Jump to content

Kristi Jackson

Trainers
  • Content count

    129
  • Joined

  • Last visited

  • Days Won

    5

Everything posted by Kristi Jackson

  1. Are you a doctoral student who used NVivo for your dissertation within the last 6 months? Does your dissertation include an education research component? (It can be in a field outside of education as long as the research addresses an education question) Earn $100 for participating in a 60 minute telephone interview about the use of NVivo. I am conducting a study on perceptions among doctoral students and their committee members regarding the benefits and problems associated with using NVivo in a dissertation. If you agree to participate in the study, you will be asked to provide me with a copy of your final, successfully defended dissertation, and you will be asked to participate in a 60-minute telephone interview (some time after your defense). The interview will focus on an exemplary qualitative research article of your choosing, the use of NVivo in the dissertation, and any negotiations you observed among your dissertation committee about the use of NVivo or the presentation of findings in your dissertation. I will also ask you to identify the committee member who was most interested in how you arrived at your findings and who might be willing to participate in a similar 60-minute telephone interview. The confidentiality of the interview content will be protected by the researcher, although the committee member and the student will be made aware of each other’s participation in the study. All participants will be compensated $100.00 for their time. Data collection will occur any time between now and August, 2013, depending on your availability and the timeline for completing the dissertation. For more information, please contact Kristi Jackson at kirsten.jackson@colorado.edu, or 303-832-9502.
  2. Kristi Jackson

    Make visible the "see also links" to memos from memos

    Absolutely agree that See Also Links should have an option to be "bi-directional" when created, or that one-direcitonal and bi-direcitonal are defaults in the Options. For people who want them to go both ways, their work is doubled.
  3. Kristi Jackson

    matrix to node

    One option is to run the matrix (with results saved in the results area of Queries), then copy the matrix result from the List view, and paste it in as a Tree Node. The columns will be turned into nodes with Rows as subnodes (so each cell becomes its own node in a tree). Hope that helps.
  4. Kristi Jackson

    What is coded

    A potential work-around is to open a Source file, then go to "View" from the main menu. Select "Coding Stripes" and "Coding density only". This will turn on one stripe, which will display a darker black next to passages that have more coding applied to them and a ligher white with passages that have little coding). If you over over one of those dark patches in the coding density stripe, it will provide a list of the codes in that area of text. Cheers.
  5. has not set their status

  6. Kristi Jackson

    Adding a model to a case

    Claire, sorry if I misunderstood you -- and Pat has provided great ways to think about a model that has a case in it (add a case to a model) and connections to other items in a database. You cannot put a model into a case, though -- just to clarify -- but again I may have misunderstood your question. All the best, Kristi
  7. Kristi Jackson

    Adding a model to a case

    Nope. But you can create a model about a case. Cheers.
  8. . Hiya Claire! Copy the projects first. Then open one and go to "File" and "Import Project". Several options allow you to select only certain project items. You cannot merge two models together but you canhave both models in the smame project.
  9. The problem is probably that your cases don't have any qualitative data in them (only attributes). To check, go to Nodes in the Navigation View, and then to your cases. Double click on a case. If it's empty -- that's the problem. You need to code all of the text belonging to a case into that case (just like a Tree or Free node contains the qualitative passages for a theme, Cases contain the qualitative data for a case). There are quite a few ways to do this, but if you have 75 or fewer cases, and you have a document representing each case in your Sources, it's probably going to be fairly easy to highlight all of the text and drag it to the Case Node. There are actually 7 different ways to create cases -- and about 4 ways to get text into cases -- but you don't need to learn them all -- I'd stick with what you already know how to do -- code the text into the case the same way you coded to other nodes. Cheers.
  10. Greetings -- I have one possible explanation. While your attributes may be mutually exclusive, you may still get incorrect counts if your CASES are not mutually exclusive. In other words, each piece of qualitative data (text, audio, video, photo) that is in a case should be in that case and only in that case. If you have a hierarchy of cases (so the parent case is an aggregate of the children), or if there are other methodological reasons you have the same qualitative data in multiple cases within NVivo, you will not get accurate case counts. If you follow the rule for mutually exclusive case data, you will retrieve accurate case counts. If there are methodological reasons you cannot make your cases mutually exclusive, you may find some alternative strategies in either using SETS or copying the project and using one unit of analysis in one project and another unit of analysis in another project. Let me know if that hits the mark or not. Kristi Jackson, kjackson@queri.com
  11. Just out of curiousity, Matt, can you tell if there is any formula notation in those cells that change color?
  12. Kristi Jackson

