Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Donald Spady <dspady@ualberta.ca> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: Extracting numbers from code |
Date | Mon, 12 Nov 2012 08:48:34 -0700 |
I am not sure if this is what you want but one approach I have used is to assign values in the following way. If you have 11 possible responses then Response 1 = 1 Response 2 = 2 Response 3 = 4 Response 4 = 8 all the way up to Response 11 = 2048 Thus if a person has responses 1 & 2, then they have a score of 3 if they respond to 1 & 4, then their score is 9 (basically) you sum the scores. Each sum is unique and from each sum you can determine exactly what a person responded to. Thus if the score sum is 525, then they had responses equivalent to a score of 512 + 8 + 4 + 1; there is no other way to get a score of 525. Does this help. Don On 2012-11-12, at 8:36 AM, sanja_lutzeyer@ncsu.edu <slutzey@ncsu.edu> wrote: > Good day. > I have received a dataset from a survey in which numerous questions > asked respondents to check all that apply. > As an example, one such question had 11 possible items that > respondents could check. > The data is coded in such a way that if a respondent checked item 2, > item 4, item 7, item 10 and item 11, then the response in the dataset > is coded 2471011. > I need to create dummy variables for each item. As such, I need to > create 11 dummy variables, which equal 1 if the respondent checked > that item. > The respondent above would have. > item1_dummy = 0 > item2_dummy = 1 > item3_dummy = 0.... etc. > > Until now, I have only encountered questions with up to 5 items, in > which case I manually coded them in the following way. > gen item1_dummy = 0 > replace item1_dummy = 1 if item == 1 | (item>= 10 & item<= 19) | (item >> = 100 & item <= 199) etc.... > This is horribly inefficient though. Is there another, easier way to > do this? Is there a way to code something like > item2_dummy = 1 if item contains the number 2. > Any advice would be greatly appreciated. > Kind regards > Sanja > * > * For searches and help try: > * http://www.stata.com/help.cgi?search > * http://www.stata.com/support/faqs/resources/statalist-faq/ > * http://www.ats.ucla.edu/stat/stata/ > Donald Spady -- dspady@ualberta.ca * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/