Leaderboard

Hi, If all 5 interviews are in a single document, you can create 5 nodes based on interviews and code respective interview text to those nodes. Then you run a Matrix Coding query between interview nodes and your theme nodes. The resulting matrix will show the number of common coding references across interviews and themes. The other solution which you've already tried is to have these interviews in separate documents. You can then run a Matrix Coding query between interview documents and nodes. Kind Regards, Bhupesh

Hi Gail, I have attached a corrupt dummy project with this email to help you recover the project from the .nvb file. Please follow the steps below to recover your data from NVB file: 1. Download and unzip attached file(based on your NVivo version) to your computer. 2. Open NVivo, click on Open Other Project, then browse and select the dummy file saved in step 1. 3. You will be presented with an option to recover from a project recovery file. Select your .nvb file from the list to recover the project. If your project does not appear in the list, you can try searching your computer for any .nvb files, these are project recovery files. 4. The project should open once recovery is complete. I hope this helps. Kind Regards, Bhupesh NV12CorruptProject (for restoring project backups).nvp.zip NV11CorruptProject(for restoring project backups).nvp.zip

Hi Nic, It’s not possible to get quite the output you are after. The closest thing you could do would be to view your dataset with coding stripes turned on to get an indication of how it has been coded. You can find information about turning coding stripes on at the following link: https://help-nv.qsrinternational.com/12/win/v12.1.86-d3ea61/Content/coding/coding-stripes.htm As for finding the number of your survey participants coded at each of your theme nodes, something you might want to explore is running a matrix coding query. For example, you could have a matrix coding query that has the 15 codes from Q1 as the rows and the 5 from Q2 as the columns. As you won’t have any references coded to both your Q1 and Q2 sets of nodes, you would need to change the query criteria to show references coded at the row criteria NEAR the column criteria and change the proximity to find references in the same dataset row. Once you have run your query, you can change the cell content to display cases coded rather than references coded. This will give you the number of survey participants who have coding at each intersection. Another one of the cell content settings you may be interested in is the row percentage. This will show the percentage of the total value of the row coded to each of the columns which would let you find information such as ‘Of the participants who say we do X better, 30% say it’s because of funding’ and so on. You can find information on how to run a matrix coding query at the following link: https://help-nv.qsrinternational.com/12/win/v12.1.86-d3ea61/Content/queries/matrix-coding-query.htm To set some of these query options, you will need to follow the instructions to run the query outside of the query wizard. After setting the criteria to search for rows NEAR columns, you can click the ‘Options’ button to change the proximity. You will need to change the proximity to ‘In Custom Context’ and then click ‘Specify’ to set the proximity to the same dataset row. I’ve attached a few screenshots showing where you can change these options. Regards, Heather

Hi Audrey, What process are you following when you are merging the projects together? You should be able to merge the two projects together following the instructions at the below link: https://help-nv.qsrinternational.com/12/win/v12.1.86-d3ea61/Content/projects-teamwork/import-items-from-another-nvivo-project.htm? When merging projects, if two project items are duplicates the project item won’t be imported again and will appear on the list of items not imported. Coding to this item will be imported and merged with the existing coding. If NVivo successfully detects that your document is identical in both projects, it’s expected that the document will be on the list of items not imported. The coding from both your coders should be present in the merged project. If the memo only exists in one copy of the project, the merged project should contain this memo. If both copies of the project contained this memo but with different text the merged copy of the project would contain both copies of the memos. NVivo will only merge the memos together if the text is identical, otherwise both copies are kept. You would need to manually edit the memos and delete the second copy. If your projects still don’t seem to be merging as expected, please submit a support request at the following link as we may need to take a look at your files: https://www.qsrinternational.com/nvivo/contact-us/customer-support Regards, Heather

Hi aatrainor, You can find the NVivo 12 help at the following link; https://help-nv.qsrinternational.com/12/win/v12.1.86-d3ea61/Content/welcome.htm The option to show or hide annotations is present in the View group in the ribbon. The tab this appears on will depend on the source type you have open. For example, if you have a document open this will appear on the ‘Document’ tab, while with a video it would appear on the ‘Video’ tab and so on. Regards, Heather

After creating a matrix coding query with cases down the rows and coding categories across the columns, NVivo allows you to filter the matrix by row or column values, which is very helpful. However, it does not let you save these filtered results with your query or export them to Excel, from what I can tell. Is there any way to export the matrix to Excel with the column/row filtering preserved? Thank you!

