New MPLUS Fanboy - Do we Have Mplus users here?

M

Moanah

New Member
#1
#1
Hello everyone,

I was happy to find this forum. I am a MSc in Clinical Psychology and contrary to 99% of my colleagues I love research and doing empirical studies.

I am currently doing a PhD in Nottingham (UK) and I have published few papers on behavioural addictions, mostly on Internet and Online Gaming Addiction.

My main area of research is behavioural addictions but I have a passion for general psychometrics, especially developing measurement tools.

As the title says, I love Mplus. However I am still learning since I recently changed AMOS for Mplus (thanks to MLR estimation and many other cool things..).

How many Mplus users are out there?
 
spunky

spunky

Can't make spagetti
#2
#2
Moanah said:
How many Mplus users are out there?
Click to expand...
two. well, actually only one-and-a-half because i prefer the lavaan package in R. which is like Mplus... but better (and free).

welcome, btw. you'll find it fun around here? (FUN STUFF HAPPENS HERE!!!! XD)
 
J

jamesmartinn

Member
#3
#3
3 :)

I did my undergraduate training in Psychology (Neuropsych, Psychometrics) and graduate work in Biostatistics.

What I like most about Mplus is the support from Muthen & co. Anytime you have a question, they answer it directly themselves (and are pretty fast too).

Welcome!
 
Lazar

Lazar

Phineas Packard
#4
#4
Check out the text editor sublime text and the mplus plug in for it. It makes working with mplus 100 times better.
 
Lazar

Lazar

Phineas Packard
#6
#6
Whenever possible I use lavaan instead of mplus. Obviously the exception to this is multilevel SEM and mixture SEM. Everything else I know use lavaan. When lavaan integrates these I will not use mplus again.

Interestingly, we have been doing some comparisons between R and Mplus. For things like large EFA, multiple imputations, etc. R is so much faster it is hard to believe. Which makes no sense to me. Mplus should be faster given its fortran base but a large target rotation EFA can take a week in mplus and a few hours in R.....and yet both give the same results to the third decimal place.
 
Dason

Dason

Ambassador to the humans
#7
#7
Lazar said:
Whenever possible I use lavaan instead of mplus. Obviously the exception to this is multilevel SEM and mixture SEM. Everything else I know use lavaan. When lavaan integrates these I will not use mplus again.

Interestingly, we have been doing some comparisons between R and Mplus. For things like large EFA, multiple imputations, etc. R is so much faster it is hard to believe. Which makes no sense to me. Mplus should be faster given its fortran base but a large target rotation EFA can take a week in mplus and a few hours in R.....and yet both give the same results to the third decimal place.
Click to expand...

Maybe the BLAS you have with R is better than whatever Mplus is doing.
 
spunky

spunky

Can't make spagetti
#8
#8
Dason said:
Maybe the BLAS you have with R is better than whatever Mplus is doing.
Click to expand...
wouldn't be too surprised about that. Mplus has begun it's own brand of 'SPSS-ification' in order to cater to a wider audience, which means it'll very soon will trade efficiency for fancy GUIs that will clog it. (doesn't really have much to do with anything but i haven't spewed out any insanity this day so i felt like something was lacking)
 
Lazar

Lazar

Phineas Packard
#9
#9
You would think they are using the established fortran based BLAS and linpack right....I mean the very same ones that are underneath matlab and numpy.
 
You must log in or register to reply here.