reshape command

[Pratibha Singh]

Hi stata user,

I have facing some problem while using reshape wide command, here is my data

. dataex

----------------------- copy starting from the next line -----------------------

* Example generated by -dataex-. To install: ssc install dataex
clear
input byte id str3(elligible diabetic) str2 cholestrol str3 stroke
1 "Yes" "No" "No" "Yes"
2 "Yes" "Yes" "No" "No" 
3 "No" "Yes" "Yes" "No" 
4 "No" "No" "No" "No" 
4 "Yes" "Yes" "No" "No" 
5 "Yes" "No" "No" "No" 
6 "Yes" "Yes" "Yes" "Yes"
7 "Yes" "No" "No" "Yes"
5 "Yes" "Yes" "No" "Yes"
end

------------------ copy up to and including the previous line ------------------


there is a duplicity in ID, I want unique id, so I tried reshape command which are as

bysort id:gen ab=_n

. 
. replace ab=0 if ab==.
(0 real changes made)

. reshape wide elligible diabetic cholestrol stroke, i(id) j(ab)
(note: j = 1 2)

Data                               long   ->   wide
-----------------------------------------------------------------------------
Number of obs.                        9   ->       7
Number of variables                   6   ->       9
j variable (2 values)                ab   ->   (dropped)
xij variables:
                              elligible   ->   elligible1 elligible2
                               diabetic   ->   diabetic1 diabetic2
                             cholestrol   ->   cholestrol1 cholestrol2
                                 stroke   ->   stroke1 stroke2
-----------------------------------------------------------------------------
end

now, variables converted into 6 to 9,

now there are 2 variable for one thing.

If I do this, output looks like:

gen elligible =  elligible1+ elligible2

. tab elligible

  elligible |      Freq.     Percent        
------------+-----------------------------------
         No |          2       22.22 
        Yes |          5       55.55      
  YesYes |          2       22.22      
------------+-----------------------------------
      Total |          9      100.00
end

please suggest what to do to convert into single variable without duplicity of ID.

[Nick Cox]

I don't follow what you want that is different from what you've done. But here is a small amount of extra technique, with incidental correction of typos.

tabm is from the tab_chi package on SSC, and must be installed before you can use it.

* Example generated by -dataex-. To install: ssc install dataex
clear
input byte id str3(eligible diabetic) str2 cholesterol str3 stroke
1 "Yes" "No" "No" "Yes"
2 "Yes" "Yes" "No" "No" 
3 "No" "Yes" "Yes" "No" 
4 "No" "No" "No" "No" 
4 "Yes" "Yes" "No" "No" 
5 "Yes" "No" "No" "No" 
6 "Yes" "Yes" "Yes" "Yes"
7 "Yes" "No" "No" "Yes"
5 "Yes" "Yes" "No" "Yes"
end

bysort id : gen ab = _n 

reshape wide eligible-stroke, i(id) j(ab)

egen eligible = concat(eligible?), p(" ")

tab eligible 

tabm eligible?

. tab eligible 

   eligible |      Freq.     Percent        Cum.
------------+-----------------------------------
         No |          1       14.29       14.29
     No Yes |          1       14.29       28.57
        Yes |          4       57.14       85.71
    Yes Yes |          1       14.29      100.00
------------+-----------------------------------
      Total |          7      100.00

. 
. tabm eligible? 

           |        values
  variable |        No        Yes |     Total
-----------+----------------------+----------
1 eligible |         2          5 |         7 
2 eligible |         0          2 |         2 
-----------+----------------------+----------
     Total |         2          7 |         9