Creating a dummy that detects any string variable

[Yoonhee Choi]

Dear Statalists,

My data set has a "name" variable, which is a string variable with individual names. However, there are many missing names in the list (i.e., zero string). I would like to create a dummy variable that shows 1 for any name, and 0 for missing. How can I create a code for that?

I tried
gen namedummy=.
replace namedummy =1 if name==" "

I got "0 changes made".

[Joseph Coveney]

generate byte missing_name = missing(name)

or

generate byte missing_name = name == ""

[Clyde Schechter]

If you have faithfully transcribed your code here, you went wrong with the name==" " part. The "zero" string is not a blank space, it is a null: so you want name == "".

However, a simpler approach is:

gen namedummy = !missing(name)

Note also that the code I'm showing here will generate a 0/1 variable (which is what you said you want), whereas your code, had it worked at all, would have produced a ./1 variable. ./1 variables in Stata are a recipe for errors down the road. Stata's logical operators are nicely equipped to deal with 0/1 variables. With ./1 variables, not so much, because . and 1 are both taken to be true in boolean expressions.

Added: Crossed with #2 which makes the same points.

[Chris Larkin]

Your code would normally work if the string is truly empty (even though it would give you a 1 for a missing value of name and 0 for a non-missing value). There might be some spaces in your name variable that you can't see though. Try trimming it first, like this:

* Example generated by -dataex-. To install: ssc install dataex
clear
input str21 notaname
" "                    
"Cdyqbqasai Austjfaisv"
"Saaxulzxlw Awxfytmqqf"
"Ogkkhudbce Ssdbgqpqov"
"Rhtckemioi Twdtmpwsuu"
"  "                    
"Bmodqoxbuv Nhuanpcrao"
"Nynquntmaj Waydyabmzh"
"Epwuvjxpoo Bxqjsrtqtm"
"Plfgniwolf Wkbeflufyp"
"   "                    
"Sobjrjoufv Hdlpucvbbh"
"Njcresrrzj Cejliozfoq"
"Zlijoquynj Dzlmoeacqs"
"Zyldxjdnkw Netofygmac"
"    "                    
"Zvextwaenc Njwyedjjsj"
"Qeyppoxepu Spmdxwncfw"
"Fedubvykbi Cmazwhkwjg"
"Qdmkzygzcx Kdugzwoeqi"
"     "                    
"Ivkcpfiecw Jzxjmddjyo"
"Jnlmdqzdxm Tajwbuysxc"
"Cuattqvdrh Qcwvyismhd"
"Numixddstu Vgrirsplaj"
"      "                    
"Oatrmputvx Fgckpduuab"
"Uehcrfsjwa Bjevpohrdo"
"Yhjjqhwmeb Gnjtaxrrws"
"Ndaokzugot Gvzxrqeiap"
"       "                    
"Yzfaclhbrw Rvoqdvnvmt"
"Zwyjawcxwm Bcemkeacoa"
"Svlnkzodvd Ozfjtpjqut"
"Predlgxysx Exayslsjth"
end

replace notaname=trim(notaname)
gen namedummy = notaname!=""

If that doesn't work you should share a snippet of your data using dataex (if you don't have it already, type ssc install dataex to install it and then read help dataex to learn how to use it)

[Yoonhee Choi]

Thank you everyone. I tried Clyde's code, and it works great!