Latent variable interactions in AMOS using double-mean centered product indicators

This video provides a general introduction on how you can use AMOS to model latent variable interactions. In particular, I demonstrate how to create product indicators using double-mean centering (see Lin et al., 2010), as well as how to include them in a path model involving a latent interaction variable. In addition to interpretation of results, I also discuss how you can probe a significant interaction using simple slopes tests (via the User Defined Estimands option in AMOS).
Download a copy of the Powerpoint referenced in the video here: drive.google.c...
Download a copy of the SPSS data from the video here: drive.google.c...
Download a copy of the AMOS model template I was working from in the video here: drive.google.c...

[@khaldounababneh8784]

Thank for spending your time and efforts in making all of your helpful videos. It is highly appreciated.

[@1jodiemartin]

Oh wow, you were my instructor at OU! What a find!

[@mikecrowson2462]

Hi Jodie, I remember you! Hope you are doing well!

[@yulinliu850]

Nice. Thanks for the sharing.

[@richiekaraburun1684]

Great one! Thanks

[@sonjarosy]

Thanks for the video and additional material! You mentioned in the powerpoint that the double mean-centering approach is preferable to Little's et al. indicator product residual-centering approach. If you don't mind answering - are there any cases in which you would yet prefer that residual centering method over the double-mean centering?

[@chakalalachico5487]

hey mr Mike, I would like to thank you for effort and time. I really appreciat it ! meanwhile I have a technical question and I need to know your opinion and I don't know where I can contact you ! thank you in advance.

[@jessperry1703]

Hi, do you need to perform boot-strapping for the simple slopes analysis? I have some missing data and not sure it will run otherwise...

[@jessperry1703]

Thank you so much btw!

[@hishonline]

Hello dr. MIKE
how can I do interaction with higher order latent variables?

[@jessperry1703]

Thank you so much for this video. Quick question regarding the simple slopes. The estimates that we get from conducting simple slopes through defining new estimands are unstandarized, right? How do we calculate the standarized simple slopes estimates?

[@mikecrowson2462]

Hi Jess, you are correct that these are unstandardized slopes. I haven't worked out details on getting standardized simple slopes yet. :) Cheers!

[@jessperry1703]

@@mikecrowson2462 Thank you very much for your reply! Do you have any guidance/ videos on plotting these simple slopes in excel/ spss? Your videos have been so immensely helpful.

[@sabrinahegner7367]

@@jessperry1703 Same here. I am looking for a way to plot these simple slopes. Do you have any recommendations?

[@jessperry1703]

@@sabrinahegner7367 hi Sabrina, I ended up doing the simple slopes analysis using spss macro as I wasn't sure with Amos. Sorry I couldn't be more help.

[@sabrinahegner7367]

@@jessperry1703 Thank you for getting back to me :)

[@jessperry1703]

Please help! Can you run interactions in a SEM with second-order latent variables?

[@jessperry1703]

By second-order latent variable, I mean a latent variable, which is explained by another latent variable.