Hello,

First of all R is a general purpose programming language with statistical application, in continue development, versatile, with lots of packages addressing a lot of topics and, more important, is open source - so you may have virtually unlimited access to all its developments and packages - while Mplus, as far as I know, is payware, with dedicated applicability (see specs on their website).

As human interface, Mplus is a combination of graphical input and command-line, while R is command-line only (if you don't consider rattle or some other packages to have a kind of GUI, with limited applicability) - which can be a bit discouraging for the new users with lack of programming skills.

This should not discourage you, however, since R has the largest number of hits on Google on related topics and growing,  meaning that there is a huge community of R users with various skills and you will easily find help, tutorials as long with ideas and solutions for your problems. The learning curve in R is quite steep, but not unbearable. In the end you should not learn everything at once, just which tools are suited for your problems and how to apply them.

Next, considering that you will find in both all the methods you need for your research, you should take into consideration what you will do in the future? If  whatever software (R, Mplus, Stata, SAS, SPSS etc. etc.) offers robust solutions and proved to be a good tool in your field, has good maintenance and, important!, it's largely in use in your (target) area, so it's worth learning and using it.

Coming to "power" the discussion is a bit off. There are tools performing better on some things (problems, machines etc.) and worse on others. The "power" is subject to implementations, problems, methods, machines and so on.

Good luck!

Dear Leng,

you have some information in internet:

https://www.statmodel.com/usingmplusviar.shtml

see the attachment file

bests

LM

Hello,

First of all R is a general purpose programming language with statistical application, in continue development, versatile, with lots of packages addressing a lot of topics and, more important, is open source - so you may have virtually unlimited access to all its developments and packages - while Mplus, as far as I know, is payware, with dedicated applicability (see specs on their website).

As human interface, Mplus is a combination of graphical input and command-line, while R is command-line only (if you don't consider rattle or some other packages to have a kind of GUI, with limited applicability) - which can be a bit discouraging for the new users with lack of programming skills.

This should not discourage you, however, since R has the largest number of hits on Google on related topics and growing,  meaning that there is a huge community of R users with various skills and you will easily find help, tutorials as long with ideas and solutions for your problems. The learning curve in R is quite steep, but not unbearable. In the end you should not learn everything at once, just which tools are suited for your problems and how to apply them.

Next, considering that you will find in both all the methods you need for your research, you should take into consideration what you will do in the future? If  whatever software (R, Mplus, Stata, SAS, SPSS etc. etc.) offers robust solutions and proved to be a good tool in your field, has good maintenance and, important!, it's largely in use in your (target) area, so it's worth learning and using it.

Coming to "power" the discussion is a bit off. There are tools performing better on some things (problems, machines etc.) and worse on others. The "power" is subject to implementations, problems, methods, machines and so on.

Good luck!

Hi Leng,

The answers already posted are already very informative, as they highlight the flexibility of R compared with MPlus. Note that this is not intended to counter those points, which I agree with as a user of R myself. However, you should note that there are some areas that this flexibility does not extend to, at least not yet.

As R requires development of specialised packages, some techniques are not available there until they become more commonly used. Currently this includes Marsh et al.'s (2014) Exploratory Structural Equation Modeling (ESEM), which is available in MPlus. This integration of traditional EFA and CFA is seeing increased usage in scale development, so it is worth checking to see whether you are likely to employ this technique.

If this technique becomes much more popular, there is no doubt that someone will write an R package for it. I believe efforts are already underway and I would be surprised if it isn't included in William Revelle's psych package within the next two years; however, you might not be able to wait for R if this is something you wish to explore. This may depend on the expected dimensionality of your measure and whether it will be important to establish the presence of scales/subscales. I would recommend you check the techniques that you might use. Of course, the same applies the other way round - there are lots of analyses that R will allow you to carry out but MPlus won't.

Best wishes

Marsh, H.W., Morin, A.J.S., Parker, P.D., & Kaur, G. (2014) Exploratory Structural

Equation Modeling: An Integration of the Best Features of Exploratory and Confirmatory Factor Analysis. Annual Review of Clinical Psychology, 10, 85-110.

R is a programmiing environment  for statistic and graphics, it is open source supported by a world wide research community with more than 6700 packges , it can work in any platform(windowse, mac, linux..), R also one of the most popular specialized programming languge

R studio is an IDE for R which makes R much easier

her is a very useful linke for R resources

http://www.computerworld.com/article/2497464/business-intelligence/business-intelligence-60-r-resources-to-improve-your-data-skills.html