Statistics 101: Nonparametric Methods, Kruskal-Wallis Test in Excel
ฝัง
- เผยแพร่เมื่อ 15 ก.ค. 2024
- In this Statistics 101 video, we continue our journey of learning about nonparametric methods (nonparametric statistics). Comparing more than two samples requires different techniques. In parametric statistics, we would use one-way ANOVA. The nonparametric version is called the Kruskal-Wallis Test. This video proposes an example problem, sets up a hypothesis test, and then we go into Microsoft Excel and do everything step by step. Enjoy!
My playlist table of contents, Video Companion Guide PDF documents, and file downloads can be found on my website: www.bcfoltz.comg/
Happy learning!
#statistics #machinelearning #datascience
Hello everyone! NOTE: I realized after rendering that I left the "r" out of Churchgate in the PPT and Excel sheet. My apologies. These things sometimes happen with proper names. Enjoy!
Hi Brandon! There is no RANK.AVG in my excel. Instead of using that can I use just RANK as an alternative? Thank uu
*use RANK as an alternative?
I have no words to thank you enough for this video! but, THANK YOU!
Wow! Very helpful and well explained, thank you!!
Great video again. Nonparametric statistics is easy with you. Thanks!
Great visual of the Chi-Square distribution. That really helped me
Thank you very much for this useful video. It really helped me!
Excellent video. After watching your video of Matt-Whitney Test, it was easy to understand. Thank you explaining so well.
Thank you, this was easy to follow step by step
Hi Brandon, love ur videos. Get more visual feel of the principle/logic behind the test. Thanks for your effort. Awsum videos!!
Suggestion:
----------------
It would be great if you could elaborate on the test stat mathematical formula at a high level as well. Maybe relate the principle/logic of the test with the mathematics of the test statistic. I have to do a lot of googling to obtain the mathematical understanding of the test stat. Having grasp of mathematical mechanics increases the understanding of the test.
Thanks, great video!
Thank you so much!
Great video thank you!!!!
Brandon, I am grateful for this video. It is very clear and understandable. I want to ask your favour, the graph in the closing part looks very good and can animate these iterations. Can you make a video to show how the graph was plotted? Thanks.
You are awesome!
beast video, very helpful
Thanks for ya help mate :)
Thank you for your help. This is great. Are we able to use a lower probability than 95%? If so, what is the appropriate lower limit?
Hi Tutor! Thank for providing the comprehensive video and practice material, I tried the sample data with JASP software but the result is slightly different. Can you please help me with this?
H = 8.734 and p = 0.013.
Thanks. But why no mention of options to perform these tests in statistical software like in so many other of your videos? Are these tests calculated by hand more often or something? I get showing by hand for the long way, but it would also be very useful to show more practical methods to run these tests.
hey brandon, i absolutely love the videos and they helped me a lot. i was wondering if you were willing to make higher or more advanced statistics playlist?
thank you
What did you have in mind?
@@BrandonFoltz for example can you go into more advanced probability topics such bayesian probability?
and also talk about topics such as Geometric mean and Harmonic mean, as well as skewness and kurtosis
and lastly time series analysis
ohh and i forgot can you talk about lognormal distributions, thank you
Hi, I have a bit of confusion. I have data on foraging strategies (descriptive data), I do not know which method to follow. I basically have to compare these strategies at two locations. All the videos I watch have some kind of numerical data, it would be great if I could get help on this. Cheers.
Hello there, is likert scale (strongly disagree, disagree, neutral, agree, strongly agree)with 4 different population suitable to use KW test?
Can we add different groups amd check the column interactions with them using this test?
Ah man I have the same question
Hii Brandon the dof for Chi Square u have taken is 2 it would be(row-1)*(column-1) i.e. 2*14 which is 28 and not 2 .Correct me if i am wrong.
What if the five parameters being assessed have different relative importance?
Hi! In that case, several options. Different survey design, weighting and a different type of analysis. The one-way ANOVA / KW test is only going to look at an aggregate single value across groups. What you are looking for would be more of a block / factorial design.
Would not the lowest possible and highest possible scores be 15 and 75 respectively, since there are 15 raters per store?
5 and 25 are the lowest and highest possible scores that each rater can give to each store, as the stores are evaluated with five criteria on a scale 1-5 (video 4:33) (I don't know if I missed your point, but trying to help :) )
@@lilianafreitas3718 Thanks for the clarification,, Liliana.
Sir, chi-square calculate
Why .95 sir instead .05
What does it mean that the P-value of H is 0.014 and it's less than the chi-sq P-value of 0.05? yet we concluded that H at 8.58 is greater than chi-sq 5.99!! Pls explain..
Hi I have a big data, nT is 1464 and I did everything as you showed but as a result H is negative. What is the problem?? I can't find p-value due to that:(
I struggle with the same problem :( did you solve that somehow?
Me too..
There is no RANK.AVG in my excel :( what do I do?
What version of Excel are you using? Averaging tied ranks is a must. So, either Excel Online or Google Sheets would work. Both have the RANK.AVG function. Thanks!
@@BrandonFoltzThank you very much!! :)
Hi Good day. Can I have a copy of your powerpoint? Thank you.
Three dudes evaluated three coffee shops? Doesn't within-subjects design mean the results are dependent and you need Friedman's test, not Kruskal-Wallis?