Depends on how ordinal your data are. Something like a 5-point Likert scale, I would probably use it. Less than 5-points, I wouldn't. To my knowledge, there isn't really an equivalent Spearman partial correlation. If you are stuck, and you have access to it, you could conduct an ordinal logistic regression to control the effects of another variable.
Most people would say 'yes''; however, it is possible to start with a non-significant correlation which then turns into a significant beta weight; in this case, there would like be a suppressor effect; they are rare, but they happen. Usually, though, the non-significant correlation is associated with a non-significant beta weight.
If the sig. value is larger than 0.05 in the partial correlation table, does that mean that (SES) is a very important or vital mediator between the (IQ and AA), no matter the partial correlation dropped or increased from the bivariate correlation ?
if there are 12 independent and 1 dependent variable all of them having high value of correlation coeffecients in this particular case if we want to check for the relation for one of these will we put all the 11 other variables in controlling variable section?
Hi, i was wondering whether the variables i want to insert in the partial correlation have to pass unavoidably the normality test (for instance Shapiro-Wilk). Thanks for your time.
If I am trying to find a correlation between 2 ordinal likert scale variables while accounting for gender would I do that the same was as in the video? Thanks
Can you use this technique if you have ordinal level data, since the output shows Pearson's correlations? Watched your vid about Spearman's rank order correlation. Have data that is ordinal on one variable and interval on the other.
I have a problem with my research, I have 5 sample for my research, and every sample I have 6 times test, my lecturer said that I she want to know bout correlation of them. What should I do?
How do you get SPSS to show you the ones that are signficantly correlated when you do partial correlation? I notice when you do partial the signifcant correlations are not flagged? Please help.
Is it possible you accidently clicked the button to not flag significant correlations? There's an option at the bottom left of the partial correlation window.
How would you perform a correlation between three variables with one grouping variable? Eg looking at the abundance of 3 parasites in different sections of the gut? :)
Use the 'Split File' Utility in SPSS: Data --> Split File --> Organize Output By Groups (add group variable to the box). Then, run the correlations as per usual.
Excellent tutorial. Good example used and you speak clearly and at just the right speed. Did not need to keep replaying to get it.
Extremely helpful. Thank you for sharing these simple steps!
thank you!
Great tutorial on partial correlation!
Thanks
Depends on how ordinal your data are. Something like a 5-point Likert scale, I would probably use it. Less than 5-points, I wouldn't. To my knowledge, there isn't really an equivalent Spearman partial correlation. If you are stuck, and you have access to it, you could conduct an ordinal logistic regression to control the effects of another variable.
Very very helpful! Thank you!
thanks for this! really nicely explained :)
when we conduct a regression analysis do we have to have a significant correlation relationships between the IVs and DVs?
Most people would say 'yes''; however, it is possible to start with a non-significant correlation which then turns into a significant beta weight; in this case, there would like be a suppressor effect; they are rare, but they happen. Usually, though, the non-significant correlation is associated with a non-significant beta weight.
If the sig. value is larger than 0.05 in the partial correlation table, does that mean that (SES) is a very important or vital mediator between the (IQ and AA), no matter the partial correlation dropped or increased from the bivariate correlation ?
if there are 12 independent and 1 dependent variable all of them having high value of correlation coeffecients in this particular case if we want to check for the relation for one of these will we put all the 11 other variables in controlling variable section?
Hi, i was wondering whether the variables i want to insert in the partial correlation have to pass unavoidably the normality test (for instance Shapiro-Wilk). Thanks for your time.
If I am trying to find a correlation between 2 ordinal likert scale variables while accounting for gender would I do that the same was as in the video?
Thanks
Can you use this technique if you have ordinal level data, since the output shows Pearson's correlations? Watched your vid about Spearman's rank order correlation. Have data that is ordinal on one variable and interval on the other.
I have a problem with my research, I have 5 sample for my research, and every sample I have 6 times test, my lecturer said that I she want to know bout correlation of them. What should I do?
How do you get SPSS to show you the ones that are signficantly correlated when you do partial correlation? I notice when you do partial the signifcant correlations are not flagged? Please help.
Is it possible you accidently clicked the button to not flag significant correlations? There's an option at the bottom left of the partial correlation window.
How would you perform a correlation between three variables with one grouping variable? Eg looking at the abundance of 3 parasites in different sections of the gut? :)
Use the 'Split File' Utility in SPSS: Data --> Split File --> Organize Output By Groups (add group variable to the box). Then, run the correlations as per usual.
how is calculated : RMSE,MPE, MBE by SPSS
thank you
bonjour,
comment en calcule le RMSE, MPE, MBE, par cet logiciel SPSS
et merci bien
While this was helpful, I couldn't bear to listen to your mouth clap every time you move it. Please invest in a better mic and pop filter.