Hi, thank you! Is there an alternative to a two-way test? I mean, I have a repeated-design data, with a continuos Y predicted by two categorical predictors X1 and X2, but my residuals are not normal distributed, so I cant run an Anova or a linear mixed model...
Thank you so much for the video, I have a quick question: How should I assign ranks if I have the same value, for example the morning and evening are both 35
I came up with the same question. I ended up using a Wilcoxon (W) signed ranks tests for matched pairs. Not sure if it is ok cause I found that there are multiple-comparison tests such as the Turkey-Kramer test which is based on signed ranks multiple comparisons. Hope you get the right answer from @DATAtab.
Nice. A question: what if I have 10 values recorded in morning but 7 values recorded in the noon and evening? Should we only exclude the extra 3 rows in the morning? Is there possible to incorporate and use that data? Thank you!
Yes the data will then be thrown out. There are methods to estimate the values, but there are many different methods that work differently well, so I can not give you a blanket tip! Regards Hannah
I don't understand. The parameter should be distributed normally (the sample distribution), so who cares about the data?????? I think Friedman test should be used when the residuals are not distributed normally, because ANOVA assumes a normally distributed residuals. When converting your data to ranks --> fitting am model, then the residuals would be normally distributed, and it is simple to understand. If the scores of every one is (1,2,3,4,5) , then your residuals will be distributed normally. It's a very common confusion to think we care about normally distributed data, but actually we don't, as the parameter is normally distributed in its sample distribution (given a sufficient sample size). The assumptions is about the RESIDUALS. Friedman test fixes it by converting scores to ranks. It is similar to the Spearman regression.
![One-Way ANOVA [Analysis of Variance] simply explained](http://i.ytimg.com/vi/mOdYddj5IG8/mqdefault.jpg)
This explanation is so clear and helpful! Kudos!
Glad it was helpful!
I learned this test today in class and watching this tutorial again, have craps the content better
Many thanks! Regards Hannah
Thank you so much🤍 God bless your channel more!!
Glad it was helpful and many thanks for the nice feedback! Regards Hannah
Well explained. Finally I got it. Many thanks!
Glad it helped!
Hi, thank you! Is there an alternative to a two-way test? I mean, I have a repeated-design data, with a continuos Y predicted by two categorical predictors X1 and X2, but my residuals are not normal distributed, so I cant run an Anova or a linear mixed model...
I have the same question too
Pls are we going to be given the value for the morning, noon, and evening?
Thank you for the simple explanation. I want to know what is the correct post-hoc test if Friedmann gave significant result?
You're welcome and thanks for the nice feedback! You can use the bonferroni correction. Regards, Hannah
So then do a sign rank between pairs and consider a p of ‘0.05/number 0f pairs’ significant ? Thank you for the great video.
What if 2 no. Are same in row how to rank tha for eg. In a row the no. Are 6,6,4 In such case how to rank
Thank you so much for the video, I have a quick question: How should I assign ranks if I have the same value, for example the morning and evening are both 35
great video keep doing the great work
can we pay for this calculator??
when calculate the hand hoe to separate same value like 1,2,3?
Here you find the calculator: datatab.net/
@@datatab thank you
can use minitab software for do this frideman mean separation
I can't access the Calculator since it needs payment to see some of the answers
Yes that is true! But it is cheap!
Is it 19.99€per month or 1999€ per month
Thank you for the clear explanation. One question: What type of post-hoc test is used for the pairwise comparison?
I came up with the same question. I ended up using a Wilcoxon (W) signed ranks tests for matched pairs. Not sure if it is ok cause I found that there are multiple-comparison tests such as the Turkey-Kramer test which is based on signed ranks multiple comparisons. Hope you get the right answer from @DATAtab.
How do we rank the values if two of them are equal?
for example if morning = 39, noon = 39 and evening = 40
can i start with the lowest value too?
Yes that should work too! Regards Hannah
Amazing tutorial, thank you so much!
Many thanks! Regards Hannah
very clear explanation.. thank you so much.
Glad it was helpful!
Nice. A question: what if I have 10 values recorded in morning but 7 values recorded in the noon and evening? Should we only exclude the extra 3 rows in the morning? Is there possible to incorporate and use that data? Thank you!
Yes the data will then be thrown out. There are methods to estimate the values, but there are many different methods that work differently well, so I can not give you a blanket tip! Regards Hannah
@@datatab I like your answer.
Mam plz explain the concept of statistics in detail it will be very helpful for me
As soon as possible !!!
Nice explanation
Thanks!
Thanks so much!
Amazing, thank you😊
Thank you
Love you from India
Many thanks! 😊Regards, Hannah
I don't understand. The parameter should be distributed normally (the sample distribution), so who cares about the data??????
I think Friedman test should be used when the residuals are not distributed normally, because ANOVA assumes a normally distributed residuals.
When converting your data to ranks --> fitting am model, then the residuals would be normally distributed, and it is simple to understand. If the scores of every one is (1,2,3,4,5) , then your residuals will be distributed normally.
It's a very common confusion to think we care about normally distributed data, but actually we don't, as the parameter is normally distributed in its sample distribution (given a sufficient sample size). The assumptions is about the RESIDUALS. Friedman test fixes it by converting scores to ranks. It is similar to the Spearman regression.
Спасибо большое! Мне очень помогло в понимании.
تشرح افضل من دكتورة مها
Rank order is wrong
Thanks!
nice (: