Difference between overlapping codes and double codes

26.06.2018, 10:35

Hi,

My documents are responses to an open survey question. The question was: About what researchtheme should we do research next? (in the area of X).
Sometimes people give two or more answers, like:
[do research about] 1) Housing 2) Care

and sometimes they give a response that indicates the intersection between two themes, such as
[do research about] residential care settings
which are institutions that combine both Housing and Care, so this is coded with both top level themes 'Housing' and 'Care'.

In both cases, the document thus gets two (top level) codes.
Is there an easy way to count how many times this are codes indicative of intersections, and how many times the codes are used independently?

Version: MAXQDA 2018
System: Windows 10
Loekarin
 
Posts: 45
Joined: 25.06.2018, 12:39

Re: Difference between overlapping codes and double codes

26.06.2018, 14:24

Hello again Loekarin,

Is there an easy way to count how many times this are codes indicative of intersections, and how many times the codes are used independently?

Yes, there is, although I would recommend to process 'Housing' and 'Care' not as top level codes, but as subcodes of the parent code "What research theme next?" (the structure of the Code System plays a role in some features, such as in "Complex Code Configurations").

To do this, activate the codes, go to the "Analysis"-tab and select "Simple Code Configurations". Paste the activated codes by clicking on the respective button at the top right, and then you'll get a table that shows you which combinations of codes co-occur and at what frequency. If you have more then one dimension/parent code, use the "Complex Code Configurations".

Another tool I can highly recommend to take a look at is the "Code Relations Browser". It'll show you which codes co-occur at which frequency in an interactive table (clicking on a square will retrieve the respective text passages). The reason why I didn't post it as my first hint was that it won't show you how often a code occurs alone.

I hope this helps! If not, or in case of any other questions, don't hesitate to ask!

Best regards,

Andreas
MAXQDA Support Team
Andreas V.
 
Posts: 272
Joined: 13.04.2017, 16:23

Re: Difference between overlapping codes and double codes

09.07.2018, 19:44

Hello Andreas,

"Yes, there is, although I would recommend to process 'Housing' and 'Care' not as top level codes, but as subcodes of the parent code "What research theme next?" (the structure of the Code System plays a role in some features, such as in "Complex Code Configurations"). "

We identified about 30 themes (now our top level codes), as answer to our research question. Within these themes/categories, we made subcategories (and sub-subcategories).

Why would you recommend to process (parent level) codes as subcodes of our research question?

Then we would have only 1 top level code (our research question), and many subcategories.
Loekarin
 
Posts: 45
Joined: 25.06.2018, 12:39

Re: Difference between overlapping codes and double codes

10.07.2018, 13:46

Hello Karin,

well, speaking generally, the relationship between the code "Research Proposal" and the code "Housing" is not on the same level, since the latter one is a subset of the former category. But if they are all top level codes, MAXQDA has no way of knowing that. Now when you use e.g. the Code-Relations-Browser (tab "Visual Tools") or the "Complex Code Configurations" (tab Analysis), those features will allow you to select a top level code, for whose subcodes you'll get information about co-occurrences. See on this page the section about Complex Configurations for example, it should make clear what I mean:

https://www.maxqda.com/help-max18/statistics-and-graphics-functions/configuration-table

So for you example, when you want to find out which proposals correlate with which of the respondent's disciplines, you need a code structure like this:

- Research proposal (top level)
-- Housing (subcode)
-- Care
-- Career

- Respondent's discipline
-- Sociology
-- Math
-- Political science

I hope that makes sense?

Kind regards,

Andreas
MAXQDA Support Team
Andreas V.
 
Posts: 272
Joined: 13.04.2017, 16:23

Re: Difference between overlapping codes and double codes

16.08.2018, 13:08

Hello Andreas,

Our respondents belong to one of 3 subgroups: person with autism, parent, or legal representative.

I would like to find out if there is a relation between the subgroup that a respondent belongs to, and the proposed research theme.

What is the best way to do this?

-Make document groups
-Make document sets
-Make a code 'Group' with 3 subcodes ('parents', 'persons with autism', 'legal representatives') and assign these subcodes to each document
Loekarin
 
Posts: 45
Joined: 25.06.2018, 12:39

Re: Difference between overlapping codes and double codes

21.08.2018, 15:46

Hello Karin,

I don't know if this is the "best way", but my own first step would be to check for expected frequencies. Assuming that the information about the groups is already present in the form of a document variable, and the information about the proposed research theme was coded, I would first transform the code into a document variable by rightclicking on the code and selecting "Transform into document variable". (If there are a lot of codes, you can also do this for all activated codes at once by selecting the "Quantitizing"-entry in the "Mixed Methods" section.)

Then I would start the Stats-Module, go to "Compare Groups > Crosstab", select those two variables, and then display both the actual and the expected frequencies to compare them.

Kind regards,

Andreas
MAXQDA Support Team
Andreas V.
 
Posts: 272
Joined: 13.04.2017, 16:23

Return to MAXQDA in Research

Who is online

Users browsing this forum: No registered users and 1 guest

We use cookies to improve your experience on our website. By clicking OK or by continuing to browse the website, we’ll assume that you are happy with their use. Click here to review our Cookie Policy. OK