Binary logistic regression in SPSS (March 2021)

[@muhammadsaqib3752]

i think no video is better than this. once watching this carefully can make u able to understand log reg and how to use it. thumbs up

[@harrybartonessel2478]

Well treated statistical technique from a brilliant scholar.

[@mikecrowson2462]

Thank you, Harry. That's is very kind of you to say! By the way, I ended up making a few edits and updates to the Powerpoint this morning. (there's were some errors in the diagnostics screenshots previously I ended up correcting; and added a few more slides). So be sure to use that version instead. Thanks again for visiting :)

[@AlizaElisheva]

Thank you very very much! And thank you for the clarification on my comment on your MLR video. I went through your powerpoint linked above carefully alongside my own output and interpretations. This is so helpful. I really like the practical and applied approach. Thank you so much.

[@mikecrowson2462]

You are very welcome! Thanks for visiting. Best wishes!

[@yomnaelfert]

Very helpful and comprehensive video. Thank you so much for this. It has been exactly what I was looking for

[@jansinatarajan7631]

Thank you for the detailed information. How many variables can be added into Binary logistic regression anlaysis at a time?

[@cortexconsulting5986]

These videos have been extremely helpful! Thank-you so much for the resources and populated spreadsheets, wish I had found this earlier in my thesis!

[@khae9566]

Thank you so much SIr. very clear and helpful .

[@MSigs]

life saving video right here, thank you!!!!!!

[@mikecrowson2462]

You're welcome! Best wishes!

[@feliciajoseph4207]

This video is very detailed. Thank you for your effort.

[@elizabethjoy934]

Thank you so much for the video. It helped a lot for one of my presentations. I was stuck for long. Have a query : 1. Is it necessary to write the logistic regression equation at the end? Also how to do it? 2. Under model summary, there is a statement on estimation terminated at iteration 4 , does it have any implication on the results? Thank you..

[@mikecrowson2462]

Hi Elizabeth, thank you for visiting my site! Regarding your questions: 1. I'm not exactly sure what you are asking about writing 'the logistic regression equation at the end'. I'm assuming you are asking if it is necessary to include the logistic regression prediction equation in your write up. The short answer is that it depends on the norms in the journals you publish in and in your field. If those things don't matter, then it may be something you add (or don't add) for stylistic reasons. Usually, the prediction equation is given as something like this: ln(odds)=b0 + b1X1 + b2X2+...+ bkXk or ln[odds(Y=1)]=b0 + b1X1 + b2X2+...+ bkXk or logit (Y=1) = b0 + b1X1 + b2X2+...+ bkXk (please keep in mind that I can't use subscripts here; and in your final model, you might substitute values from your regression table in for the intercept and the regression slopes). 2. Regarding your question about termination of iterations, there's no problem. SPSS uses maximum likelihood estimation to estimate model parameters when you are performing logistic regression. The idea is to try to estimate the population parameters that most likely gave rise to the observed data. There is no closed form solution (such as in the case of OLS regression) that will produce your estimates; so the program goes through (usually) several iterations to arrive at a solution. I hope this helps! Cheers!

[@elizabethjoy934]

@@mikecrowson2462 Thank you so much for the quick reply. It helps a lot. Yes in qn 1 it was about the log regression prediction equation.

[@elizabethjoy934]

I have another query as to is it necessary to check multicollinearity when using logistic regression analysis?

[@mikecrowson2462]

@@elizabethjoy934 Hi there. It's generally a good idea to check assumptions. As with linear regression, you also do not want to have collinearity among your predictors in logistic regression either. Cheers!

[@elizabethjoy934]

@@mikecrowson2462 Thank you. The PPT you have provided is very helpful to understand all concepts thoroughly. Thanks again.

[@psjmartinez]

That you Mike, you are very kind. The explanation is excellent. Just a question. Where can I find how to report the results?

[@harjaskaurji]

I tried the same method. When I clicked OK, it appeared 'The dependent variable has more than two non-missing values. For logistic regression, the dependent value must assume exactly two values on the cases being processed. Execution of this command stops.' Please tell the solution. What should I do now?

[@tasyanabila9013]

Hello thank you for the informations, I tried opening the spss file but it can't open. Can you help me?

[@m.p.kumara2738]

LOVELY PRESENTED.............

[@CarlosEObertoMachado]

Excellent explanation, expression of hard work; however, I must point out that in the third paragraph of slide no. 31 there is an error that is obviously unintentional due to the context of the explanation. For those students who would not pass the test, they were assigned a greater and equal sign, when they should have written less and equal. I hope it is corrected in the original Power Point. Greetings.

[@mikecrowson2462]

Hi there. Thank you so much for catching this! I apologize for the typo. That slide is updated based on your feedback! Best wishes!

[@Robt0805]

Thank you so much, incredibly clear and helpful! I was wondering if there is a way of doing binary logistic regression using categorical variables, covariates (as both shown in the video) but also using random variables (to control variation)? Thank you

[@abdulhamidadem7128]

Thank so much. It is really helpful. One question: what do we do or what it mean if the two classification tables are similar?

[@psygeek]

Great tutorial

[@mikecrowson2462]

Thank you! Cheers!

[@emmasplantz]

Thanks for the explanations ! My question is : how to compare two logistic regression models ? Do you simply compare the classification table and choose the model that has a higher percentage? thanks in advance.

[@juliuszulu2263]

Thanks doc, but where can I download the dataset and PowerPoint from? Thanks

[@mikecrowson2462]

Hi there. There are links in the video description. Cheers!

[@anamahmed3580]

Very useful video.

[@bruktawitbekele5747]

Thank you , it was helpful video.

[@Deborah28277]

My hero

[@elisabethstrijdhorst]

Dear mr Crowson, I watched the video and still have a question. I am doing Process Macro Model 1, with a moderator (age) and a binary outcome variable. I have a main effect of the Y and X, and an interaction effect of the moderator. But, when I perform the process macro with another moderator (sex), my main effect is different. Do you know why that is? I thought the main effect would not be different, because this is seperate from the moderator. I hope you understand my question. Thank you. Elisabeth