Hello Janet, Tim,

I had a look at the review and noticed that its set-up is a bit unusual. Each of the questions that would normally make up a coding tool have been placed in their own coding tool. This may be contributing to what you have described as you have a mix of 'completed' and 'uncompleted' coding for a particular coding tool.

If you look in the Review statistics tab you can see that coding for some items is marked as complete (with respect to a question/coding tool) while others are 'uncompleted'. The items that are marked as 'complete' will be the ones that your third reviewer is seeing when they try to code the items.
If you want that coding to be invisible to the third reviewer you are going to need to 'uncomplete' that coding. This can be done in the 'Review statistics' tab by expanding the appropriate coding tool and clicking the red disc next to the reviewers name. This will 'uncomplete' that coding so the third reviewer will not be able see other' reviewer's coding when they are carrying out their own coding.

As for the error message, could you let me know what it says. Or, if possible, could you send us a screen-shot to eppisupport@ioe.ac.uk so we can see what might be happening.

Best regards,
Jeff
 

Hello Tim,

I didn't realise you were talking about using the line-by-line pdf coding functions in your original email but that won't change the meaning of red as indicating 'not complete' and green as 'complete'. If you are stepping through the items in the 'Document details' window the colours will indicate the items status with respect to the codeset.
The reason they are flipped in the 'Review statistics' tab is that it is indicating what you can change it to the opposite status. By clicking on the red disc in the 'Completed' coding table you can change the item's status to 'not complete'.

What I meant by how the codes are set up is that the multiple coding tools you have created would normally be contained in a single coding tool. The tool would have a number of sections with each section containing a number of questions and each question would contain a number of possible answers. You can see an example of this in the example review that was placed in your account when you created your user account. If you look at the data extraction tool you can see what I mean. What you have done is not 'wrong' but it might make things more confusing.

Are we talking about the 'PCPC' review? If I look at that review I can see there are two reviewers in it (Deborah and Janet). From your original message I was understanding that a third person was involved. Was I mistaken about that?

 The line-by-line pdf coding functionality will work as expected when your coding tool is set for 'comparison' coding (two heads on the coding tool).  Because the coding is not 'complete' it will not be available to the program's searching and reporting functions. This will include the 'show text coded with this code' report. That is why you are not seeing anything when they click on 'show text coded with this code'.
 If they want to see their coding for a particular item they can click on the 'Coding record' tab (in the 'Document details' window), find the correct codeset and name in the table and click 'View' for that row. Their selected codes and pdf text will be displayed for that codeset.
 
 If you have any questions about this please let us know.
 
 Best regards,
 Jeff

Hello Tim, I have had a look at this and I think I understand what is happening. It appears that if the codetype is not set to 'selectable' then the highlighted pdf text is not included in the report when you select the reviewers and click 'Run comparison'. I tried this on your section J and when I changed the 'none' code to selectable (so a checkbox is visible) then Janet's code became visible (as that was the only code that she had applied text to). What you will need to do is to make all of your codes that you are applying pdf text to as selectable (this is a property of the code). You can tell if it is selectable because it will have a checkbox next to it so if you change the properties while in the 'Document details' window you will be able to easily see if the code is selectable or not. I don't remember if this behaviour is by design so we will be looking into it further. Best regards, Jeff

Hello Tim,

If you look in the 'Review statistics' tab and expand each of the codesets you can see that Deborah has used all of them. If you look at codeset A (in the 'Review statistics' tab) you can see there are two items where she applied codes to. If you click on the '2' to list those items and go into the first one you can see the coding that Deborah has done using that codeset. You can do this by going to the 'Coding records' tab and click 'View' in the row that is codeset A and has Deborah's name. You can also compare the coding on this item between Deborah and Janet.

Perhaps you were looking at an item that didn't have coding using codeset A and that was why you didn't see any coding for Deborah?

Best regards,
Jeff

Hello Tim,

If we are only talking about the item with 'CMA' as its title then there are only 3 codesets that Deborah has used in coding that item (codesets E,F and J).

The only scenario I can think of where she could have coded the CMA item using other codesets but not have that information displayed is if she was coding the item using a codeset whose coding was already marked as 'complete'. In that situation Deborah's coding would then become part of the 'completed' coding and not show as her own coding (but rather the person who's coding is the 'completed' version). Is that a possibility?

Best regards,

Jeff