Hi Ruben, To try and resolve your issue, try merging your project into a blank new project. To do this: 1. In NVivo, create a blank new project. 2. Click the Import tab and the click the Project icon. The 'Import Project' dialog box appears. 3. Click Browse and navigate to and select your project file from the Desktop. 4. Click Import. Once the import finishes, see if this issue is resolved in this new project file, which contains your entire project data. If the issue still persists, can you please send us the project file by submitting a support request at https://qsrinternational.com/nvivo/contact-us/customer-support ? Kind Regards, Bhupesh

I have a coding matrix of cases by a few nodes. I want to produce a report of the text coded at the specified nodes for each case. It should look something like this: Case 1 Node 1: Text..text..text … Node 2: Text..text..text … Case 2 Node 1: Text..text..text … Node 2: Text..text..text … Etc. Does anyone know how to do this?

Hi cgev, Currently there is no direct way to do this in NVivo, however it may be possible with a Matrix Coding query. The query can be set with columns being your 36 nodes and rows as your responses. As the responses are just text at the moment, you will need to make them somehow a project item, such as a node. Once the query is run, you can display if the item in a row is coded to a particular node (column) by right clicking on the matrix and from the menu selecting Cell Content > Coding Presence. These query results can also be exported to an Excel file. For more information on matrix coding queries, see the following help link: http://help-nv11.qsrinternational.com/desktop/procedures/run_a_matrix_coding_query.htm Hope this helps.

Hi everyone, I know the Nvivo is about qualitative analysis. Nevertheless, when using the cluster analysis function, we need to explain the validity of those results. For example, how to show that Pearson coefficient of 0.46 is significant? The p-value logic could be applied here? If it is not possible to know all the parameters to calculate the Pearson coefficient, how to justify the use of Nvivo as a reliable tool? I really need some help with this. Thanks very much!

Hi Heather, Thanks for the answer. Please, I don't know how to calculate manually because the parameters and the approach used are unknown. Is there any way to ensure or qualify the Pearson correlation number provided by NVIVO? For example, what is the N ? What the significance level for one-tailed test? Is there some online tool to get this values using the text files? Thanks.

The two options for cluster analysis appear to be Source and Node. Is it possible to do a cluster analysis for sources that I've grouped using case classification attributes? For example, let's say School 1, School 2, School 3, ... n each have a # of sources (records, interviews, etc.). I made these sources into cases, then assigned the names of schools as case classification attributes (so School 1 is the attribute value for all its sources and so forth). I want to see which schools (= its collection of sources) have similar coding.

Hi Bhupesh, Thanks for your reply. (1) When I run a cluster analysis on sets, the diagram shows all the sources in that set (e.g., all interviews)-- the analysis looks the same as selecting all sources -- it does not compare set vs set (e.g., School 1 vs School 2). Am I doing something wrong? (2) Matrix queries have been useful for comparing cases, but I'd hoped there was a way to run a cluster analysis. I guess I could export it to Excel for the cluster analysis. best, Lisa

Hi dr_cnkimball, There are a few approaches you can take to organising your data. Possibly the simplest approach would be to rename your sources in NVivo so that the time/date is a part of the source name, or use a different source folder for each time. You could also use sets to organise your sources, so you could have a different set for each time. Sets are different to folders in NVivo as a set just contains a “shortcut” to your project item. This allows you to have one project item included in multiple sets. You can find some information about this at the following link http://help-nv11.qsrinternational.com/desktop/procedures/create_and_manage_sets.htm Another approach would be to add the time as an attribute to a source or case. This would let you run queries using this attribute. You can find details on classifying sources at the following link http://help-nv11.qsrinternational.com/desktop/procedures/classify_sources.htm Regards, Heather

As a Trainer, I feel difficult to demonstrate a practical example of cluster analysis. Could you form support create some example of using and reading the output of a cluster analysis. The creation process is quite simple but the interpretation fo results is not that easy. If you do so, make it for the three different kinds of criteria (words similarity, coding similarity and attribute values similarity). Regards, Ernani