    Read-only document

    You can't edit if the coding stripes are on. If you turn them off can you edit?
  13. I think I understand your question, but I'm not 100%. At any rate: The matrix intersection is now a Matrix Query, and when you create it, you'll want to "save results as a new matrix" instead of "preview only" (from the third tab where you enter the instructions for handling results). After it runs, go to Queries in the navigation view, and then to the results subfolder -- copy the matrix and then paste it into your tree -- the individual cells become individual tree nodes.
  14. Can you do an intersection of the document and the node first, and then use that Result as the filter?
  15. Kristi Jackson

    exporting node set

    I might not be fully understanding the scope of your needs, but I often copy a project, then delete all the documents from the copy. It's a brand new project with only the nodes. Import new files and you're off and running. The other alternative is to purchase Merge for NVivo2, and then do the same as I suggest above (take the project with the desired nodes, copy it, and delete the source documents), but Merge the new, empty project (that has the skeleton of the nodes) into the project you already have up and running (in which you might already have done other coding on some of the relevant documents). This will allow you to import the new nodes, while retaining your old nodes, and only incorporates the documents from one project. Also one final note: Researchers often split projects up when they don't need to. The software has wonderful abilities of sorting things for you (only looking at a certain set of docs in a certain set of nodes, for instance). So, when possible, I recommend people put as much as possible (remember that NVivo2 has some size limitations) into one proejct, and make NVivo do the filtering for you. That way you don't have to run the same report in all 5 projects when you want to compare across data sets -- you simply run 5 different reports (and sometimes only one report with the comparisons built in!) in a single project. Cheers.
  16. Just to add to Leonie's wise (and always kind) words --- if you have multiple speakers in a file, you simply need to create a node for each speaker (usually a Case Node, and usually accomplished through auto coding, but it sounds like you might have done this manually, which is also just fine). You'd scope the search on "Selected Items" and then use the "Select Project Items" window to choose the NODE for that speaker, rather than for a paricular file that contains all of their text. Cheers.
  17. Kristi Jackson

    Customize reports

    David -- thank you for articulating this feedback, as many of my clients are requesting improvements in the options for reporting as well. It's not so much that there's anything wrong with the current reports (for the most part), just that researchers (and especially research teams) have a lot of project-specific issues that they need in reports. I think there wasn't as much of this feedback early on as people began using the software, but now that there are more users of NVivo7 and more need for output (in projects that are hitting the six month or year mark), I'm also hearing this need for additional reporting options more often. I've found QSR to be very responsive to this kind of feedback (the issues with coding stripes and the service pack additions to coding stripes early on is an example), so I'm trusting we'll have a few more options in NVivo 8. I just wanted to echo your request to QSR, to communicate that I've been hearing a crecendo of these reporting requests from my clients, and also to say that it's absolutely wonderful what this software can do -- so thank you to QSR for all the improvements that have already been made!!!!! Best.
  18. Kristi Jackson

    Coding problem

    If you are coding via "drag and drop" this shouldn't happen, but I've seen it happen when people use the right click, menu, or coding bar and they find the option to "code sources" rather than to "code selection". Could that be the problem? You don't want to "code sources", ever, if you are only hoping to grab small bits of a file and send those to a node.
  19. Three things that I've done in the past -- but they might not meet your needs: 1) code the files, and be sure to give the files slightly different names (sounds like you've already done this). Get these files into the same project (by importing one project into the other if needed), and then run a matrix intersection on your documents BY your nodes. This allows researchers to talk through data either one document at a time, or one node at a time, but it does not provide the coding stripe map, as you've requested, in order to work from a transcript. 2) Instead of looking at all the coding stripes at once, create sets of coding stripes for different parts of your coding structure - and then look at a more manageable list of stripes in the two files (which you can also do easily on-screen, but you may need to "undock" the two open files from their List View. 3) Run the Coding Summary report for those documents -- this will provide the entire document, but sliced by node.
  20. Kristi Jackson

