Most often, if you want to know the degree or strength of a relationship, you would use a correlation coefficient (or maybe the square of that, which tells you the proportion of variance in the outcome variable that can be predicted by the other variable). Regression is generally better when you want to make a prediction about what the score would be, regardless of how strong the relationship is. There is a lot of crossover between the two, but that's how I would do it.
Because a bivariate regression is such a simple procedure, you can usually report the results in one or two sentences, like this example from Jeffrey Kahn (just google "apa regression results" and it's the top result): "Social support significantly predicted depression scores, b = -.34, t(225) = 6.53, p < .001. Social support also explained a significant proportion of variance in depression scores, R2 = .12, F(1, 225) = 42.64, p < .001."
Hello, Bart! I have watched a lot of your videos recently and, while I learned a good deal of SPSS, I still have this doubt, hope you don't mind answering it I want to see correlation between the "anxiety" and "experience" variables, being anxiety 0 or 1 (no or yes, respectively) and experience ranging from 0 to 3 (0: smaller experience in relationships, 3 maximum). What kind of correlation should I apply? Even better if you can redirect me to a video of yours explaining. Thank you very much!
Well, there's not much to do with a bivariate regression because it's a very simple procedures. You should, however, make sure that your data are in good shape: check for coding errors, check for outliers, check for normality, etc. A large sample helps, of course. Other than that, if you want to get more sophisticated, you would need multiple regression. Does that help? Bart
I think you need to use simple logistic regression, not simple linear regression, because your dependent variable is a proportion (literacy rate from 0 to 1), not a continuous variable. I was hoping this video would show me how to do one in SPSS.
thanks a lot for video :D I'v performed a binary logitic regression analysis with 2 independent (categorical) variables, the p valuse were significant but OR ratio were less then 1, what migth be the reason ? what does it mean please help ((
@gilletteTBAMCG Also, I know this isn't multiple regression... I got here from your multiple regression tutorial, trying to wrap my head around all the basics. Stats aren't my strong suit!
Good Vid! but remember people, you cant confuse 'cause' with 'correlation' ;) Higher literacy rates don't necessarily cause a higher average life expectancy.
Thanks alot for this! very clear explanations and your voice isnt as irritating as the other people in stat help videos :) keep em videos coming aye
Your tutorial is saving my thesis!! Doing a massive multiple regression as an undergrad has been very overwhelming :(
Most often, if you want to know the degree or strength of a relationship, you would use a correlation coefficient (or maybe the square of that, which tells you the proportion of variance in the outcome variable that can be predicted by the other variable). Regression is generally better when you want to make a prediction about what the score would be, regardless of how strong the relationship is. There is a lot of crossover between the two, but that's how I would do it.
This was very helpful and easy to understand. Thanks
Really well done videos. Thank you very much Bart!
His name is Robert Paoulson!! Pretty good demonstration of complex statistical procedure
Because a bivariate regression is such a simple procedure, you can usually report the results in one or two sentences, like this example from Jeffrey Kahn (just google "apa regression results" and it's the top result):
"Social support significantly predicted depression scores, b = -.34, t(225) = 6.53, p < .001. Social support also explained a significant proportion of variance in depression scores, R2 = .12, F(1, 225) = 42.64, p < .001."
Man, this is very helpful !! Thank you so much fro uploading this video :)
Hello, Bart! I have watched a lot of your videos recently and, while I learned a good deal of SPSS, I still have this doubt, hope you don't mind answering it
I want to see correlation between the "anxiety" and "experience" variables, being anxiety 0 or 1 (no or yes, respectively) and experience ranging from 0 to 3 (0: smaller experience in relationships, 3 maximum). What kind of correlation should I apply? Even better if you can redirect me to a video of yours explaining. Thank you very much!
Well, there's not much to do with a bivariate regression because it's a very simple procedures. You should, however, make sure that your data are in good shape: check for coding errors, check for outliers, check for normality, etc. A large sample helps, of course. Other than that, if you want to get more sophisticated, you would need multiple regression. Does that help?
Bart
Do you have an example of how to report the results in APA format?
When I go to analyze, regression, but it did not give me the option to choose binary logistic regression, how do I get that option?
perfect!
I think you need to use simple logistic regression, not simple linear regression, because your dependent variable is a proportion (literacy rate from 0 to 1), not a continuous variable. I was hoping this video would show me how to do one in SPSS.
how to improve linear regression model and results?
what if i had more than one answer for each data ? how to put it in spss or excel? #.Cause as i 'v seen, one data for one answer.
thanks a lot for video :D I'v performed a binary logitic regression
analysis with 2 independent (categorical) variables, the p valuse were
significant but OR ratio were less then 1, what migth be the reason ?
what does it mean please help ((
@gilletteTBAMCG Also, I know this isn't multiple regression... I got here from your multiple regression tutorial, trying to wrap my head around all the basics. Stats aren't my strong suit!
Good Vid! but remember people, you cant confuse 'cause' with 'correlation' ;) Higher literacy rates don't necessarily cause a higher average life expectancy.
His name was Barton Poulson