Interpreting Odds Ratio for Multinomial Logistic Regression using SPSS - Nominal and Scale Variables
HTML-код
- Опубликовано: 22 авг 2024
- This video demonstrates how to interpret the odds ratio for a multinomial logistic regression in SPSS. In this example, there are two independent variables: one nominal variable with three levels and one continuous (scale) variable. One dependent variable with three levels is included.
This is probably the best video that explains how to work Multinomial Logistic Regression for SPSS. It would have been nice if a video was also produced to explain the Full Factorial option since it seems like nobody online has spoken about it :(
I have been researching MLR for months for an explanation on how to interpret the results. Dr. Todd, by far, did an excellent job covering all of the MLR bases. Thank you.
You're welcome, thanks for watching -
Maybe you just saved my Bachelor Thesis! Thank you. Keep up the great content. Furthermore I liked the calm way you are explaining. It is easy to follow.
A true hero and life saviour! I enjoy watching ur videos a lot, very professional.
Very clear and easy to understand interpretation of the results that made it easy for me to follow (even if I do not use SPSS).
This is the video you were looking for. 👌
Sir, you have done a great work to humanity, thanks a lot, but can you give us more justifications for the selection of a particular variable (among the 3 dependent variables) as a reference variable and can you also help us to use stata software to explain this same tool? Thank you in advance...
Thank you Dr. Grande for a great video! It is very easy to understand. Can you please upload a video about nested logit model? Thank you.
Your explanation was very clear and a great help! Thank you very much!
Thank you very much Dr. Grande.
when the predictors are categorical variables with more than two categories should I recode them into dummies?
Hello Dr GRANDE, Thanks a lot for your videos, one question, should we assume that the odd ratio obtained is really representative of the population if the parameters estimates aren't significant?
I have the same question. Did you get an answer?
Thanks, Todd. I have seen several posts from you. It seems this regression is based on LOGIT transformation (default in SPSS), is there any way to change to the PROBIT model?
How you choose the number to divide in? In the last 5 minutes.
Please answer me
Iam finding your vedios very useful 😊⭐💖
You really are lifesaver!!
How can we caculate the odd ratios for Type 2 relative to Type 0 and Type 1?
As usual, great video! very helpful!
I am glad you found it helpful, thanks for watching -
Thank you very much for the helpful video. If I want to compute the odds of selecting CBT over Psychodynamic, can I just divide the beta of CBT by the beta of Psychodynamic?
I believe you cannot compute just by dividing the beta of CBT and Psychodynamic only because you need one predictor/ independent variable such as age or type (type 0, type 1, or type 2), good luck.
Hi Dr. Grande, thank you for your detailed explanation. I have a question and hope you can help me out. In your example, type 2 (family) was anchored as the reference variable, how do we compute the odds ratio for type 2 (family) then? Thank you, deeply appreciate your help.
dr it's very great to have you here, but i have a foolish question, whether we must move our categorical or nominal variable to the factor box and conti variable to the covariate box ? or is it not like what i am understanding ? thank you in advance
Why is the type 2 is set to zero. ??
This is very nice approach! here it should have continoues part. please, send any more pdf and other sources of interpreting spss result in MLR analysis.
thanks a lot sir. very helpful lesson. live long...
You're welcome!
Prof. this is your first lecture, i could not understood very well despite i watched it 5 times.
I have a question, how can i get Odd ratio comparing two variables with three variables? I mean i have two variables of depression by scale, and three levels of response like none, moderate and high. and what type of regression is applied on it? is there anything like 2x2 table, any 3x2 table for OR calculation, or by an mean i can calculate OR for them? thanks
Thanks for the video, it's excellent.
Can robust variances be calculated for prevalence ratios or odds ratios in multinomial regression models for cross-sectional studies? If so, can prevalence ratios be calculated in multinomial regression models in SPSS? Please, I await your help. Thank you!
Hi Sir, thanks for your videos they are very helpful. I am actually working on my thesis now and i have i am trying to perform a multiple binary regression on the outcome of passing or failing a financial literacy score. My predictors are all categorical with some assuming nominal or ordinal scales. Could you please upload a video on how to perform this type of regression.? Thanks in Advance. By the way i suscribed already :) .
I still dont get it -__- and im writing my chapter 4 for my dissertation
I'm writing my chapter 5 and I'm still struggling to understand what are the benefits of using Exp(B) rather than directly interpreting based on the values of B :-) I'm not a statistician btw.
I want to ask, so it is okay right to treat ordinal variable as covariates?
Hello Dr. Grande, I have a question. Is there a precedence of taking a random sample of one category of the dependent variable so as to have similar proportions to the second for a three category dependent variable. My category proportions are 0.77, 0.20, and 0.027. Is there any other way to model the three category dependent variable for these proportions.
Hi Dr. Grande, thank you for your clear explanation. I am wondering, when you have more than one categorical predictor variable in a multinomial regression analyses; which category is held as the constant when interpreting the other variable. In this video you say; we interpret 0 compared to 2 when age is held constant. So my question is, how do I know at which level of the second binary variable it is held constant when interpreting the first binary variable. More so, when you include the interaction term for these two binary variables, at which level of the one binary variable do you interpret the other one? Thank you so much! Any help would be greatly appreciated. Best, Marieke
Hello sir, It was awesome but i want to know if i want to put predictor's reference so how do i do it???
If Type has 3 categories, shouldn't we define (3-1)=2 dummy variables?
If you have a categorical variable you always define dummy groups less than one in categorical variable. For example, you have a categorical variable with three levels . you have to define two dummy variables. Omitted category set as reference group.
Hi Sir, how can I calculate the adjusted odd ratio in spss? Thank you.
I have odds ratio that is 19.5. how do I convert it to percentage.
1850% greater.
( (Y2 - Y1) / Y1 ) x 100
Thank you!
You're welcome - thanks for watching.
Dr. Grande: If i have a question how can I reach you?
Hi Sir, i have one question hope you can help me! I have a continuous independent variable(eg. annual premium paid for life insurance) with three independent variables( dummy variable for demographic, likert scale for another two IVs), should I run multiple linear regression?
Your question is unclear what is the type of your dependent variable? continuous or categorical (two or three or more?) ?