    Full meaming of NVIVO

    To see Pat Bazeley's very excellent summary of the history of NUD*IST, see page 4 of her newly released book, Qualitative Data Analysis with NVivo, from Sage. It's a fabulous book. Get it. She discusses the evolution of NUD*IST which is an acronym (Non-numeric, Unstructured data * Indexing, Searching, Theorizing [or Theory-building -- I can't remember which]), and the arrival of NVivo alongside NUD*IST in 1999. In vivo coding, as implied below, is a focus on inductive rather than deductive coding. One way of being inductive is to use the words of participants, but it is not the only way. NUD*IST is an acronym, but it was changed to N5 and N6 (taking off the remaining letters) for obvious marketing reasons as the software became popular. (When you goggle "nudist" this isn't the first "hit"). When NUD*IST became truncated, NVivo came out, so the parallel naming of N5 and NVivo was handy and it helped with the branding of both products. NVivo isn't really an acronym, like NUD*IST was -- it is a play on words, as already mentioned, pulling from the use of In Vivo coding (which preceded NVivo as a software). Hope that helps.
  21. Kristi Jackson

    Full meaming of NVIVO

    To see Pat Bazeley's very excellent summary of the history of NUD*IST, see page 4 of her newly released book, Qualitative Data Analysis with NVivo, from Sage. It's a fabulous book. Get it. She discusses the evolution of NUD*IST which is an acronym (Non-numeric, Unstructured data * Indexing, Searching, Theorizing [or Theory-building -- I can't remember which]), and the arrival of NVivo alongside NUD*IST in 1999. In vivo coding, as already mentioned, is a focus on inductive rather than deductive coding. One way of being inductive is to use the words of participants, but it is not the only way. NUD*IST is an acronym, but it was changed to N5 and N6 (taking off the remaining letters) for obvious marketing reasons as the software became popular. (When you google "nudist" this isn't the first "hit"). When NUD*IST became truncated, NVivo came out, so the parallel naming of N5 and NVivo was handy and it helped with the branding of both products. NVivo isn't really an acronym, like NUD*IST was -- it is a play on words, as already mentioned, pulling from the use of In Vivo coding (which preceded NVivo as a software). Hope that helps.
  22. I'm assuming that you already know you can change the default to 200 stripes by going to the main menu to View --> Coding Stripes --> Number of stripes. Now, to the problem: While you can view up to 200 stripes in a source (or node) while you're in detail view, the only nodes that will automatically appear in the stripes are ones that you have already applied in that detail view of the source. My workaround: I take the first word in the file (and sometimes the description of the file if I have one at the top of the page) and code that text at every node in the system (for this, it's easier to select the text, right click --> Code --> Code Selection at Existing Nodes --> then pick all free and tree nodes). Now they're all activated in that file. WARNING: This can cause huge problems when you run reports/frequencies/queries, so I always place an "X" at the end of my source names to indicate that they have this "dummy" coding early in the file, and I remove the X when I am finished coding and as a last step have "Uncoded" that first word or description in the file from all the nodes. So, be mindful of the problems you can create if you use this strategy (it's not a flawless approach, and you'll see that it is annoying having to look at some nodes that are NEVER applied to the file, because you've done the "dummy" coding at the top -- but if you keep using the "recently coded" option for the stripes, it could be what you need). I do believe QSR has received the suggestion that it would be nice if these automatically refreshed. Cheers.
  23. Not sure if this helps, but you can bring in any shape from the available options, and then convet that item into something from the project. So, for instance, you can convert a diamond into a node (with a link to the node, but still in diamond shape). However, you cannot bring a node in as a diamond. The project items have default shapes assigned to them. So, if your goal is to alter the defaults given to you by the program for specific project items, you select the desired shape first, bring it into the model, and then convert (link) it to the desired project item.
  24. Kristi Jackson

    N6 and NVivo compatibility

    You can bring an N6 project into NVivo, but not the other way around. However, you can use free demo software for NVivo to examine the data. If you have the demo of NVivo2, the demo software lasts forever, but you cannot save any changes to the data (so you can search, report, etc., but not permanently "build" anything). If you have the "demo" temporary license of NVivo7 (the software went from 2 to 7 with no 3, 4, 5, 6 because it takes the best of N6 and NVivo2 and has built into an entirely new package), then you can save, build, etc. BUT the NVivo7 temporary license only works for 30 days, so you want to choose your 30 day window wisely. Both packages (NVivo2 and NVivo7) allow you to convert an N6 project into them. Hope that helps.
×