Welch (Unpooled Variance) t Tests and Confidence Intervals: Introduction

An introduction to Welch (unpooled variance) t tests and confidence intervals. (Inference for two means.)

[@HolaHappyHooman]

Hi, could you make a series on how you use software to do calculations and take a million samples and make box plots etc?

[@Simrankalra21]

can you please explain what command you have used for R software to find degrees of freedom of welch procedure?

[@fin-pundit9631]

Hi i have some doubts is 1)"Two population t-test with Hypothesis"=Welch (Unpooled variance)=Pooled variance t test 2)In 6:47 Alternate hypothesis helps to determine wether its Left tail or Right tail."or Two tail So as per the time line 6:47----->The diagram shows left tail .Shouldnt it be "Right tail"(Ha:U1>U2) 3)4:04-->Degree of freedom--->In the numeratior AND denominator we have done WHOLE SQUARE ()^ Someweher i have seen that "degree of freedom" numeratoor and denominator is not whole square.Is there a possibility that of not having whole square? Please clarify

[@jbstatistics]

1) I don't know what you are asking here. 2) At 6:47 the test statistic happens to fall in the left tail. That can happen with any alternative. The alternative hypothesis does not depend on the value of the test statistic. 3) I don’t know what you are asking here. That formula can be expressed in different ways. The form I give at 4:04 is one of them.

[@szucs2020]

Awesome!

[@sujitsubhash5290]

do you have a video for the wilcoxon sum rank test ?

[@jbstatistics]

Not yet.

[@drma1206]

Hello, First, Thanks a lot for this great video. Second, is the Welch-Satterthwait test is the same thing as this video is talking about? Also, I have unequal size of groups (e.g. 700 males vs 200 females) + the data is not normally distributed; which of these test is the best to use please (1- Henseler’s MGA Test, 2- Parametric Test, 3- Welch-Satterthwait Test). Many thanks,