How to perform paired t test in Excel

[@LJ-jj2lu]

I'm a little confused about which P value to look at (one or two -tail) ? Thanks so much :)

[@SocratGhadban]

If you want to check for a difference without specifying the direction (greater or smaller), you use a two-tailed test. If you want to check for a difference while specifying the direction (whether it is greater or smaller), you use a one-tailed test.

[@jeffreyagaya2705]

Is the paired T test a best tool to backtest the LGD and EAD of ECL model?

[@SocratGhadban]

Thank you for your comment, but this goes beyond my expertise. I asked the question to Chatgpt and it seems it's not appropriate. To verify

[@robinsingh5252]

What should i select in output range?

[@syncgalaxy7326]

The cell where you want to generate your results

[@SocratGhadban]

@@syncgalaxy7326 Exactly

[@marialuisasanchez1745]

Thank you so much sir. More power to your channel

[@SocratGhadban]

[e-mail] please send me an email, i will forward it to you

[@jaysons.1022]

Hello, i am doing a experimental research and i also use pre test and post i notice that there might some outliers in our data how can i identify it

[@SocratGhadban]

Hello, Observations that lie more than 2 or 3 standard deviations from the mean are often considered outliers. another method you can use the Interquartile Range (IQR) Method: Calculate IQR = (Q3 - Q1) and define outliers as observations outside the range [Q1 - 1.5*IQR, Q3 + 1.5*IQR].

[@jaysons.1022]

Thank you for clarifying my doubts.

[@SocratGhadban]

My pleasure@@jaysons.1022

[@living6671]

should we also include the control group's results for before and after in the test?

[@SocratGhadban]

yes like this you can confirm the existing relationship

[@TserJords]

How did you make the graph Sir?

[@SocratGhadban]

I didn't make it, I have as a picture!

[@brenda_pma]

Great video! I'm a foreign language teacher working on a research project. To demonstrate that the strategies I teach in class improve students' language skills, would you recommend using a one-tailed paired t-test to show that post-intervention scores are higher than pre-intervention? How is it done?Can you (please 🙌🏻 )briefly explain? Thanks in advance!

[@SocratGhadban]

Hello, thank you. In your case you have to use a one tailed test, because your hypothesis is having a direction (higher >) if the p one tail is less than 0.05 then there is a significant difference.

[@brenda_pma]

Again, thank you so much ☺️! Your video and your explanation have helped me a lot!

[@SocratGhadban]

This is the purpose :) please share in the future the results of your research@@brenda_pma

[@hashemahmed1402]

if you have multiple measurements for the same patient (Before) and multiple measurments for same patient (After) , does this work with paired t test?

[@SocratGhadban]

This is exactly what the paired t test is designed for!

[@almariefuentes9506]

Hello. Can I ask a copy of the graph?

[@SocratGhadban]

Hello, please share your email.

[@arslansiddique32]

Good video. Well explained

[@SocratGhadban]

Thank you. Glad you liked it!

[@cachazulfa]

hi how do you make the graph ?

[@SocratGhadban]

Hello. This graph wasn't made on Excel, if you want to have it just leave your email here

[@aly7559]

hi can i know how you made the graph for the paired t test?

[@SocratGhadban]

Hello. This graph wasn't made on Excel, if you want to have it just leave your email here

[@aly7559]

@@SocratGhadban [e-mail] , thank you :))

[@SocratGhadban]

Graph sent! my pleasure!

[@marialuisasanchez157]

@@SocratGhadban hello sir.. can I have it too? Looking forward to your response

[@SocratGhadban]

@@marialuisasanchez157 Hello, i need your email!

[@marialuisasanchez157]

I dont know why everytime I put my email address here it was being remove

[@joli3978]

I have used this video in my class for a few semesters now, and this week a student pointed out to me that the hypotheses are not what we talked about in class, and was confused. Just an FYI, null should be less than or equal to, a typo you might want to correct.

[@SocratGhadban]

Hello Jo Li.. Thank you for your comment. In fact, I'm not sure if I got your point. If you're talking about the null hypothesis, you can write either ways mean equal to mean or difference equal to zero.. The alternative hypothesis could be one of two options : different (≠) where it's two tailed, if I'm just interested to know if there a difference, or one tailed like the case we're having in the video (here greater)..

[@ITsmotorjo]

Great video doctor 💪

[@anthonyelsebaaly6251]

The best on youtube 👍

[@SocratGhadban]

Glad you liked it! time for application now ;)

[@marialuisasanchez1745]

Sir I cant put my email address here I dont know why seems there is a problem. I tried it many times

[@jaylordliquigan2594]

Do a two-tail sir plss

[@SocratGhadban]

Check this video, it will help you understanding the difference between once tailed and two tailed test kzbin.info/www/bejne/eKOpc2xndpyLmck

[@kenesu1281]

❤💖

[@mantasha6402]

8.663 is not less 0.05 . P value is 8.663 and it is greater than 0.05

[@SocratGhadban]

it's 8.663 * 10^-5