Hello Lesley,

I had a look at your review and I believe the numbers you are seeing are correct; it's just how to interpret the numbers that may be confusing.

I created a new comparison (the one dated 25/09/2012) so I was sure I was looking at the most recent data. If you 'View' stats using this comparison it will show 121 items for Jessica and 72 items for Lesly. In the Coding assignment table it shows 121 items started for both of you. The reason for this discrepancy is that the comparison was generated after some items in the original work assignment were 'completed'.  If you 'Run' a quick report using that new comparison you will see 121 codes under Jessica's name and 72 under Lesly's name and 49 in the Agreed column. The 49 in the agreed column would have originally been in the Lesly column but once they were 'completed' they moved to the 'Agreed' column. The 49 items in the 'Agreed' column are items that have been marked as 'complete' under the assumption that the selected code is what the two coders have agreed is the correct response. You now only need to look at the 72 items in the Lesly column.

The 72 items in the Lesly column are in most cases agreements. In the 'View' stats window you can see that 59 out of 72 items are in agreement and 13 out of 72 are disagreements. Items that are in agreement still need to be marked as 'complete' before they are moved to the agreed column. You can do this in the 'View' stats window by clicking on the 'Complete' button (to the right of the 59/72 button). Once you click on that button those 59 items will be marked as 'complete'. At that point you just have the 13/72 items to look at. Those 13 are disagreements and will need to be reconciled and marked as 'complete' individually (similar to what you did with those initial 49 items).

Although I haven't looked at the other group of coders I would suspect we are dealing with a similar situation.

Your confusion over this has drawn to our attention some things that we may need to address. In the 'View' stats window we probably need to be more specific about what 'Number of documents coded by' means. Perhaps we need to specify 'number of coded by' and 'number marked as complete' separately for each person. As well,  in the 'Run' quick report, perhaps the 'Agreed version' column should explain that these are the items marked as 'complete (and not necessarily items that were in agreement originally)'.

Best regards,

Jeff





 

 Hi Jeff,

I have read through your instructions and understand that we have what appears to be 13 disagreements remaining to reconcile. I can see those documents. In order to 'complete' the coding for the 59/72 - you wrote that we can simply click the "Complete" button beside it.

"...In the 'View' stats window you can see that 59 out of 72 items are in agreement and 13 out of 72 are disagreements. Items that are in agreement still need to be marked as 'complete' before they are moved to the agreed column. You can do this in the 'View' stats window by clicking on the 'Complete' button (to the right of the 59/72 button). Once you click on that button those 59 items will be marked as 'complete'..."

I am having some trouble with this because when I go to 'View' and click the "Complete" button to the right of (59/72) - it says "Complete. 0 Records Affected".

Please advise. Thank you!

Jessica Chan

Hello Jessica,

We looked into your issue a bit further and have found the reason that you were unable to 'complete' those items. In a software update that we did in July a issue was introduced to the completion code in the comparison table. In a particular scenario, if the item(s) to be completed was present on another comparison, using a different codeset and was already marked as completed in this second codeset, then it was getting overlooked by the completing function. This is why it was telling you 0 items were being completed when you tried to complete those 59 items.

We have made a correction to the program so you should now be able to complete those items. If you have any other problems please let us know.

Best regards,

Jeff