Welcome back!!! We have missed you. :) Kickass video as usual. This kind of post-hoc testing is something that I have wanted to know for a while. Thanks a ton for finally shedding light on it.
Many thanks for this great video. my chisq pvalue=0.0073, meaning there is a link between lines & columns. But when i do the post hoc test (as in your video), all pvalues are greater than 0.05. How can i explain this?
In SPSS --> Crosstabs --> Cells, there is also an option called Compare Column Proportions (z-test) and Bonforroni correction for the same. It does not do the same thing?
It's difficult to say, exactly. The 'Compare Column Proportions' option simply ascribes a letter to the cell frequencies, where different letters suggest different proportions within a column. My hunch is that most people would find this insufficient. The procedure I demonstrated in the video allows one to estimate precise p values for each cell and the opportunity to apply any p value correction that may be considered appropriate.
+Indrajeet Patil +how2stats Yes... how does one read the Bonferroni correction in the compare column proportion option? Do you have a video about that? Thanks for the great explanation.
I have two set of data on vocabulary items provided by two groups of students. The total numbers of vocabulary items are not equal. The provided words are divided into two groups: correct and incorrect. Group 1 (n = 218) - correct = 48 / incorrect = 170 Group 2 (n = 352) - correct = 82 / incorrect = 270 I want to know whether the difference between the correct and incorrect answers across the two groups. Can I use a chi-square? If no, please suggest a proper statistical test. Thank you for your advice.
This guy rocks. His voice is so calming. He talks HUMAN!
Welcome back!!!
We have missed you. :)
Kickass video as usual. This kind of post-hoc testing is something that I have wanted to know for a while. Thanks a ton for finally shedding light on it.
You are amazing. I love your voice (hate my tutor's voice, so here (hear) we are....). Thank you. I think you've helped me pass.
You are a life saver, exactly what I was looking for >> watch all 3 videos
Can the same thing be done after a Fisher's exact test?
Many thanks for this great video. my chisq pvalue=0.0073, meaning there is a link between lines & columns. But when i do the post hoc test (as in your video), all pvalues are greater than 0.05. How can i explain this?
You are a God amongst men, my dude
The reason we use post hoc test, is it bcz it has more than 2*2 table?
In SPSS --> Crosstabs --> Cells, there is also an option called Compare Column Proportions (z-test) and Bonforroni correction for the same.
It does not do the same thing?
It's difficult to say, exactly. The 'Compare Column Proportions' option simply ascribes a letter to the cell frequencies, where different letters suggest different proportions within a column. My hunch is that most people would find this insufficient. The procedure I demonstrated in the video allows one to estimate precise p values for each cell and the opportunity to apply any p value correction that may be considered appropriate.
+Indrajeet Patil +how2stats Yes... how does one read the Bonferroni correction in the compare column proportion option? Do you have a video about that? Thanks for the great explanation.
Does this method apply to a 2x2 contigency table as well?
I have two set of data on vocabulary items provided by two groups of students. The total numbers of vocabulary items are not equal. The provided words are divided into two groups: correct and incorrect.
Group 1 (n = 218) - correct = 48 / incorrect = 170
Group 2 (n = 352) - correct = 82 / incorrect = 270
I want to know whether the difference between the correct and incorrect answers across the two groups. Can I use a chi-square? If no, please suggest a proper statistical test. Thank you for your advice.
use z test?
How do you do post hoc following chi square with PRISM? (comparing correlation of 3 SNPs with 3 phenotypes... categorical data)
Thanks! Is there a citation for this technique?
Awesome!