Jump to content


  • Content count

  • Joined

  • Last visited

  • Days Won


ofabro last won the day on November 23 2011

ofabro had the most liked content!

Community Reputation

1 Neutral

About ofabro

  • Rank
  1. Hello, I am working with interviews, in which the audio is synced with the transcripts of each interview. I created one node to code each speaker, and it works great. The problem is that if after syncing the audio and text, and after coding each paragraph with the respective speaker node, if later I want to merge paragraphs, a strange thing happens. Apparently, I can maerge, adjust the in/out time frame, and code it. The problem is that the same paragraph (now a merged results of two or more paragraphs) keep showing in the node as multiple countings (e.g. instead of say "reference 1", it says "reference 1-2"). The paragraph does not appear twice in the screen, it is just the multiple counting of paragraphs that is a problem, because when I run a matrix query, the same paragraph is coded twice. I've tried everything, and cannot remove this... any idea? thanks, fabrolino
  2. Another suggestion: Open Nvivo, but do not open your project yet. Go to Tools/Compact and repair. If there is any inconsistency in the file, a query is likely to find it, and crash. See if this helps. Fabrolino
  3. My suggestion would be: a. Create a CASE for each interview b. Create TWO nodes, one for Males, and another for Females c. See if you can format each contribution from the interviewer in one paragraph, and use a symbol to mark those paragraphs that refer to male, and those that refer to female participant [eg. [MM] [FF]) d. Use a Word query to move the paragraphs in to the correspondent NODE. Then, when you need to analyse it, you can run MATRIX query and find the information you need. Hope this have helped. Fabrolino
  4. Hi Bruce, The only solution that works for me is to include a number before the node name. A node called "like", "dislike", "unsure" would became 1.dislike , 2.like , 3.dislike . If I want to make something at the top, I use double final stop (e.g. "..dislike"). Not smart, but it works... fabrolino
  5. Hi there, One of my projects was too big, and I had to slice it in smaller pieces to work in a slower machine. I just created an empty file an imported the node tree to it, which worked fine. An odd thing is that I cannot filter the importing process by which sources I wanted to drag into the new project, and I had to work around it. Nvivo apparently just allows you to filter importing sources by user, thus I've created a fictional user, changed the sources in the big original file with this user, and then I could filter and import just what I needed. The problem is that I could not import the queries from one project to another? Did I got it right or Nvivo does not allow to do it? I am having to replicate the queries one by one... any ideas? Thanks! Fabrolino
  6. As far as I know, Nvivo will not "read" the filename and create nodes accordingly. However, if you don't have many sub categories (knoun in nvivo as nodes) I guess the best solution is to create the nodes in a tree-format, ans then import the files directly to each node (you can use drag and drop). One tip that might help: if you want to associate specific attributes to these sources, create the attributes/cases and change the "default attribute value" related to each node before importing the sources. This way you will have the files in the right nodes, and will have the attributes automatically coded to the correct attribute value. Best. fabrolino
  7. Dear Mike, I miss the two options as well. What I did in the first case (speakers name + quote) is to leave both coding in the same paragraph. Thus I used the "in the same paragraph" scope option to do the trick. In my case, as all the interviews had few participants in it, I actually used the filter option in the transcript/source screen to do the coding (I also used a uncommon symbol notation, break parenthesis, to differentiate when [Fabrolino] is the speaker and when Fabrolino is mentioned by someone else). This is nevertheless probably not what you (or me) ideally wanted. I like in Excel the fact that you can search and select a text that is either to the left or to the right of a specific criteria. I think this is missing in Nvivo. In the second case (de-select part of the coded node without uncoding), I do not have a solution. I realised that sometimes I just want to eclipse something to avoid reading it over and over, and sometimes I really want to code sentences that have in the between something I don't want to include in the node. I opted to do the latter option (exclude from coding the sentences I don't want) because this will not affect the coding results, while the former option will. However, I miss a feature to READ the coding results in a different way. I would love for example that I could simply continuously read all the codes that are in the same paragraph, or in the same source. Just like what we do when we quote a sentence from a book with words apart from each other and we use the (...) notation to group ideas in a simple visual way. That I miss. Hope to have helped somehow. But in general terms, I was surprised to note that I am much more able to merge/split/select/edit textual sources in Excel than in Nvivo. It seems that the textual formulas in Excel (which are per se very limited compared to what you can do with numbers or databases) can still do more several tricks that Nvivo can't. Nvivo is great software nevertheless, I just miss that. best, fabrolino
