samples affects the standard err mean, from 0.018 down to 0.0009014 if increase number of samples from 25 to 10000. I am confused how to calculate this standard err mean? you mentioned that is the standard error of the mean of all sample proportions. But what is the equation for that? Also, if number of samples does not affect the standard error of the sample proportion, we take 25 samples n = 30. Is that the same as 1 sample with n = 750? (750 = 25 x 30). In that case, we took many samples, which is the same thing as increase sample size? Thanks.
I had the same question and was looking at the comment section if anyone has already asked. Thanks for posting! From what I recall from the past videos is that std_error of the mean is the std deviation of the sampling distribution. So essentially, if we're calculating the standard deviation of the distribution of the sample proportions for each of the 25 (or call it the standard error), shouldn't the standard error and standard error of the mean be the same? I think I'm kind of in the weeds here. @Brandon, we really appreciate if you could help us here! Thanks very much in advance.
Hi Xi, so essentially, the standard error that we are calculating is synonymous to the sampling standard deviation (don't confuse with the std error of the mean). For a sample of 25 proportion, we'd calculate standard error as sqrt((x - mu)^2 / (n-1)). So for a sample size of 25, this will give 0.0920346 (sample standard deviation of 0.0920346). Now you should also assume that in the absence of population standard deviation, sample standard deviation provides a reasonable estimate. So, we'll fit in this sample std deviation in the formula (sigma_x_bar = s/sqrt(n)) to calculate std error of the mean. So for a sample of 25, plugging in the values, the std error of the mean (this is the std deviation of the sampling distribution) will be 0.0920346/sqrt(25), which gives 0.018. Now you can do the same thing for the rest of the sample sizes (25 to 100 to 1000) and you'll see that the std error of the mean reduces from 0.018 down to 0.0009014. Hope this helps! @Brandon, please correct me if I'm wrong, or if I'm missing something.
hi, Brandon flotz I saw your video. I want to learn on Random walk, hitting time, mean first passage time. any members can help. I want to learn it. please help me i will wait
Hi! Thanks for watching. There are tests for the equality of multiple proportions (p1 = p2 = p3) but that would be a slightly different topic than this. Hope to make a video about that sometime soon.
@@BrandonFoltz yes..thank you so much Brandon..I would appreciate, you may do this statistics on R or Some tool like SAS , SPSS by taking real-time problem. Again thank a lot
![BOLIVIA vs. COLOMBIA [1-0] | RESUMEN | ELIMINATORIAS SUDAMERICANAS | FECHA 9](http://i.ytimg.com/vi/zg15uVWjuf4/mqdefault.jpg)
Your the Best Brandon!! Im just wasting my $$$$$ at the university.
Thanks for the videos. But just a quick note: you should have mentioned that we use the P bar when we do not know the standard deviation and the mean for the whole population, which is the university's students.
Thanks again for your helpful videos :)
Hey Brandon, You are just awesome! kudos!!!!!
Pls, explain the standard eror = 0.0922 at 5:08 where it come from, what formula did you use? I used formula √p(1-p)/√n but the result did not fit with your calculation.
awesome video! thank you!
Hi Brandon, your video was very helpful.
But I am still confused about how the idea of proportion is related to the binomial distribution.
Specifically, how population proportion, p, and the probability, p (from binomial distribution) are related?
Are these two the same?
Hi, Brandon, thanks for your great classes. I learned a lot from them. One more thing, could you please recommend one or two textbooks?
the only issue about this topic is using the term "sample" and "sample size". You can use the term subgroup for sample while each subgroup has a sample size.
Hey dear Mr. Foltz,
It would be nice if you make videos about Partei and Semi partei correlation and also about Phi coefficients, Y coefficient ( gamma) , discordant und concordance for ordinal data correlation .
Thank you so much. You are the best @
Thank you Brandon...
11:49 This is the part where I'm confused. A few videos ago you mentioned the "Standard err of the mean" is inversely proportional to the sample size, so why is "Std Err Mean" decreasing when sample size stays constant here?
