I have a survey (around 800 respondents) with open ended questions (regarding suggestion to improve environmental problems in agriculture) which I coded (based on content analysis using a inductive category system).
Now I made a frequency table based on the codes for each age category (variable in the survey) and would like to run a statistical anaylsis:
Example:
the theme "catch crops" was coded 27 times in all open ended responses with the following distribution in the categories age:
under 18: 0 (times)
18-25: 3
26-30: 1
31-35: 1
36-40: 4
41-45: 3
46-50: 4
51-55: 1
56-60: 5
61-65: 3
66-70: 2
over 70: 0
It is possible that for one person catch crops was coded twice as each respondent had the possibility to give three answers.
I would like to test if there is a statistical significance between age groups with respect to how many times a certain topic was coded.
I would be very thankful for a suggestion, which test I could run on my data.
Thanks a lot!
https://us.sagepub.com/sites/default/files/upm-assets/82020_book_item_82020.pdf ( Chi-Sq testing ).
This RG link maybe useful
https://www.researchgate.net/post/Which_Test_statistics_should_I_use
In the first place, I think that I should carry out a "semantic reduction" by combining the various responses received in "similar, homogeneous or similar semantic frames or of the same sense or option"; After that, it would calculate the frequency of each "semantic frame" obtained, it would rank them by virtue of the number of subjects who opted for such "semantic frames" and it would offer the resulting MODA ("semantic frame" with more subjects).
Another possibility would be to do a "Delphi-type analysis" of the responses obtained, as is done in the "brainstorming"
