How to handle multiple response questions
May 2nd, 2006
In a lot of research, multiple answers can be given to a single question. For example:
“What kind of food do you like?”
o soup
o rice
o salad
How do you analyse this type of question? Let’s assume you want to make a table with the answers. How do you combine them? Even though it may seem like the most easy thing to do, this is pretty difficult stuff. SPSS is good at analyzing unique combinations of variables (answers to questions) combined with unique cases (people in a survey). The combination of more than one answer per person, does not fit into that logic. So we have to be a little creative.
There are – at least – four different ways to analyze these results. They all have advantages and disadvantages, so it really comes down to your preferences:
1) Create separate variables for each answerÂ
varsoup (0 = not ticked, 1 = ticked)
varrice (0 = not ticked, 1 = ticked)
varsalad (0 = not ticked, 1 = ticked)
If you’re not entering the data yourself, chances are that this is what your data-set will look like if there was a multiple question in the survey. With three separate variables, you can create three tables using the frequencies-command. The advantage of this approach is its simplicity, the disadvantage is that you have three separate tables. You cannot tell if the three answers were in any way linked to one another.
2) Make a ‘grouping variable
a. Follow the steps in option 1, creating separate variables for each answer
b. Create a new variable: the ‘grouping variable’. Select analyze > tables > multiple response sets… Select all variables from the list you want to group together and click them into the right window ‘Variables in set’. Choose the ‘counted value’: this is the value that you want to count as ‘yes’. In our example, this is the value 1. Give the new variable a name, for instance ‘varfood’. Click ‘add’, and ‘OK’. Your new grouping variable will appear with a $ in front in de list to the right of the screen: ‘$varfood’. The $-sign tells you this is a variable containing several variables.
To make a table, select analyze > tables > multiple response tables… In the window to the bottom-left of the screen, you see grouping variables. Click the one you want and click it into the ‘Rows’ box. Click OK and you’re done.
3) Several variables with a hierarchy
In order to be able to use this method, it would be nice if your survey also asked to rank the three items available. If that’s not the case, you will have to decide for yourself which answer is most important and label that one first answer.
This will lead to:
varfood1 (1 = soup, 2 = rice, 3 = salad) label “First answer”
varfood2 (1 = soup, 2 = rice, 3 = salad) label “Second answer”
varfood3 (1 = soup, 2 = rice, 3 = salad) label “Third answer”
4) Create a variable with a single value for each possible combination
This will lead to:
varfood
1 = soup,
2 = rice,
3 = salad,
4 = soup and rice,
5 = soup & salad,
6 = rice and salad,
7 = soup and rice and salad
The disadvantage is that this can be quite some work when you’ve got more than three answers. And you also run the risk of not being able to interpret your resulting table at a glance: the number of cases per option can be quite small. This is really only a serious option if you want to know exactly what each person answered. Pay attention to the order of the answers: if you start off with those combinations that were pretty popular, they will be at the top of your table, making it easier to interpret the results.
(Thanks to Sander for answering this question)
Entry Filed under: 0. SPSS, where to start,Questions and answers
36 Comments Add your own
1. Martijn | July 21st, 2006 at 3:06 am
A useful overview. However, it would be nice if the 5th option could be removed, as it is an exact copy of the 4th option.
2. Patty | July 25th, 2006 at 7:46 pm
I have a similar problem.
let’s say I have a combination of
br21= 0 equals no, 1 equals YES
br22= 0 equals no, 1 equals YES
br23=0 equals no, 1 equals YES
br24=0 equals no, 1 equals YES
br25=0 equals no, 1 equals YES
I would like the frequenies for all the combinations
there are about 25 different combinations and I don’t want any duplications.
please advise.
3. tan | August 10th, 2006 at 9:00 am
How do you exclude multiple responses, to just keep single responses? Options 4 and 5 are not efficacious when working with large datasets and exponential combinations. Any ideas?
4. andris | August 15th, 2006 at 5:12 am
@Martijn, thanks for your tip. It’s edited out.
5. andris | August 15th, 2006 at 5:13 am
Patty and tan, could you maybe clarify a little bit more what you mean?
6. kk | October 2nd, 2006 at 11:00 pm
when doing multiple response, the frequencies are based on the total no. of answer by the whole sample size, say if i have 28 respondents, they could choose up to 3 options, but some just choose one, so the N may vary according to the no. of answers the respondents given. is it still ok to do the test and anlaysis on the percentage? If not, what else can i do?
hope i have make it clear, thanks!
