question about free tekst results merging

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

studentrotterdam
Posts: 1
Joined: Wed Feb 15, 2012 6:06 pm

question about free tekst results merging

Postby studentrotterdam » Wed Feb 15, 2012 6:17 pm

I have the following problem:
in a research the use of medication in multiple patients is investigated. Every patient could list 3 different medicines in free text. Therefore, the data is shown for instance as follows:

patient medicine1 medicine2 medicine3
A valium diazepam zolax
B zolax valium asperin
C aspirin celecoxib oxynorm

How can I make a frequency-table that takes categories medicine 1, 2 and 3 together? I also want to be able to see that the combination on the first line (valium, diazepam and zolax) belongs to patient A, so my first question is: how can I take de frequencies of the 3 medicine coloms together, and secondly: how do I do this without losing the connection to the patient on the same line?

I have tried to do this with first recoding the free text into numerike data, but then I still can't figure it out..
GerineL
Moderator
Posts: 1477
Joined: Tue Jun 10, 2008 4:50 pm

Re: question about free tekst results merging

Postby GerineL » Thu Feb 16, 2012 2:30 pm

2 questions:

- approximately how many different kinds of medications did the patiens write down?
- are all medications spelled the same (e.g., not valium and valuim)?
studentrotterdam1
Posts: 1
Joined: Thu Feb 16, 2012 3:17 pm

Re: question about free tekst results merging

Postby studentrotterdam1 » Thu Feb 16, 2012 3:21 pm

There are approximately 35 different kinds of medication, which are also spelled differently, so valium could be spelled as 'valium' or 'vailium' or 'Valium'

Thanks for your fast reply!

Who is online

Users browsing this forum: No registered users and 1 guest

cron