Ok never mind I think I get it now. There's another layer of sampling being introduced, so collecting more samples in the 1st layer is equivalent to increasing the sample size in layer 2:
1st layer: Sampling distribution of the sample proportion of size n=30 (with 25, 100, ... samples). Standard error only dependant on sample size, not number of samples.
2nd layer: Sampling distribution of the mean in layer 1. Our n now is the 25, 100, .. samples in layer 1 so it now has an impact on Standard Error of Mean of this layer.
Gr8, would help if you start non parametric series also
Hi Brandon, can you please tell which formula you used at 5:08 to calculate the standard deviation?
Greetings. Have you figured out the solution to this question? I have the same one. Using the standard deviation formula, which Mr. Brandon introduced in Statistics PL03 - Descriptive Statistics II.
Regarding the software calculations it seems that the Standard Error of mean (0.0144) is calculated using n as 25 rather than 30. This would mean that n is the number of samples rather than sample size. Please clarify which is correct. Thanks
In 13:59, why are you comparing the Mean to the Theoretically calculated Std Error? Shouldn't the Std Deviation (generated by the software) be compared with the theoretical value?
very very very helpful:)
Hi Brandon, is proportions synonymous with average or mean? Or am I missing something here?
Thank you so much for the videos.
I re-watched the first video, I must have missed something :)
why are we looking at mean for the standard error
So each time we are taking 30 different students for the sample and same 25 students are doing the sampling??
Hi Brandon,
Are all these samples mutually exclusive?
Does the sampling distribution of proportions normally distributed?
Sure is!
@@BrandonFoltz Thank you😊
Hi Brandon,
At 11:49, you mentioned the number of samples affects the standard err mean, from 0.018 down to 0.0009014 if increase number of samples from 25 to 10000. I am confused how to calculate this standard err mean? you mentioned that is the standard error of the mean of all sample proportions. But what is the equation for that?
Also, if number of samples does not affect the standard error of the sample proportion, we take 25 samples n = 30. Is that the same as 1 sample with n = 750? (750 = 25 x 30). In that case, we took many samples, which is the same thing as increase sample size?
Thanks.
I had the same question and was looking at the comment section if anyone has already asked. Thanks for posting!
From what I recall from the past videos is that std_error of the mean is the std deviation of the sampling distribution. So essentially, if we're calculating the standard deviation of the distribution of the sample proportions for each of the 25 (or call it the standard error), shouldn't the standard error and standard error of the mean be the same? I think I'm kind of in the weeds here.
@Brandon, we really appreciate if you could help us here! Thanks very much in advance.
Hi Xi, so essentially, the standard error that we are calculating is synonymous to the sampling standard deviation (don't confuse with the std error of the mean). For a sample of 25 proportion, we'd calculate standard error as sqrt((x - mu)^2 / (n-1)). So for a sample size of 25, this will give 0.0920346 (sample standard deviation of 0.0920346). Now you should also assume that in the absence of population standard deviation, sample standard deviation provides a reasonable estimate. So, we'll fit in this sample std deviation in the formula (sigma_x_bar = s/sqrt(n)) to calculate std error of the mean. So for a sample of 25, plugging in the values, the std error of the mean (this is the std deviation of the sampling distribution) will be 0.0920346/sqrt(25), which gives 0.018.
Now you can do the same thing for the rest of the sample sizes (25 to 100 to 1000) and you'll see that the std error of the mean reduces from 0.018 down to 0.0009014.
Hope this helps!
@Brandon, please correct me if I'm wrong, or if I'm missing something.
hi, Brandon flotz
I saw your video. I want to learn on Random walk, hitting time, mean first passage time.
any members can help. I want to learn it. please help me i will wait
Thanks, but I think more examples to each topic is required.
What If three or more nominal like good, bad and medium,. How do we proceed sampling concept.
Hi! Thanks for watching. There are tests for the equality of multiple proportions (p1 = p2 = p3) but that would be a slightly different topic than this. Hope to make a video about that sometime soon.
@@BrandonFoltz yes..thank you so much Brandon..I would appreciate, you may do this statistics on R or Some tool like SAS , SPSS by taking real-time problem. Again thank a lot
What! Biden knows stats!