  8. Hi there, I am running a compound query, and I have one question. I have some nodes based on key words (e.g. mention to a specific newspaper) and some nodes based on broad themes (e.g. definition of what newspapers are). I coded both of them independently, and now I want to run a query just to make sure that: all the definitions included at least one newspaper label AND that all newspaper labels are used in one definition node. Because I have nodes that differ in length (i.g. one is a word, the other is a group of sentences) they will never overlap completely. that said, I am trying to use "near by" to locate this instances. If there is a paragraph with a newspaper label, and no definition node coded, it is likely that I missed this one. And this is exactly what I am trying to double check. The thing is: I can use the query to locate things that are "near by", but not to identify things that are NOT near each other. In sum, I want to find the cases in which the two observations do not overlap. However... this is puzzling me .. I could run a "not near by" query... Any ideas? thanks, fabrolino
  9. Perfect! It will work like a charm! thanks, fabrolino
  10. Hi there, Maybe a good idea... Sometimes I just want to split a transcript row in two. When I want to merge two rows in one, I do have an option for that. But if I want to break one long piece in two, I have to manually insert a row, cut and paste the text inside of it, and change the timing for each block row.... I would be happier if I could at least split the row in two automatically... I would just need to adjust the start/end time or ... Nvivo could just prompt me for this information, I type it in, and the job is done. Best, Fabrolino
  11. Hi there, If I have one entry in one node, and I want to copy it to another node I can use drag and drop and do the job. But what about if I want to MOVE instead of copy the entry? Is there any way to do it? (holding ctrl while I move the entry, for example)? Sometimes I want to recode and move one entry from here to there, but I do not want just to copy it; instead I want to delete it from them original entry and paste it in the new node. thanks, fabrolino
  12. Hi there, One of my tree nodes is based on a pro/neutral/con classification. For example, I have one node coding all the mentions to impartiality. It is important to know how many times impartiality is mentioned by each interviewee, but I also want to know if they are saying something in support or against impartiality (or maybe, their comment is neutral about it). That said, I want to divide my node in three parts: pro/neutral/against. I could create three nodes (+Im, 0Im, -Ip) and use drag and drop to recode it. However, it seems more reasonable to transform the Impartiality node in a Case node and use attributes do recode it. The problem is that I need to read each entry coded and from that decide if the entry is pro, neutral or against. Because of that the casebook window is not good for me. What I would need instead is a drag and drop option to code the Impartiality node.... the thing is, can this be done in NVivo? In sum, I want to use drag and drop relate attributes to nodes in order to avoid using the CASE NODE window. Any ideas? thanks fabrolino
  13. Dear Anne-Marie, If you properly coded the nodes as cases , this should be working... You did create attributes, right? But have you attributed values to the nodes (indicating which of your nodes are related to each attributes, or for example which are not to be assigned?). I am not sure how advanced you are in coding, but the Nvivo online tutorial is quite good in showing how to do this in case you need help. fabrolino
  14. Hi there. I have a solution that is not great, but works. You should simply create a bogus node (a free node code, for example) and before you stop to work you simply drag and drop some words from your document where you want to reopen later in the bogus node. Next time you enter NVivo, you open the bogus node, see where the last entry is and right click/open the source... It is not the best solution, but should do the trick. best fabrolino
  15. Edit from node

    Hi there, Maybe other people would like this feature as well. My main sources are made of transcribed interviews, and because they might have more than one interviewee in it (and because I cannot use drag and drop coding directly from transcribed sources) I created nodes for each interviewee and use these as the source for coding. What is happening quite quite often I spot typos that I would like to correct, but I cannot do it because source edit is not possible in nodes, only in sources. However, it would help if I could select the text I want to change in the node and simply be able to enable a "temporary edit the selection" option, correcting the typo straight from the node. The work-around I have right now is to right click, select "open referred source" and go to another tab-window and correct it. But this is not as quick as editing the text straight in the code window... best fabrolino