ANOVA 2 - Calculating SSW and SSB (Total Sum of Squares Within and Between).avi

[piivanov]

SSB=SSE

[doit10]

yes its the same as SSE

[thecolincolincolin]

thank you Khan. i learnt the a whole semester of my uni's business stats unit using your concepts and explanations in 3 weeks.

[TheOak1983]

Thank you Khan! Keep going! YOu rock!

[JackF1116]

Is the SSW another way of saying SSE? SSE is what I learned, but they look the same.

[e0f3z]

My prof taught me nothing and final exam is tomorrow

You taught me so much

Thank you so much

[gungs001]

@e0f3z word nigga

[sbs860]

It was really helpful. Thanks :)

[smmul89]

Thanks to Khan Academy I no longer attend me STAT class. Why? Because this guy is a million times better then my professor whos been teaching statistics for over 15 years.

[PeaceLoveHappiness09]

thank you to the moon and back!!!! I was sick friday and missed this lecture in class!! you just saved me from being extremely confused!!!! :)

[eternalko]

Don't really get the df.

Why in the prev video:Contingency Table Chi-Square Test

we had the same table and df=(rows-1)(cols-1).

We could, presumably, apply the same logic here. But we don't



[eternalko]

@eternalko

Sorry, the comment was to the prev. video. Now I get the logic behind it :)

[bezzer1185]

@thelastbattle

degrees is something I dont think I will ever understand. but i look at it like this.

the only true way to calculate sd is if we know the probabilities of each value in our sample. since we dont, and assume all values have the same probability we have to make our standard deviation a little bigger by introducing degrees of freedom.

[thesixthsense1]

For Sal "Excellence is not an act but a habit".. Thank you SAL :)

[OrnellaJulia]

Priceless:)

[rdboaventura]

Just what I need. Thanks a lot!

[tenso2002]

hi.. thank u so much for all the help.. Can you please upload a video on GRACH model, OLS, ADF, PP Test.. Thanks a lot

[thelastbattle19]

i cant understand  degrees of freedom :(

[Breadfan01]

@thelastbattle19 It's the size of your sample minus 1. So if you are sampling 20 peoples' test scores, there are 19 dF. Sal gives the reasons in these vids.

[CogitoErgoCogitoSum]

Why dont we just compute variances of the different subsets? And compare them. Isnt that a better analysis of variance? Combining it all into an f-score seems to obfuscate the data... its just questionable as a valid concept.

[EDUCATE2007]

Thank you sooo very much!

[XenomorphLV426]

beauty!

[FHomeBrew]

You're video came a week to late :( last week I had a test on this :(