Removing observation that contain certain words

[Nick Baradar]

I am trying to remove observations that contain several words, but I recieved a message about "invalid syntax". Can you please advise where the mistake is? I also tried dataex to list the observations but I recieved the following message: "data width (999 chars) exceeds max linesize. Try specifying fewer variables"

The code I am trying currently is the following:

local wordlist "refines" "enhances" "flourishes" "matures" "expands" "grows"


gen newvar = 1

foreach w in wordlist' {
    replace newvar = 0 if subinstr(lower_KeyDevelopmentSituation, "`w'") != 0
}

drop if newvar == 0

[Justin Niakamal]

looks like wordlist' should be foreach w in `wordlist' (notice the ` character). Also, subinstr() is used to incorrectly.

You could also try

drop if ustrregexm(lower(lower_KeyDevelopmentSituation), "refines|enhances|flourishes|matures|expands|grows")

[Nick Baradar]

looks like wordlist' should be foreach w in `wordlist' (notice the ` character). Also, subinstr() is used to incorrectly.

You could also try

drop if ustrregexm(lower(lower_KeyDevelopmentSituation), "refines|enhances|flourishes|matures|expands|grows")

Thank you very much, Justin! I just have too many words, I don't want to use all of them in the drop code; the 6 words that I provided are just for the sake of example. Could you please also help me by specifying what's wrong with the subinstr() function so that I can correct it?

Thank you very much in advance!

[Justin Niakamal]

subinstr() is used for replacement. Eg subinstr(some_var, " ", "_", .) would replace all spaces with underscores.

Try this for your problem

local wordlist   "refines" "enhances" "flourishes" "matures" "expands" "grows"

foreach w in "`wordlist'" {
    drop if ustrregexm(lower(lower_KeyDevelopmentSituation), "`w'")
}

[Nick Baradar]

subinstr() is used for replacement. Eg subinstr(some_var, " ", "_", .) would replace all spaces with underscores.

Try this for your problem

local wordlist "refines" "enhances" "flourishes" "matures" "expands" "grows"

foreach w in "`wordlist'" {
drop if ustrregexm(lower(lower_KeyDevelopmentSituation), "`w'")
}

Thanks a lot for the explanation, Justin!
I used to code that you suggested, but it deleted every observation I had, which cannot be true.