Hello,
I have millions of records with an alcohol variable (drinker30d) as drinking in last 30d as YES or NO. I want to convert the "drinker30d" into "drinker365d" using another survey's drinking distribution %. This imputation has to be done "by chance". For the drinker365d I have a distribution as 10% abstainers, 10% former, 20% non-30d drinkers, 60% 30d drinkers.
I have given a smaller sample of 20 and i have drinker30d data and I want to achieve drinker past year. Basically, i want to convert the 0 in drinker30d into either 1 or 2 or 3 by chance given the distribution of 1/2/3 (as10%, 10% 20%). Drinker365d should look like the 3rd column.
I am not sure how I can use random function in SPSS. Please help!!!
| id |
drinker 30d |
drinker past year |
| 14 |
0 |
1 |
| 18 |
0 |
1 |
| 16 |
0 |
2 |
| 4 |
0 |
2 |
| 10 |
0 |
3 |
| 20 |
0 |
3 |
| 15 |
0 |
3 |
| 17 |
0 |
3 |
| 2 |
1 |
4 |
| 5 |
1 |
4 |
| 8 |
1 |
4 |
| 9 |
1 |
4 |
| 6 |
1 |
4 |
| 19 |
1 |
4 |
| 12 |
1 |
4 |
| 7 |
1 |
4 |
| 11 |
1 |
4 |
| 13 |
1 |
4 |
| 1 |
1 |
4 |
| 3 |
1 |
4 |
|
|
|
|
|
ASSIGN BY CHANCE |
| 10% |
abstainer |
1 |
| 10% |
former |
2 |
| 20% |
non-30d drinker |
3 |
| 60% |
30d drinker |
4 |
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