7. Ahad | February 20th, 2007 at 6:09 am
Superb! This tut is a life saver, thanks!
8. Nicolas | May 17th, 2007 at 9:33 am
I want to follow the 4th option. use 1 variable with many possible answers. my question is how do i put the answers in one cell of the SPSS. for example: somebody answered soup and rice. do I put in the cell of SPSS 1,2 (separated with commas?)or how else?
9. murad | May 23rd, 2007 at 1:39 am
Dear Nicolas
first find all Possible compenations which is equal to 7 then start coding them from 1 to 7 and enter your data as the codes of these combenations
10. Lisa | May 24th, 2007 at 3:39 am
I’m not sure if someone has already answered this but say you have a question which says tick all answers which apply….how can that be put on SPSS and analysed?
Thankyou
11. andris | May 31st, 2007 at 3:52 am
Lisa, this article explains four ways to put the MC question in SPSS and analyse it.
12. Mange | December 10th, 2007 at 9:47 am
Is´nt there any easier way to do this? really oldschool for a helpful statistic program like SPSS. Feels like I´m messing up a very nice db with this, but I cant find any other solution, soo… Hopefully the developers of SPSS fix this in a near future.
13. Stan van den Heuvel | December 11th, 2007 at 6:56 am
Hi,
i’m having problems defining the multiple resonse set. It works as described in step 2. Only i cant make the table. It generates one automatically when you’re done,. But it displays nothing.
Also i cant find the option “mutlitple response tables”. I hope someone can help me, please reply by email.
i used the website http://www.thesistools.com so when you have multiple response questions the website divides the answers in different cells and different answers so suppose someone answered option 1, 2 and 3. You get the answers in three cells and with the numbers 1,2 and 3. I changed the 2s and 3s in 1s and tried step 2 but it partially worked.
14. Stan van den Heuvel | December 11th, 2007 at 7:05 am
i use version 14.0 and Vista btw. It states on the website that it wont work, but im getting data en tables if i want. Maybe this has something to do with it?
15. andris | December 12th, 2007 at 4:55 am
Hi Stan,
Please use the SPSSforum.com to ask your question.
16. Pratik | March 25th, 2008 at 6:02 am
Thanks a lot! This forum saved my life
17. Shwetha | May 5th, 2008 at 1:31 am
I want to follow the 4th option. use 1 variable with many possible answers. my question is how do i put the answers in one cell of the SPSS. for example: somebody answered soup and rice. do I put in the cell of SPSS 1,2 (separated with commas?)or how else?
18. Simon | May 19th, 2008 at 8:53 am
If I have an output file for mulitple responses which generates a table in the output viewer and I want to use the same script that this generates again for the same question with different answers. How do I use the output script as an input script???????
19. Thayaparan | October 29th, 2008 at 12:31 am
dear sir plse explain how do we analyze the multiple response analysis in SPSS to get the regression?explain about the multi classification analysis and multi classification analysis?
20. arif7d | November 24th, 2008 at 2:36 pm
nice tricks to deal with the situation. I was having this problem and earlier used first option to handle it, but after visiting this thread, I think I will give method 4 a try. seems better to me in concluding results. thanks for helping out.
regards,
arif [Pakistan].
21. GEORGE | February 24th, 2009 at 3:51 pm
Can you please tell me how to sort data in order to help myself with “the hierarchy” of responses to one question? to be more specific: perhaps the question is: What are the problems with your internet connection? and the participants reply from a list (i.e. 1. bad signal, 2. low speed, 3. busy server etc). But the problem is when you say to them to actually write down their type of problem but in a way of ascending importance for example someone puts more serius the signal problem and then thew rest, while another chooses the server problem to be moreimportant to him…. How do I analyze this data. I have to come to a conclusion to write down in some order of importance (hierarchy) all the problems that my participants pick from the list… (my real list has something like 13 different cases!!!)
22. Veronica | March 13th, 2009 at 11:18 am
Cool~ I’d like to know if I could present the result by using a graph like bar chart..
23. Anne | July 6th, 2009 at 9:48 pm
George, I have the same problem with you?
Do you know how to rank multiple responses?
Tx
24. Lauren | October 28th, 2009 at 2:38 pm
Very Useful!
Thank you!
