Jump to content

stan

Members
  • Content count

    2
  • Joined

  • Last visited

Community Reputation

0 Neutral

About stan

  • Rank
    Casual Member
  1. I am quite familiar with NVivo and have used it for many qualitative research projects with interview data. Now I am using it for social media analysis (~5000 comments of varied length). I would like to ask what is the best approach to code the data where multiple codes may apply to a Tweet or forum post etc. I have collected Tweets, Facebook posts, forum posts etc and each is a different sources i.e. Tweets are a source, Facebook comments are a source, a forum is a source etc. So a source contains multiple rows of data. My question is how to code these and which is the best approach: Assign only one code to a comment? Assign multiple codes to a comment? Assign multiple codes but only to parts (e.g. a sentence) of a comment? Assign a code per sentence of a comment? A challenge is that the same code may be assigned several times in a forum post (e.g. someone may refer to the same phenomena in different ways in several different sentences) and I since I am dealing with a large amount of data I am unsure on how to best account for this so that when I do some basic statistical work the results make sense.
  2. I have extracted social media data (CSV file) over the course of a year (e.g. Tweets, facebook posts etc). I have extracted date of post and the comment. I would like to run a query to determine how the codes allocated to the data change over time. Is there a way to run a query so that I can see which were the most popular comments and nodes in any particular month? I am interested in months rather than specific dates, so do I need code the month (e.g. as April, May etc) so that I can do cross tabs? TIA
×