Jump to content
The mixed methods coach

Coding Stripes Show More Than They Should When Working With PDFs

Recommended Posts

Hi everyone,

 

Further to my post yesterday on the problem of coding a databite to more than one node while having to move the scrollbar of the List View, another glitch appears today which really wears off my enthusiasm to use NVivo to code and analyse PDF.

 

In a PDF source, if two separate databites coded at a node are located at the top and the bottom of a page, the associated coding stripe does not correspond to the actual databite sections in the text. Rather the coding stripe expands throughout the page. See picx 1 for the databites coded at the node attitude.

 

This problem also occurs when viewing the coding at attitudes within another node, selecting the PDF view Tab in the Detail View– see picx 2.

 

Could we expect this problem to be resolved soon, please?

 

Thanks, Marie-Hélène

post-9402-131005964909_thumb.jpg

post-9402-13100596825_thumb.jpg

Share this post


Link to post
Share on other sites

Hi Marie-Hélène,

 

Thank you for raising this with us. I have managed to replicate the issue using pdfs with multiple columns. The issue has been raised and will be considered for a fix in a future release of the software. BTW, I also noticed that this issue didn't happen if I coded the top and bottom in one column pdf page. It seems the issue happens when coding across multiple columns, for example: viewing coding stripes of coded content from bottom of column 1 and top of column 2. The application will consider that the content is continuing from 1 column to another, resulting in coding stripes highlighting the whole content. However, the coding references seem to be correct with only the bottom of column 1 and top of column 2 coded. Thank you.

 

Best Regards,

Andri Purmawinata

QSR International

Share this post


Link to post
Share on other sites

Hi everyone,

 

Further to my post yesterday on the problem of coding a databite to more than one node while having to move the scrollbar of the List View, another glitch appears today which really wears off my enthusiasm to use NVivo to code and analyse PDF.

 

In a PDF source, if two separate databites coded at a node are located at the top and the bottom of a page, the associated coding stripe does not correspond to the actual databite sections in the text. Rather the coding stripe expands throughout the page. See picx 1 for the databites coded at the node attitude.

 

This problem also occurs when viewing the coding at attitudes within another node, selecting the PDF view Tab in the Detail View– see picx 2.

 

Could we expect this problem to be resolved soon, please?

 

Thanks, Marie-Hélène

Share this post


Link to post
Share on other sites

Hi Marie-Hélène,

 

Ben Meehan here in Ireland (nice gig in Cork recently - well done). My solution is to obviate NVivo when working with PDFs because to date it simply does not work well despite the recent visual improvements in 9.1. Although the PDF looks okay because it appears to retain its formatting, it is only an aesthetic improvement because it still codes as raw text and can’t, for example, code images at all. Thus your problem with tables or columns! Nvivo does not work right with PDFs and for this reason and you will not be thanked by trainees for recommending it for their lit review as another example of user need. Many academic papers include images and tables and these documents need to be coded in a holistic fashion just like any source. This is currently not possible in NVivo. You can’t even use heading styles for auto-coding because you can’t format a PDF after importation.

My solution is to use a PDF editor (Smart PDF Pro is very cheap) and convert the PDF to an editable MS Word document and then format and import. Problem solved! 

Hope you are keeping well.

Warm regards,

Ben

www.qdatraining.eu

Share this post


Link to post
Share on other sites

Hi Marie-Hélène,

 

Ben Meehan here in Ireland (nice gig in Cork recently - well done). My solution is to obviate NVivo when working with PDFs because to date it simply does not work well despite the recent visual improvements in 9.1. Although the PDF looks okay because it appears to retain its formatting, it is only an aesthetic improvement because it still codes as raw text and can’t, for example, code images at all. Thus your problem with tables or columns! Nvivo does not work right with PDFs and for this reason and you will not be thanked by trainees for recommending it for their lit review as another example of user need. Many academic papers include images and tables and these documents need to be coded in a holistic fashion just like any source. This is currently not possible in NVivo. You can’t even use heading styles for auto-coding because you can’t format a PDF after importation.

My solution is to use a PDF editor (Smart PDF Pro is very cheap) and convert the PDF to an editable MS Word document and then format and import. Problem solved! 

Hope you are keeping well.

Warm regards,

Ben

www.qdatraining.eu

Share this post


Link to post
Share on other sites

Hi Ben,

 

In fact you can code images in PDFs in NVivo 9.1, by changing your PDF Selection mode from Text to Region.

 

Take a look at the following help topic: http://help-nv9-en.qsrinternational.com/nv9_help.htm??#procedures/basic_coding_in_documents_pdfs_externals_and_memos.htm#MiniTOCBookMark3

 

Hope this helps.

 

Regards,

Share this post


Link to post
Share on other sites

Dear Ben,

 

Many thanks for your reply. I did go the PDF Editor route two years ago when I converted 250+ academic papers into text files to analyse them in NVivo. Although this seems the logical thing to do when one wants to use NVivo for text analysis, the work that it involved was incredibly cumbersome since you never really know how bad the PDF will look like when you turn it into a text file, and how much time you’ll need to invest so that the source can be coded and analysed.

 

Also, after importing the text files in NVivo, the fact that the visual context of the PDFs was gone negatively impacted on the coding. I felt it was like coding the sound of a video recording without being able to look at the video shooting. So, when comparing these disadvantages with the PDF glitches in V9, I still prefer to code the original PDF. Hopefully the glitches I mentioned will be solved soon.

 

Hope all is well with you and yes, Cork was lovely!

 

Take care,

 

Marie-Hélène

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×