labeling value multi_responses variable

[Thein Zaw]

Hi All ,

I am now labeling the variables : there is some issues for multivalues
I have data set like this variable : available water sources ? the options are as follow , allow multiple selection so responses in the data set are likes 1 3 ...4 5 .....1 2 3...... 178 , etc because they are selected by multiple options so how can label value those kinds of multiple response . please help .

1	Surface water (river, pond etc)
2	Rainwater
3	Borehole
4	Unprotected spring or well
5	Protected spring or well
6	Tanker truck or cart
7	Public tap
8	Piped into house
9	Bottled

[Joseph Coveney]

Is this variable a string data type, say, with multiple selections separated with a space? For example, did you mean to write "1 7 8" and not "178", or is it that there may be (is allowed) to have multiple choices present in the variable as run-together and not always space-separated?

[Fahim Ahmad]

Thein, if you your variable represents multi response question, it means respondents can choose more than one or all of the categories listed in your question.
then while entering your data in to Stata you must have as many variables as number of categories of your multi-response question (In your data set it must be 9 variables)
then define a new value label which includes all of them (lab def LABEL 1"Surface Water" 2"Rain Water" ... 9"Bottled"), then assign this value label to all of your variables.

[Thein Zaw]

Hi Fahim ,
Thanks for your explanation ,
You mean firstly doing (lab def water_source 1 " " 2" " ....9" ) and then (lab val q_31 water_source ) . I already did like that but this is not work .These are all strings and one column have more than one responses (multiple response) , do I need any preparation before this step ? .Please see like this my data set :

q_31
2 3 7
3 7 8
2
8
2 9
3
1 5
4 7

[Nick Cox]

You can't label string variables in Stata. For discussions of this problem of multiple responses to survey questions see e.g. http://www.stata-journal.com/sjpdf.html?articlenum=pr0008 and http://www.stata-journal.com/sjpdf.h...iclenum=st0082

Your existing variable can be used: search this forum for mentions of tabsplit from tab_chi (SSC) but I agree with Fahim that also using separate indicator variables would be a good idea.

forval j = 1/9 { 
       gen q_31_`j'  = strpos(" " + q_31 + " ", " `j' ") > 0 
}

The leading and trailing spaces are needed for more challenging examples of this kind. In your case, this works

clear 
input str5 q_31 
"2 3 7"
"3 7 8"
"2"
"8"
"2 9"
"3"
"1 5"
"4 7"
end 

forval j = 1/9 { 
    gen q_31_`j' = strpos(q_31, "`j'") > 0 
} 

list

Please supply data examples as requested in FAQ Advice #12.

[Thein Zaw]

Hi ,
I am a new leaner so I have so many things to know about stata , the question : q_31 .available water sources ?
The options are :

1	Surface water (river, pond etc)
2	Rainwater
3	Borehole
4	Unprotected spring or well
5	Protected spring or well
6	Tanker truck or cart
7	Public tap
8	Piped into house
9	Bottled

Actually in my data set presents like this

q_31

q_31_1

q_31_2

q_31_3

q_31_4

q_31_5

q_31_6

q_31_7

q_31_8

q_31_9

q_31_10

q_31_11

q_31_12

q_31_13

q_31_88

2 3 12

FALSE

TRUE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

3 7 12

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

2

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

8

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

2 12

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

3 88

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

I would like to make label value ,

Please help me this ...

[Fahim Ahmad]

Please have a look on FAQ on how to post an example of your data-set, anyway send me your data-set through this email ([email protected]) to look on it.

[Joseph Coveney]

the question : q_31 .available water sources ?
The options are : . . .

I would like to make label value

If the options were as you show (only 1 through 9), then how did you get values of 10, 11, 12, 13 and 88? (And 178?) What are the labels for those values?

[Thein Zaw]

Sorry , I make missing copy ... here is correct options :q_31_What different kinds of disasters and emergencies can you think of?
The options ;

1	Landslides
2	Flooding
3	Typhoons
4	Drought
5	Storm surge
6	Volcanic eruptions
7	Earthquakes
8	Infectious Disease Epidemic
9	War/violence/conflict
10	Displacement
11	Mass Exposure to Toxin/Chemical
12	None
88	Other, specify

In data set , present like this ;

_id

q_31

q_31_1

q_31_2

q_31_3

q_31_4

q_31_5

q_31_6

q_31_7

q_31_8

q_31_9

q_31_10

q_31_11

q_31_12

q_31_13

q_31_88

6832555

2 3 12

FALSE

TRUE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

6832556

3 7 12

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

6832557

2

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

6832559

8

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

6832560

2 12

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

6832561

3 88

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

6832562

12 88

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

TRUE

6832563

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832564

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832565

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832566

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832569

2 4

FALSE

TRUE

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

6832570

88

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

6832573

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832574

2 12

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

6832575

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832576

1 88

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

6832577

13

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

6832578

2 12

FALSE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

TRUE

FALSE

FALSE

6832579

1 2

TRUE

TRUE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

FALSE

6832612

2 5 8 12 88

FALSE

TRUE

FALSE

FALSE

TRUE

FALSE

FALSE

TRUE

FALSE

FALSE

FALSE

TRUE

FALSE

TRUE

How can I do label value for those kind of multiple response ?
Help me .

[Joseph Coveney]

You might want to check again: what's 13?

Anyway, you'll need to say just what it is that you're after. What do you want things to look like when you're done? I doubt that you want blanks wherever there is "FALSE" and a value label for wherever there is "TRUE" . . .

[Thein Zaw]

Hi Joseph ,
13 Means 1 3 . Because these all data are downloaded from KoBo data collection apk . We have used KoBo application for data collection.

[Fahim Ahmad]

Thein Zaw, I have send you the commands in your e-mail, check your inbox.

[Enrico Zorzi]

Hi Thein Zaw clear input str10 q_31 "2 3 7 88" "3 7 8" "2" "8" "2 9" "3" "1 5" "4 7" end split q_31 drop q_31 destring q_31*, replace lab def LABEL 1 "Landslides" 2 "Flooding" 3 "Typhoons" 88 "Other, specif" lab val q_31* LABEL list

[Enrico Zorzi]

I am sorry for mistake

clear 
input str10 q_31 
"2 3 7 88"
"3 7 8"
"2"
"8"
"2 9"
"3"
"1 5"
"4 7"
end 
split q_31
drop q_31
destring q_31*, replace
lab def LABEL 1 "Landslides" 2    "Flooding" 3 "Typhoons" 88 "Other, specif"
lab val q_31* LABEL
list