Thank you for this video and all your others! They were most helpful among the videos I found online when analyzing my data. I especially like the clear and logical step-by-step explanations as well as your calm voice.
You only need to include 2 dummies in the LR model since the 3rd is implied by a double absence. e.g. just use DayTreatment and Inpatient. Outpatient is then the reference level (intercept)
Hi Todd - 2 questions if I may (1./) my exercise sheets talk about dummy coding n variables and turning into n-1 variables. And the regression analysis then goes on to show this e.g. only entering 2 dummy coded items into the analysis. Is that correct? In your presentation we actually turn 3 variables into 3 dummy coded items? (2./) what about if you have been given gender with 2 categories, but they are already coded as 1 = male and 2 = female. Do you need to dummy code so that values are changed to be 0's and 1's? Or can you run regression as is? Appreciate your time. Regards,
Thank you for your great explanation Dr. Grande! I do have one question, when I have different (dummy) variables created for the variable highest achieved education one of them is used as a reference variable/category when doing a regression analysis. Meaning, one of the six (in my case) variables are included in the constant variable so only five are shown in the output of the regression analysis. Do you work with reference categories when you calculate Pearsons correlation? Or is there no reference category when calculating pearson correlation and are all the 6 variables of level of education shown in the output table?
Thank you very much for this video. I still have a doubt about this procedure. If my nominal variable has only 2 levels and its coded by 1 and 2 (e.g. 1 = Has the characteristic; 2 = Don't have the characteristic) I still have to create dummy variables to run linear regression analyses?
Hi Dr. Grande: Thank you for the tutorial. I am trying to complete a multiple regression test using a categorical variable, which I have converted into dummy variables. However, they are still listed as categorical variables in the data variable view. So when I try to use them as independent variables, they are not scaled. Is that correct? I thought that once I converted them to dummy variables, their level of measurement would change from categorical to scale. I am really confused, and I don't think my test is calculating properly. Any guidance you can provide is appreciated.
Sir I wanted to ask about my categorical variable. My categorical variable named types of smoking behaviour has 3 categories that is smoker, non smoker and ex-smoker. I named this variable as (sn) and give values like 1 for smoker then 2 fir no smoker and then 3 for ex-smoker. When I dummy code this variable I have three separate variable.i dummy code them because I wanted to apply Pearson correlation analysis on my categorical variable. But in correlation analysis it was a rule that only dichotomous variables are used. So please suggest me that how should I apply correlational analysis on categorical variable ?? Because I want the results of all the 3 categories with a separate correlation with other variables. Please sir help me out.
thankyou so much,...could help me entering data so that I could get name and occupation side by side when there are many variables in occupation and its crating much bigger unnecessary table..thankyou
However, I still don't understand why the group that is coded 0 in the k-1 category is considered reference group. since when they become k-1 independent variables.
Thank you Todd! I love how slowly you went. You really talked me through this. This is the best SPSS video I've seen. Your calming voice is great!
You are quite welcome - I'm so glad you found the video to be helpful -
Thank you for this video and all your others! They were most helpful among the videos I found online when analyzing my data. I especially like the clear and logical step-by-step explanations as well as your calm voice.
You only need to include 2 dummies in the LR model since the 3rd is implied by a double absence. e.g. just use DayTreatment and Inpatient. Outpatient is then the reference level (intercept)
Very useful comment. Thank you!
Thank you so much for your clear explanation, Dr. Grande! Appreciate it as it is very helpful to complete my analysis of my dissertation.
Thank you so much, your presentation is very easy to understand and easy to apply. Great way of teaching. Thanks. Asha Mehta
your presentation is very easy to understand and easy to apply. Thank you so much.
You're welcome!
Cannot thank you enough for all of your videos and help.
Hi Todd - 2 questions if I may (1./) my exercise sheets talk about dummy coding n variables and turning into n-1 variables. And the regression analysis then goes on to show this e.g. only entering 2 dummy coded items into the analysis. Is that correct? In your presentation we actually turn 3 variables into 3 dummy coded items? (2./) what about if you have been given gender with 2 categories, but they are already coded as 1 = male and 2 = female. Do you need to dummy code so that values are changed to be 0's and 1's? Or can you run regression as is? Appreciate your time. Regards,
Thank you, sir. You are really good!
Thank you. This was clear and concise
Thank you, the process was very clear.
Thanks, man! You saved lives here..
Thank you for your great explanation Dr. Grande! I do have one question, when I have different (dummy) variables created for the variable highest achieved education one of them is used as a reference variable/category when doing a regression analysis. Meaning, one of the six (in my case) variables are included in the constant variable so only five are shown in the output of the regression analysis. Do you work with reference categories when you calculate Pearsons correlation? Or is there no reference category when calculating pearson correlation and are all the 6 variables of level of education shown in the output table?
Hm? I thought with k levels you use k - 1 dummies, and use one as the reference group (0, 0).
Excellently clear and very helpful, thank you!
You are quite welcome - thanks for watching -
Thank you very much for this video. I still have a doubt about this procedure.
If my nominal variable has only 2 levels and its coded by 1 and 2 (e.g. 1 = Has the characteristic; 2 = Don't have the characteristic) I still have to create dummy variables to run linear regression analyses?
Hi Dr. Grande: Thank you for the tutorial. I am trying to complete a multiple regression test using a categorical variable, which I have converted into dummy variables. However, they are still listed as categorical variables in the data variable view. So when I try to use them as independent variables, they are not scaled. Is that correct? I thought that once I converted them to dummy variables, their level of measurement would change from categorical to scale. I am really confused, and I don't think my test is calculating properly. Any guidance you can provide is appreciated.
great tutorial! I have one question: are the 3 new variables ordinal? Or scale...?
Verrrry helpful. THANK YOU!
You're welcome, thanks for watching -
Thank you for valuable vdo. It is very helpful.
You are quite welcome!
Can these recoded variables from the one categorical variable with 3 levels be applied to binary logistic regression also?
Sir I wanted to ask about my categorical variable. My categorical variable named types of smoking behaviour has 3 categories that is smoker, non smoker and ex-smoker. I named this variable as (sn) and give values like 1 for smoker then 2 fir no smoker and then 3 for ex-smoker. When I dummy code this variable I have three separate variable.i dummy code them because I wanted to apply Pearson correlation analysis on my categorical variable. But in correlation analysis it was a rule that only dichotomous variables are used. So please suggest me that how should I apply correlational analysis on categorical variable ?? Because I want the results of all the 3 categories with a separate correlation with other variables. Please sir help me out.
Thanks a lot....very helpful.
You're welcome!
thankyou so much,...could help me entering data so that I could get name and occupation side by side when there are many variables in occupation and its crating much bigger unnecessary table..thankyou
However, I still don't understand why the group that is coded 0 in the k-1 category is considered reference group. since when they become k-1 independent variables.
Shouldn't we deduct one level when we do dummy coding?