25. stella | November 4th, 2009 at 5:57 am
I have a question where I ask teachers to note three strategies they use in their lesson and rate their effectiveness from 1-5 (1=V.Ineffective- 5=V. Effective)
How do I imput the responses in the SPSS of questions such as this
26. Said Qasim | November 23rd, 2009 at 7:57 pm
What to do with a multipple response table with several interconnected things, like
Please tick the causes of Land degradation (7 choices).
What are the direct causes of …. (for each one above).
What are the indirect cause of…..(for each one above).
What are the effects of each one….(above).
Can you give solution, Please.
27. kenny | December 11th, 2009 at 12:08 pm
thanks for sharing.. i got the idea to continue my tasks..
tqvm
28. star | January 14th, 2010 at 8:43 am
May I know how to analyze the data after I create a variable with a single value for each possible combination?
eg.
1 = soup,
2 = rice,
3 = salad,
4 = soup and rice,
5 = soup & salad,
6 = rice and salad,
7 = soup and rice and salad
but in my result table, I only want show out the numbers of soup rice salad.
eg. soup 4; rice 4; salad 4.
29. Arthur Stevens | January 20th, 2010 at 6:44 am
Hi.
First of all thanks for the help.
However i still have a problem.
I have a multiple response question with 18 possible answers.
I used method 2.
Now i get the table and can make my graph which is great.
only it shows in actual count and I want it to show in %%.
However if I click on %% I get 100 % for every answer since the other answer ( 0=no) has been removed during the merging of the answers.
30. Nick Reynolds | February 3rd, 2010 at 9:20 pm
This is relatively easy to achieve.
Assign the values to each of your responses so that the new value for a response is double the value before.
1 = soup
2 = rice
4 = salad
8 = …
16 = …
Then create a new variable summing the values. This gives a unique code to each combination such that
1 = soup
2 = rice
3 = soup and rice
4 = salad
5 = salad and soup
6 = salad and rice
It works for any number of alternatives.
31. L. F. | February 23rd, 2010 at 3:14 pm
Hi,
Imagine you have a field in which you want to know how people find most motivating learning a foreign language. The variables are in this case 18 (eg 1-Internet, 2-Movies, 3-Publicidade…12-Magazines…). If someone responds options 1 and 2 do the input of numeric values without any space, as someone told me to do, which is 12. Result … the SPSS option will take the 12 instead of taking the 1 and 2 as planned.
Could someone help me?
Thanks
32. Wolfie | April 15th, 2010 at 9:06 am
how do we enter data for a questions that need you to rank from 1 to 7, 1- least prefered, 10- most prefered
33. ANTHONY GIKURI | May 12th, 2010 at 5:59 am
I am fond of researching however I lack skills in analysis of data especially when exposed to analysis tools like SPSS. In particular I need your help in analysis of multiple responses.
For example. What are the sources of your income?
1. agriculture
2. small business
3. family/relative support
4. others
Suppose a respondent ticks more than one how would I go about coding and analysing these data?
Thanks
34. Stephen | May 13th, 2010 at 11:53 pm
Thanks for the overview,
I used a multiple response set and it worked great. My research involves a questionnaire where the participants choose their favorite 3 objects (in order) from a list of 27. In my case, I have 3 variables (1 for each response) in the response set, using the categories option and 1-27. I can get the frequencies and the cross-tabs and all, but I am wondering if there isn’t a statistical significance test that I can run on the data? I would really like to be able to report the findings as significant or not…I know that a chi-squared test would not be powerful here, as a multiple response set breaches some of the assumptions. Does SPSS have a way of getting a significance level for these sets?
35. Stephen | May 14th, 2010 at 12:00 am
Follow up to my last comment and for anyone who has the same problem:
I think I found the answer. The custom tables feature of SPSS allows you to use a multiple response set and run statistical significance tests. In my case, I am comparing preferences and gender, where preference is a multiple response set and gender is the usual nominal Bernoulli variable (I didn’t make it clear what I wanted to show was significant).
36. Hannah | May 25th, 2010 at 7:21 am
I have created a clustered bar chart using a ‘multiple defined set’ variable as the category variable (x) within SPSS chart builder. A problem arises when I wish to have the bars show percentages (based on legend variable category), the percentages shown do not correspond.
Any suggestions on how to show the corrrect percentages are much appreciated! (It works fine in Custom Tables … can it be a bug in SPSS?)
Leave a Comment
Some HTML allowed:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>
Trackback this post | Subscribe to the comments via RSS Feed