Excel Techniques - 11 - ANOVA - Two Factor without Replication.avi
Вставка
- Опубліковано 9 лют 2025
- Excel Techniques for Statistics DGGB 6820 - Summer 2010
ANOVA - Two (or more) Factor without replication
(Please be aware that the audio is a little out of sync)
Download the Excel Workbook at:
docs.google.co...
Download the Stats Notes at:
docs.google.co...
![Two Factor ANOVA [without repeated measures]](/img/n.gif)
Excellent explanation. The example, and your conclusion at the end really helped me understand.
Thanks a lot!
Thank you for your documents and videos. They're very cool. Thanks again.
So what are we testing for? What is the null and what is the alternative? Does it make a difference that Rows were no a significant p value, but the Columns did have a significant p value? How does that work with the null and alternative hypothesis, whatever they are?
Thanks a lot! It helped me a lot!
Good explanation! I really appreciate the help!
You saved my life 😂
Lol that tiny roast in the conclusion
It is a verry good video. Thank you. I kile the last explenation. They are better suited for Life Guards rather than Marines :))
What if those recruits were tested again in all 3 tests to see if they have improved (say for example after 2 weeks), giving you 3 pairs of before-and-after scores? What test should you use if they improved in any of those performances?
it's called a repeated measures anova (2 x 3) for before and after (2) and 3 tests per individual.
Can we do a tukey test to compare between the subjects to see who is the best among them ?
I did everything you said but it comes up with an error message "this input contains non numeric data". I did click labels, I did highlight as you did, not sure why it won't do it for me. Help?
When you select blank cells (samples with different number of data points), the non-numeric data error pops up.
Hi guys, I hope you can help me solve this problem.
I want to verify the independence of two factor without replication, can i use '' ANOVA - Two Factor without replication'' ? and how?
In this example, I can verify the independence of factor " Recruit " and "Score", its relationship to the result.
Thanks a lot!
Thank you so much
Thank you, much appreciated!
hey how do you find SSE what is the formula?
The design of this data file suggests this is a ONE-WAY ANOVA with 3 levels (tests 1-3). If each person takes each test, then test is a repeated ('replicated') factor. Differences between tests should not be tested as if it is a 'non-replicated' factor because it is theoretically not true and you lose power to detect differences when you treat them as if different groups of people were taking each test.
Testing for differences between individuals is NOT sound practice, as each persone would have to be treated as a factor, or perhaps a factor of 'person' with six levels. This whole example is very botched in my opinion.
Truth. It's a poor demo that he should revise.
thank you so much!
Thanks!
Your analysis is inappropriate for the data. This test assumes that all the values are equally independent of one another, but in fact you have scores for SQ, RQ and TQ that all come from the same individual and are related to one another. This is a repeated measure of a subject. You can see this is true by scrambling the order of, say, the TQ values--you will get the same statistical result from the two-way ANOVA, whereas if the test were appropriate for a repeated measure on a subject you would get a different statistical result. This error is on many videos on youtube. I see now that weakside made the same comment below.
Except that weakside advocated for one-way ANOVA, which is also inappropriate for the same reasons.
Hi what is error how is it computed ? Can we use tukey test for columns.?
statistisignifcaldiffernce