Hypothesis tests on one mean: t test or z test?

A look at at what influences the choice of the t test or z test in one-sample hypothesis tests on the population mean mu. I work through an example of a t test, and compare the p-value of the t test to would have been found had we (incorrectly) used a z test.

[friedricetrash]

this man is the only reason I am not gonna pass/fail my stats class

[greyveil]

ur a life saver, i was trying so hard to find out the true rules of using t and z values

[jbstatistics]

Thanks!

I do everything on a Mac. The underlying presentation is done in Latex/Beamer, and I record and edit in Screenflow. It doesn't take too much to get up and running at first. But I've definitely learned a lot as I've gone along, and refined things quite a bit. Best of luck!

[BNT]

First of all, thank you so much for sharing knowledge, for free in your videos! I should say, I always enjoy and learn from your videos!

However, I have a few questions:
1. I couldn't see the significant level to set up your rejection area in the question. Which one did you use here?
2. Assuming you have chosen the 'usual/traditional' one of 5% (0.05), how did you come to reject Ho based on P-value (0.568) you obtained in your calculations? (When P-value is larger than significant level, your t-statistics tend to fall in "fail to reject area of Ho" that is: Ho= hypothesized mean).

I would appreciate for any insights from anyone here.

Thank you!

[scottsmith]

> Be a Stats professor
> Students asking for suggested online videos for clearer explanations about which distribution to use in word problems
> Not sure what's unclear but okay
> Tons of videos where they say to use z-test any time n >30
>
> Find this video, go to 2:21
> THANK YOU!!!

[RRR]

You are an awesome teacher. Gifted I would say! :) 

[CommandoX]

5.7% versus 4.9% — not insignificant, though, conceptually, a small difference.

Thank you for sharing this video — I found it to be very helpful.

[RECKANOWHERE]

Excellent explanation. Thanks for posting.

[bethanianoriega]

Great explanation but how do I calculate the P-value on TI-84?

[statsteacher]

Very good video!
How do you make your videos? It is not PowerPoint is it?
I teach at a college and will be teaching an online course and would like to make videos like this one.

[RRR]

Wow that is a very strong statement. But when n > 30, there isn't THAT much harm that could be done by using the normal distribution. The t table is a pain in the ass to use as the areas are limited. Probably why people try not to use it unless they absolutely have to.

[thatoyaonekedumetse]

thank you very much, it was a bit confusing when i thought about the sample size, but thanks for clearing it up

[Jack_Omush]

Does one need a level of significance (alpha) to get a p-value from the table? How are you getting the 0.0568 from the tables?

[coolgreensnake]

thx for ur effort man we really appreciate it, but i have a question about the sample size required for the hypothesis test and if there is a formula for every case to determine it? 

[sirl]

Can you send the t table you used to find the p value

[irafundafunda]

So if we can't use the sample size to determine whether to use a t Statistic or Z statistic then what should we use? Can't find this in your follow up video

[worship]

we were taught that if sample size exceeds 30, we should use the z-score... can you pls expatiate on this pls...tanx

[superturbomegasonic]

wow, i understand it now. thank you very much!

[richardl.]

Pls send link for the t table that you use

[xuelei]

thx a lot for the explaination, but i got a question. can i substitue μ0 for xbar when calculating s? (since i have made the hypothesis that μ0 is the mean)