Thank you very much! Your videos are unparalleled - nothing else about multilevel modeling in jamovi. However, i remain confused by the terminology. Level 1 is represented by the studentID and socioeconomic status is a Level 1 factor. Level 2 is represented by schoolID and school size and school type are level 2 factors. However, your analysis in this video focuses exclusively on Level 2 (schoolID) but takes a Level 1 predictor (socioeconomic status)....
Hi Mihai, you have to think about your data as falling at multiple levels. Since students are nested within schools, hence the student-level variables fall at level 1 within the data hierarchy , whereas any school-level variables would be considered as falling at Level 2. The outcome measured at Level 1 achievement is measured at the student level, however, it is related to the level 2 predictors via the random variation in intercepts/means across the Level 2 school units. I'm a bit confused by your question. However, it looks like you may be asking about the 'ses' and 'meanses' variables. Student ses was measured at Level 1 - that is the individual student level. The 'meanses' variable was an aggregate of individual student data within schools. Specifically, it was computed as an average level of student 'ses' associated with each school (so each school - level 2 unit, will have a school-average of student ses). In this way, we can model a school-context variable (average student ses) as a Level 2 (school-related) predictor of achievement. FYI, this is oftentimes done in multilevel models (where you aggregate individual data from Level 1 to Level 2) in order to capture potential between-school differences on the Level 1 predictor. So the 'meanses' predictor is a school-contextual factor that may account for variation in student achievement, apart from their individual level ses at Level 1 (and they do not necessarily have the same effect on school achievement) . The school size and type variables (both measured at Level 2) were not used in this video,but are addressed in the second video of this series: ruclips.net/video/qXgrWDkJ__c/видео.html . You might also consider looking at another video and related info in another video using SPSS: ruclips.net/video/x5Z5KYODwNk/видео.html
Hello Mike. Very informative video for learning multilevel modeling. However, I am a little confused with my model. I have measured my IV with 6 items and DVs with 3 and 5 items each, at organizational level by data collected from managers (N=67). my mediators is a five item construct measured with data from individual employees. my moderators is again organizational level dichotomous variable. I don't how to run this model. If you could help in this regard
Thank you so much! I am still wondering how to restructure my dataset, as I want to analyze longitudinal data that is in the wide format and needs to be transformed into long format. I don't have access to SPSS and am wondering if Jamodi oder JASP offer possibilities for restructuring. It seems quite complicated to me. Any advice would be much appreciated :)
![Summer Walker - Heart Of A Woman [Official Lyric Video]](http://i.ytimg.com/vi/SFddctbZzw8/mqdefault.jpg)
Much appreciated! Happy Holidays and best wishes in 2020!
You're very welcome, Yulin. You have a great holiday too! Best wishes!
Thank you very much! Your videos are unparalleled - nothing else about multilevel modeling in jamovi. However, i remain confused by the terminology. Level 1 is represented by the studentID and socioeconomic status is a Level 1 factor. Level 2 is represented by schoolID and school size and school type are level 2 factors. However, your analysis in this video focuses exclusively on Level 2 (schoolID) but takes a Level 1 predictor (socioeconomic status)....
Hi Mihai, you have to think about your data as falling at multiple levels. Since students are nested within schools, hence the student-level variables fall at level 1 within the data hierarchy , whereas any school-level variables would be considered as falling at Level 2. The outcome measured at Level 1 achievement is measured at the student level, however, it is related to the level 2 predictors via the random variation in intercepts/means across the Level 2 school units. I'm a bit confused by your question. However, it looks like you may be asking about the 'ses' and 'meanses' variables. Student ses was measured at Level 1 - that is the individual student level. The 'meanses' variable was an aggregate of individual student data within schools. Specifically, it was computed as an average level of student 'ses' associated with each school (so each school - level 2 unit, will have a school-average of student ses). In this way, we can model a school-context variable (average student ses) as a Level 2 (school-related) predictor of achievement. FYI, this is oftentimes done in multilevel models (where you aggregate individual data from Level 1 to Level 2) in order to capture potential between-school differences on the Level 1 predictor. So the 'meanses' predictor is a school-contextual factor that may account for variation in student achievement, apart from their individual level ses at Level 1 (and they do not necessarily have the same effect on school achievement)
. The school size and type variables (both measured at Level 2) were not used in this video,but are addressed in the second video of this series: ruclips.net/video/qXgrWDkJ__c/видео.html . You might also consider looking at another video and related info in another video using SPSS: ruclips.net/video/x5Z5KYODwNk/видео.html
Hello Mike. Very informative video for learning multilevel modeling. However, I am a little confused with my model. I have measured my IV with 6 items and DVs with 3 and 5 items each, at organizational level by data collected from managers (N=67). my mediators is a five item construct measured with data from individual employees. my moderators is again organizational level dichotomous variable. I don't how to run this model. If you could help in this regard
thank you sir. is it okay if ICC = 1.000 ?
Thank you so much! I am still wondering how to restructure my dataset, as I want to analyze longitudinal data that is in the wide format and needs to be transformed into long format. I don't have access to SPSS and am wondering if Jamodi oder JASP offer possibilities for restructuring. It seems quite complicated to me. Any advice would be much appreciated :)
Hi Dr. Mike,
Can Jamovi program be used as an alternative to HLM 6 or 7 softwares?
for individual employee (N=211)