How to calculate Sample Size

A video on how to calculate the sample size. Includes discussion on how the standard deviation impacts sample size too.

Related Video
How to calculate Samples Size Proportions

[statisticsfun]

Questions are always good. SD is the standard deviation and it tells how disperse the data. If you have a high SD you will have a bigger samples size.

E or error is the acceptable error. It is usually given or the researcher decides on it.

Good luck in your classes.

[AishaDiane]

Short, sweet and straight to the point. thank you. So easy to follow.

[statisticsfun]

Yes you are correct. Either equation should give you the sample size. Do make sure you watch the video on sample size for proportions (see link in video description).

[Zahara]

You Rawk! Test on Monday and this is helping so much! Thanks for making Stats easy to follow!

[statisticsfun]

If you are looking at populations that are not normally distributing you should think of ways you could segment the data so it more closely fit a normal distribution. For example, if I was looking at heights of all adults versus heights of adult females and heights of adult males.

[statisticsfun]

I am not sure what is "RCT" or perhaps it does not matter. The sample size does not depend on the type of study, but the number of dropouts that are excluded would impact the sample size. From what you have told me I would say 1 is false.

[statisticsfun]

Sounds like you are frustrated, but don't be hate'n! I post a lot of videos which are available for free.

[jaytee_]

Great video. Let's say you don't know the standard deviation in a population....how do you determine what sample size will be adequate?

[statisticsfun]

That would actually be equation for sample size for proportions. I added a link in the video description to a video that discusses using this formula.


Make sure you like MyBookSucks on FaceBook (see link in video description). This will help other students find the educational videos.

[statisticsfun]

Good luck on your test, you can do it!

[ejshelby]

The calculation of sample size in order to...decide on sample size, is kinda chicken and egg. You first need means and SD, which means you needed to conduct the experiment already. So what's the point? It only helps on the next experiment.

[Songvbm]

I have a weird problem in fitting negative binomial where the mean and variance is given and I have calculated the P(X=x). But cannot determine E=N. P(X=x), because N is not given.

[jonrrobinson]

Many thanks for your videos! They are really informative.

Can I just check my understanding on one point. Is the error rate different from the margin of error?

[morriscolenbrander]

Your voice sounds like Russel from ownage pranks. Besides that, very helpful video.

[balajis]

When power analysis to be used to identify the sample size? Why it is not discussed in the above video?

[islamaah]

How do we calculate a standard deviation before we start our thesis

[bonakalisiwemasinga]

I wanna know that the formula does not change if it requires the minimum of sample size?

[madscientistmadscientist]

Prof i am probably naive asking this.
You have a population of say 100 cases and worked out standard deviation, standard error and finally the sample size which worked to be say 75.
Would you select 75 or 100 cases from your population for your study?
If you chose only 75, would you be selective or randomly chose the 75?

[lexingtoninterpreters]

Where did the 3.84 come from? You didin't explained! Thanks!

[gianlucalepiscopia]

Morning Sir
1.96 is for Gaussian or distribs that approach Gaussian. What if we are out of this domain? What if we have a Pareto distrib ?