SPSS #18 - Multiple Linear Regression - Moderator - Theory

Hi guys, welcome to episode 18 of my SPSS tutorial video's. Please leave a like, subscribe to this channel and if you have any questions, feel free to ask them in the comment section down below.

[@contessaa1671]

Thank you Bas for your great explanation. I do have a question though; I do have a positive interaction term and want to further run the analysis / understand the relationship by stratifying my moderator. Please a short explanation will be very helpful on how to proceed. Like for instance do I add my interaction term in the analysis with my stratified moderator? Do I keep my other independent variable continuous or do I group it as well? Please help

[@doge6154]

thank you

[@kebrai9196]

Hi Baspss I was wondering what I should input when making the moderator variable; I have two subscales to measure the locus of control I am trying to see the interaction between attitude and intention to behaviour and using locus of control as a moderator Do I write it as Attitudes*Internality + PowerfulOthers? Or Do I turn them into separate moderators? Hope my question makes sense Thank you

[@xLindsayvanReijn]

Hi, je videos zijn ontzettend leerzaam, dankjewel! Ik vroeg me af hoe het werkt als je in je onderzoek 2 modererende variabelen hebt: kan je dan de tweede modererende variabele in het vierde model zetten? Of kan je dit alleen los van elkaar onderzoeken (dus twee verschillende toetsen en dus verschillende modellen)?

[@tracyquetzal9477]

How to run moderating effect by stratifying and how to determine if the strata are significantly different?

[@yoshcookie4091]

Hello Bas, Is this than the same as a hierarchical regression or is this a single and multiple linear regression at the same time?

[@mmajdanac]

Hi Bas! Wow, I am really admiring your knowledge of statistics! I am struggling to make sense of my data, particularly interpretation of two sets of moderators that I have. It's like Japanese to me. Do you give private tutoring? :)

[@baspsstutorials9783]

Hi, appreciate the compliments! Sadly I dont tutor privately at the moment, really busy with my regular job. Im pretty sure someone in your real life network can help you! And small questions can always be asked here in the comments

[@alintje11]

How do I do this but with 1 IV (7 point Likert scale), 1 Moderator (intervention vs control), and 2 DV's (both 7 point Likert scale) ?

[@jhabindrapokharel]

how to perform regression including moderating effect if there are more than one predictors?

[@SalihFCanpolat]

You can add others as they are or create interactions for them as well. However, you should plan your model beforehand and which variables moderate and which variables predict.

[@karlaveras1793]

Hi! How come you didnt have to create dummy codes for job in the regression?

[@baspsstutorials9783]

Hey! Officially job is an ordinal variable indeed, which would result in making dummy variables.. But just for the sake of these videos I treat it as an interval variable, which doesnt necessarily conclude in making dummy variables

[@c.vriesde3867]

Nice video! Does the moderator needs to be binary?

[@baspsstutorials9783]

No it can also be scale/ratio! Age could be one for example: "With every year you grow older, the main effect increases with B extra" But it's harder to interpret, keeping it binary keeps it easier haha