# tabstat _all, ignoring strings

#### Soporiferous

##### New Member
Hey all,

I'm working with a large dataset, containing both numerical and string variables. To get a quick overview of means, range etc. of my numerical variables, I wanted to run tabstat _all with an option to ignore the string variables. Is there a way to do this? If not, what is the easiest way around it? Should I just destring everything, run the tabstat and then restring again? Or is there a way to sort variables by var type and then specify the tabstat for the numerical part of the list? Thanks in advance!

Best,

Viktor

P.S. I know about the summarize function, but it doesn't offer all the stats I'm looking for

#### xbender

##### New Member
If you are planning on using the large set of variables over and over, I recommend to create a local macro (see macro or Stanford manual). You can write
Code:
.local controls age agesq education income
which will create local macro "constrols" and then you use that when refering to variables - eg.
Code:
.regress outcome treatment controls'
you can also you an asterix to mark all variables beginning with certain string - eg. when you have variables (avgp1 avgp2 avgp3 avgp4 total), you can use
Code:
.regress total avgp*`