No, the sample size does not need to be equivalent. What you need is enough observations per parameter to produce stable estimates, so that depends on how complex your model is.

FYI you will find various statements about how many observations you need per parameter -- starting with a minimum of 5.

The sample size estimation depends on your information about population, is it finite population? or infinite population.

In the case of finite population, you can use the attached table file to estimate the suitable sample size.

But in case of infinite population: 

The sample size for any study depends on the standard deviation of the variable ( from previous studies ) and the margin of error you decided . Although there were many online sample size calculators and many software have an option to calculate sample size, but I preferred to use G*Power .

I think that G*Power software is effective tool to calculate sample size for a different range of experiments. G*Power is free to download and easy to use after reading the manual, the download link:

http://download.cnet.com/G-Power/3000-2054_4-10647044.html

The link to download G*Power manual:

http://www.gpower.hhu.de/fileadmin/redaktion/Fakultaeten/Mathematisch-Naturwissenschaftliche_Fakultaet/Psychologie/AAP/gpower/GPowerManual.pdf

Good Luck

Thank you very much Khalid. Your information was really useful.

With best regards,

Filipa

My pleasure to help you.

Good Luck

Dear Khalid,

Many thanks once again. In addition, do you know how to perform multi-group analysis in Mplus? Because for my comparative study, I need to perform it. Once again, many thanks for your help,

With kind regards,

Filipa

Dear Filipa

I don't have experience with Mplus, I prefer SAS and SPSS for statistical analysis, this software effective and widely used.

Good Luck