Confidence Intervals for a Proportion: Determining the Minimum Sample Size

I discuss determining the minimum sample size required to achieve a given margin of error when estimating a population proportion. (For intervals based on the normal approximation.)

[TitanTubs]

YES, these videos from 8 years ago are STILL VERY HELPFUL.

[borenxu]

this is series in the BEST of all on the topic of proportion hypothesis test, Thank you!

[yitbarektenaw]

the information you posted is very important

[otabekrazzakov]

Thanks! these lectures really helped me to refresh my knowledge!

[davidhenry]

You can always make the complex simple!

[daleeps]

thanks a ton, very clear explanation.

[RichardGreco]

SQRT[p(1-p)/n] is the proportion standard deviation.  This concept could be applied for any statistic?  I could apply it to find the minimum sample size for the sample mean of a distribution for a given margin of error?  x_bar is the statistic and [sample mean standard deviation]/ sqrt[n] is the error. 

[OriginalJoseyWales]

Thats great and very instructive but I have a question which I would really appreciate an answer to. Staying with your first example the sanple size works out at 1068. But what if my population was in fact 1068. That is to say if I had a population of 1068 to choose from and I wanted to sample from this population also with a margin of error of 0.03 and a confidence interval of 95%? Using that formula it still works out at 1068 - my whole population. So how can I adjust for poulation size?

[jenroberts]

At ~ 4.30, what does the z-apha/2 mean for this example? I thought that meant 1.96 would be divided by 2? What is the purpose of the /2 above the m in this equation?

[haithamabdelrahman]

What is the difference between the formula her and the formula np≥15, n(1-p)≥15 ?, It is soo confusing
Other question, why we are keeping using z value rather than t value, the standered devition is not known, are we shoud using t value instead ?