Jamovi 1.2/1.6 Tutorial: Logistic Regression (Episode 19)

Подписаться 8 тыс.

Просмотров 7 тыс.

50% 1

In this Jamovi tutorial, I go through a quick example of a binomial outcome logistic regression. The outcome is dichotomous, rather than continuous, so you have to use logistic regression, and the usual general linear model regression. I discuss the output and go through a quick interpretation of this data's results.
Jamovi stats: www.jamovi.org/
NOTE: This tutorial uses the current build of Jamovi, 1.6.16 on MacOS. Version 1.6 contains all the new features to the program as of this recording date. Version 1.2.27 is listed as their most stable build.
Find me on Twitter: / profaswan
Go to my website: swanpsych.com
Twitch streams on psych & related topics: / cogpsychprof

Опубликовано:

13 мар 2021

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 17

@voltairecarlogarcia8620 2 года назад

Thank you! will be using jamovi for my Thesis especially Binomal Logistic regression

@anastasiakondrateva7519 2 года назад

Dear Alexander, thank you very much for the tutorial. May I ask you what package have you installed with so many functions for LR?

@AlexanderSwan 2 года назад

I'm not sure what you're referring to. When you open logistic regression on your Jamovi, does it not have all the options that I explore in this video? It could be a version issue. Or it could be that I have a ton of additional packages added through the program (plus sign icon at the very top right).

@ApoorwVikramDwivedi Год назад

Hey Alex, thank you for the video. Why is the significance or p value not provided for reference variable in factors (categorical independent variables)? It's provided in SPSS.

@AlexanderSwan Год назад

No idea why it’s not provided, sorry. That’s a dev question :)

@sandon4447 Год назад

Hey Alex, awesome video! I'm from Aus and am studying my third year of psychology and currently doing stats! My question is, how do we know if the model is a good fit to the data? Is it based on the model fit measures table? And if so, how do we interpret it? Thanks so much!

@AlexanderSwan Год назад

Yes, the model fit table tells how good your regression model worked. It's been a while since I looked at logistic regression in Jamovi, but this page gives a good breakdown of the model fit stats Jamovi generates: www.r-bloggers.com/2015/08/evaluating-logistic-regression-models/

@kmahawar111 3 года назад

Thanks. Great tutorial. How do we get data into Jamovi?

@AlexanderSwan 3 года назад

Take a look at some of my earlier episodes on this playlist to find out how to import your data into the app

@MichaelSturgeon0317 3 года назад

I have been looking for something like this. Is the dataset you used in your demo available ?

@AlexanderSwan 3 года назад

Unfortunately, no. It's not my dataset, so I cannot distribute it. It's many many years out of date too.

@galactic-nucleus 3 года назад

@@AlexanderSwan Hi Alexander, thanks for all the Jamovi tutorials! They are a big help to me in learning Jamovi. In the future, can you please use datasets that we can all access? There are so many from within Jamovi from the Learning Statistics with Jamovi and R data sets. It makes it so much more useful learning-wise if we can tweak the parameter settings based on what you're teaching us. Again, thanks for all your tutorials! Keep them coming.

@AlexanderSwan 3 года назад

I definitely do try to do that. The only reason I chose this dataset for this particular example is because I had my work from my time in graduate school, so I knew what I was looking for. I'm less confident in this particular set of analyses, so this was more about showing off how you do it in jamovi rather than teaching how to interpret the outcomes and values. The vast majority of my tuts use the data library stuff from either jamovi or JASP.

@yazanalkhatib5725 Год назад

And ís this what we consider to be a multivariate analysis or just multiple univariate analysis??

@AlexanderSwan Год назад

Nah, multivariate generally refers to how many outcome or dependent variables you have, so when you only have one DV, in logistic regression for example, then it’s still univariate. You can have multiple predictors though, so your second label is more correct.