NVivo 9 for Psychology Dissertation–need help please

[Anonymous]

Hello,
I am attempting to use NVivo 9 to code qualitative data, though it is not the traditional in-depth interviews. Participants in a recovery program responded to 3 open-ended questions, each question having 400-600 participant responses. I am looking for NVivo to help identify themes (to see if they correlate to themes I have identified), but I am REALLY struggling to figure out how to get this to code. Any thoughts would be much appreciated!

[Danelle]

Sorry…I hadn’t registered yet. This question is from Danelle. Thanks for any help!

[QDATRAINING Admin]

Hi Danelle,

First thing to consider is, is the database set up correctly. You don't say how the data is currently formatted. Is it in spread sheet, text files etc…? If you have background demographics or other profiling information, these should be  set up as classifications and linked to case nodes?

Next, if all you need to do is conduct a thematic analysis of each question, then you could code all the responses to question 1 to a single node (you can probably automate this bit). Then, open the node and code it's content to sub or child nodes clustering them by theme. Have a look at the tutorials here as a good starting point:

http://www.qdatraining.eu/tutorials

When you have looked at the basic tutorials we might log on and take a look at your set up. Let us know if you need this?

Kind regards,

Ben

[Danelle]

Hi Ben, You are a life saver! Thank you so much for your reply. I will look at the tutorials now. I would be happy to send my datasets to you: I have the documents in Word, with the question at the top and participant # followed by participant response listed underneath the question. This is the same format for each of the 3 questions. I coded the data myself; now I'm just hoping to add to my dissertation by using a program my committee hasn't used before. My goal is that the program will identify themes that I found too, so that I can have support for my findings. I am so thankful for your reply–I am in a time crunch and feeling overwhelmed bc I've been really struggling to figure out the program. I can't thank you enough. My email address is (e-mail address removed to protect from the user from spammers)  and I am in the U. S. Thank you again very much! Please let me know what (if anything) you would like me to send, and I will let you know how I do with the tutorials. (E-mail address removed to protect from the user from spammers) A million thanks! Danelle

[Danelle]

Hello again,
These tutorials are great! I think I know where I’m getting stuck, but I’m still not clear on how to rectify the situation.

The autocode features will group together responses to the same question across interviews, so that way we can look within each question to find themes ourselves. Definitely much easier and a great feature. For my data, I have 3 questions, each saved in a separate Word document that I uploaded into the program and made folders for. There are lots of responses for each question, but the questions don’t relate to one another–so I’m looking for themes within the same document, instead of across.

I do just need to conduct a thematic analysis within each question. I was hoping that the software would find themes within each question so I could compare the themes it finds with what I previously found…or do I need to identify all the child nodes under each question bc this isn’t something the program is able to do? Even if I identify the child nodes, does the software look for themes using my child node as a key word or phrase and code the data that way? I’m so sorry for my confusion! I feel silly not being able to figure this out, but it’s just not intuitive to me. I will keep trying…and I do appreciate the help!
Danelle

[QDATRAINING Admin]

Hi Danelle,

I think a discussion is in order rather than trying to deal with this in a series of forum replies. If you go to this page:

http://www.qdatraining.eu/support 

You need to read the instructions and download the requisite file (this wil make more sense when you have read the instructions). We then need to arrange a time to meet on-line and I can demonstrate your options to you. Bear in mind Danelle that we are in Europe and probably about five hours behind you so your afternoon is optimum because it will be our evening here. 

This is the most efficient way manner to proceed from here.

Kind regards,

Ben

[QDATRAINING Admin]

Not a good plan to post your e-mail address on a Forum. It may be picked up by spammers. If you need to contact via e-mail you can use support@qdatraining.eu and we will respond. I have removed your address from the previous post. This account is protected from spam.

Kind regards,

Ben

[ejh2]

Danelle, have you got some a priori codes from the literature that you can set up and then use to start your coding. Other codes will emerge as you start thinking about the data.
The other thing I found worth doing on my dissertation was to annotate the data with whatever thoughts struck me when I first read it, and later. Months after I reread those annotations and was thus reminded of issues that I’d forgotten about but that fitted together after so much more thought and analysis.

[Author]

Posts