Splitting a string variable delimited list into individual binary variables in SPSS

Question

I have a string variable created from a checkbox questions (Which of the following assets do you own?)

I am trying to create individual binary variables for each type of asset based on whether that number is present in the string list.

The syntax I am using cannot differentiate between 1 and 11.

do repeat wrd="1," ",2," ",3," ",4," ",5," ",6," ",7,"/NewVar= W3_CG_asset_TV_1  W3_CG_asset_radio_2 W3_CG_asset_payTV_3 W3_CG_asset_tel_4 W3_CG_asset_cellphone_5 
     W3_CG_asset_fridge_6 W3_CG_asset_freezer_7.
 compute NewVar=char.index(W3_CG_HouseExpen1, wrd)>0.
end repeat.

do repeat wrd= ",8," ",9," ",10," ",11," ",12," ",13," ",14," ",15," ",16," ",17," ",18," ",19," /NewVar= W3_CG_asset_electricstove_8 W3_CG_asset_primusstove_9
    W3_CG_asset_gasstove_10 W3_CG_asset_electrickettle_11 W3_CG_asset_microwave_12 W3_CG_asset_computer_13 W3_CG_asset_electricity_14 W3_CG_asset_geyser_15 
    W3_CG_asset_washingmachine_16 W3_CG_asset_workingvehicle_17 W3_CG_asset_bicycle_18 W3_CG_asset_donkeyhorse_19.
 compute NewVar=char.index(W3_CG_HouseExpen1, wrd)>0.
end repeat.

Answer 1

I have tested this on SPSS 28. make sure the column W3_CG_HouseExpen1 is string and length of it long enough to hold the data.

Then I added execute

data list list/W3_CG_HouseExpen1 (a50).
begin data
"1,2,11,12,"
"2,12,"
"1,2,"
"1,11,12,"
end data.

do repeat 
wrd="1," ",2," ",3," ",4," ",5," ",6," ",7," ",8," ",9," ",10," ",11," ",12," ",13," ",14," ",15," ",16," ",17," ",18," ",19,"
/NewVar = W3_CG_asset_TV_1 W3_CG_asset_radio_2 W3_CG_asset_payTV_3 W3_CG_asset_tel_4 W3_CG_asset_cellphone_5 W3_CG_asset_fridge_6 W3_CG_asset_freezer_7
      W3_CG_asset_electricstove_8 W3_CG_asset_primusstove_9 W3_CG_asset_gasstove_10 W3_CG_asset_electrickettle_11 W3_CG_asset_microwave_12 W3_CG_asset_computer_13 W3_CG_asset_electricity_14 W3_CG_asset_geyser_15
     W3_CG_asset_washingmachine_16 W3_CG_asset_workingvehicle_17 W3_CG_asset_bicycle_18 W3_CG_asset_donkeyhorse_19.
compute NewVar=char.index(W3_CG_HouseExpen1, wrd)>0. 
end repeat.
EXECUTE. 

Answer 2

My suggestion is to run through this in reverse, erasing the values you've already recognized. So if you've got "11" and erased it, when you later search for "1" you won't find it in an "11".

I recreated a tiny exaple dataset to demonstrate on (EDIT-improved example):

data list list/W3_CG_HouseExpen1 (a50) .
begin data
"1,2,11,12,"
"11,12,"
"2,11,"
end data.

Now I do the whole process on a copy of the original W3_CG_HouseExpen1 variable so I can eat it away without damage to the original data:

string #temp(a50).
compute #temp=W3_CG_HouseExpen1.
do repeat wrd="12," "11," "2," "1,"  /NewVar= W3_12  W3_11 W3_2 W3_1.
    compute NewVar=char.index(#temp, wrd)>0.
    compute #temp=replace(#temp, wrd, ""). /*deleting the search string from the full string.
end repeat.
exe.