DOING PROC FREQ FOR TWO DIFFERENT TABLES

[bersys02]

Hello,

I have to do a freq for two variable but they are in different tables. Is there a way to do:

Proc freq data=table1 table2

     title "asdfg";

     table var1*var2;

run;

I want to do a frequency for those two variable but they are in different tables I created from the same dataset.

[PaigeMiller]

You have to combine the tables into a single SAS data set. You can do this in a DATA step using either MERGE or SET commands, depending on how the tables should be combined.

[bersys02]

I created two tables from the same dataset. The first I took 4 rows and the other one is only 1 column. I can't find a way to merge them because one is numeric and the other is character. Is there a way to merge it without same ID'S OR INDEX ?

[Tom]

---

@bersys02 wrote:
I created two tables from the same dataset. The first I took 4 rows and the other one is only 1 column. I can't find a way to merge them because one is numeric and the other is character. Is there a way to merge it without same ID'S OR INDEX ?

---

Does not make much sense. How many observations and variables did the original dataset have?  How many variables does the "first" generated dataset have?  How many observations does the "other" dataset have?  Why does whether the variable is character or numeric matter at all? PROC FREQ will not care.

 What did the original dataset look like? Why not just run PROC FREQ on that instead of the other datasets.  Or why not just run PROC FREQ twice, once on each of the two new datasets?

[ballardw]

---

@bersys02 wrote:
I created two tables from the same dataset. The first I took 4 rows and the other one is only 1 column. I can't find a way to merge them because one is numeric and the other is character. Is there a way to merge it without same ID'S OR INDEX ?

---

Subsetting your data into separate data sets is often a beginner error. Generally it is better to add a variable to indicate desired group membership and use that variable to create report tables grouped as needed.

[PaigeMiller]

---

@bersys02 wrote:
I created two tables from the same dataset. The first I took 4 rows and the other one is only 1 column. I can't find a way to merge them because one is numeric and the other is character. Is there a way to merge it without same ID'S OR INDEX ?

---

Why split the data set this way, and then try to recombine? Wouldn't it make more sense to not split the original data set and run the analysis on the original data set?

But in any case, we have reached the point where you need to SHOW US a portion of this data set so we can see what you are talking about.