Dyadic Data Analysis in SPSS and R

[@Tomahawk1999]

Thank u Dr Wess for a wonderful video. we need more of these get down in the ditch and show the data file and do analysis kind of videos. just going all theory upfront is not just confusing and boring, its downright demotivating and instills no confidence at all. thanks look forward to more please!

[@choheeshrader9843]

Yaaas! Science Queens! Thank you! This is incredibly helpful! Thank you for advancing science and making it accessible and digestible!

[@echowang4774]

Thank you so much! it really helps me understand the dyadic data.

[@andyk5802]

Fantastic presentation! Thank you so much

[@mojimshahvali4053]

Thanks for the video. Question. How are t-tests run differently for dyadic data? In other words, how do we control for the dependency of our data before running paired t-tests.

[@avshalomkoren9104]

very good presentation! Thank you! Is there a way to get access to this dataset (TSD_person_pairwise.sav) for practicing and training?

[@LudwigVera]

This is so great, thanks so much! If the same participant took part in several dyads (i.e., each participant is part of several dyads, because they sequentially get paired with new partner, akin to speed dating), how would I change the model? (include participant as a random factor, too?)

[@Paolo-tu2ft]

Hello Dr Wess. Brilliant video. That helped me a lot. Just a question: I have dyadic data with two outcomes, one for the patient, one for the caregiver. How do I compute the effects of patients predictors on their own scores and then on the caregivers scores? Thanks in advance! Paolo

[@oanadanieladumitru2134]

Hi Paolo, to get at your question, you have to use the Actor-Partner Interdependence Model (you can find more information about it online or on Dr. David Kenny's personal website). You'll have to structure your data in a person period pairwise format, such that every row has the actor (e.g. caregiver) variable and the corresponding partner (e.g. patient) variable. We have a video on restructuring data in our playlist on here. Hope this helps!

[@Paolo-tu2ft]

@@oanadanieladumitru2134 Very clear thanks Prof!

[@Paolo-tu2ft]

Hello Dr. Wess, just one more question. I had an example of an analysis that was conducted on HLM; it was basically an APIM model with more than one IV (10 in total). Thanks to your suggestion I am now able to conduct this type of analysis in SPSS, however, the paper tells about this tau correlation, defining it as the interdependence of the outcome within-dyads. Is this the same as the CSH rho you obtain in the SPSS output? In case it is not, how can we obtain it with SPSS? Thanks!

[@diverseflavors5634]

Hi Dr Wess.. There is a question if the respondents in a dyad respond on different questionnaires...how to you make a data set in that case?

[@oanadanieladumitru2134]

You can easily merge two datasets in SPSS or R. If the two dyad members answer the same questions on different questionnaires, you would have to make sure the variable names are the same and you would use the "add cases" function in SPSS. If they answer different questions, you would use the "add variables" function and it would create different variables. Hope this helps!

[@diverseflavors5634]

@@oanadanieladumitru2134 thanks a lot...in my research project...I have a standard dyad (1 on 1). I have asked and member of the dyad about rating the other's creativity. While the other member has been asked regarding rating the other's leadership. Both different surveys with different number of questions. I wondered how I can manage both in the data sheet.

[@oanadanieladumitru2134]

@@diverseflavors5634 sorry for the delayed response! in that case, you can just merge the data sets and (in SPSS) use the "Add variables" function. You'll want to merge the smaller data file into the larger one.