I need to fix a problem on Nvivo. All my interviewees' audio files are grouped as sources and I can't classify them according to gender, age, etc. Any suggestion on how to solve it?
Raven's Eye is an online natural language analysis tool based. Raven's Eye provides its users with dynamic, powerful, and instantaneous automated analyses of written language. Raven’s Eye facilitates an incomparably intimate and multifaceted understanding of people, as well as the ability to speak to them in highly resonant ways.
65 languages to analyze data with! Multi analytic functions and features! Graphic user interface. SaaS pricing, super cost effective!
Have a nice day and good luck with your research!
http://youtu.be/hdsqMVY118U
Here's a link to a YouTube video on how to code audio files. Good luck!
Best,
Wendy
http://youtu.be/hdsqMVY118U
Hello Cristiana,
By now you will no doubt have solved your problem but my tip may be useful to anyone else looking at your question with a view to analysing audio interviews (or indeed in any other media).
The quickest and easiest strategy is to
1. create a spreadsheet (excel if working in windows environment) with a row for each participant, with columns for id/name/pseudonym plus each item of demographic and other characteristic (eg location) of interest to analysing the interview themes and findings.
2. Import the spreadsheet into Nvivo as a classification sheet and select the option to automatically create individual 'person' nodes - best done into a separate folder in the nodes section.
This results in a node for each participant with a linked classification sheet, which you can later use to run (classification) queries after coding the interview and other data.
Before coding the audio, or other media, file to themes
a) prepare key overarching thematic nodes from the interview questions if you have used a structured or sem-structured interview format or the theoretical framework of your study if interviews were open format.
b) create sub-nodes under each main question/issue /heme node, for instance positive, negative, fence-sitting, no idea... code the interview responses first to the appropropriate sub-node relating to that theme. It is much easier/quicker to get the aggregate view (check the aggregate property on the top level thematic node) than to have to go back and divide the answers down into different aspects.
If each 'source' file is from only one interviewee, code the whole file to the individual./person node, when you load it. Play the file and code content segments to thematic (sub) nodes.
If you have an audio file with mutliple participants e.g. from recorded focus group, you need to load and play it in order to code segements to a) the respondent/person nodes as well as b) appropriate thematic nodes. The YouTube video Wendy recommends above should show how to do this, alternatively Nvivo documentation explains the procedure with an example.
The approach above also makes it easier to run queries, using the respondent classification fields (e.g. gender) and nodes with sub-nodes, to produce tables, stats and graphs of the sort: no/percent females/males with postive/negative view on topic under question.
There are alternative approaches which work, such as making nodes for particular characteristics (gender, age group, location, position), but these are more time consuming to manage and especially change, as your undertanding of what matters changes during the coding.
regards, Lorinne
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Software