Resolving Data Inconsistencies in Survey Responses

[Santosh Dash]

Hi,

I am currently tabulating a string variable, and I've noticed an issue: the enumerator recorded "Don't know" in two different formats. I'm having trouble resolving this. I've attached the results for your reference.

 . tab vwsc_functionality

vwsc_functio |
      nality |      Freq.     Percent        Cum.
-------------+-----------------------------------
  Don't Know |         54        4.21        4.21
  Don’t know |         65        5.06        9.27
          No |        280       21.81       31.07
         Yes |        885       68.93      100.00
-------------+-----------------------------------
       Total |      1,284      100.00

I’m not sure how to fix this. I look forward to your guidance.

I have attached a screenshot for your reference also.

Thank you.

[Clyde Schechter]

Look closely at the apostrophe's in the two different versions of "Don't know." They are different. Then also look at the k's: one us upper case and the other is lower case.

As for fixing the situation, it depends a bit on information not available from what is shown. The first question is whether this variable is a string variable or is a value-labeled numeric variable. If it's numeric, just run -label list XXXX- (replacing XXXX by the name of the value label attached to the variable.) Then use -recode- or -replace- to change one of the numeric values of "Don't know" to the other. You might also then want to modify the value label to whatever spell-out of "Don't know" you prefer. (My personal preference would be Don't know.)

If it's a string variable, then just run

replace vwsc_functionality = "Don't know" if inlist(vwsc_functionality, "Don't Know",  "Don’t know")

To be sure that you properly capture the different apostrophe's in this command, it is best to copy/paste them from the -tab- output rather than trying to type them.

[Santosh Dash]

Clyde Schechter, Thank you for your response. I apologize for my delayed reply, and unfortunately, your previous suggestion didn’t help.

However, I found the solution in your other reply to this query:

https://www.statalist.org/forums/for...-%E2%80%99-and

  
 replace varname = usubinstr(var1, "`=uchar(8217)'", "'", .)

Thank you once again!