| Issue Number | 204 |
|---|---|
| Summary | [Reports] Citations Rejected/Accepted for Publishing - Add Grand Total |
| Created | 2014-06-09 17:31:29 |
| Issue Type | Improvement |
| Submitted By | Juthe, Robin (NIH/NCI) [E] |
| Assigned To | Kline, Bob (NIH/NCI) [C] |
| Status | Closed |
| Resolved | 2014-08-03 08:16:46 |
| Resolution | Fixed |
| Path | /home/bkline/backups/jira/oceebms/issue.129189 |
We would like to add a grand total of unique rejected citations to the citations rejected/accepted for publishing report (this report is under the Citation Reports menu).
This report currently provides the number of rejected and accepted citations by Board and for a given review cycle. However, the grand total is not a sum of these numbers. Instead, we'd like the grand total to reflect the total number of unique citations that were rejected for all summary topics during the medical librarian initial review.
Please add this total to the top of the report output when the report is generated for all Boards.
Adding Cynthia and Minaxi.
Ready for user review on DEV.
The grand total is displaying and seems to be calculating the total the way we want it to be calculated. Once this is in QA, we will be able to get a better idea if the total is correct by comparing it with previous statistics.
Cynthia and Minaxi have reviewed this and provided the following feedback:
Grand total rejected on the Rejected/Accepted report for Jan 2014 is
1671 when we generate this total using the “include rejected only” check
box on the search the database screen we get 1681. These numbers should
theoretically match.
How is this number being generated? The rejection is linked to the
summary topic therefore any given citation could have more than one
rejected topic for more than one board. Are we totaling the number of
rejections or the number of citations rejected? And if a citation has
one topic that is rejected and another from the same board that is
accepted, is it counted as rejected?
For our monthly report, we have been generating a grand total number of
rejected citations manually by subtracting the number published from the
number that first appeared in my queue just after import (which excludes
citations NOTed out by journal). Our total for Jan 2014 is 856. This is
the best we could come up with and it makes sense but what is this
number really? Generating totals for rejected citations with in each
board is an accurate statistic but once we started merging boards the
numbers loose meaning due to all the overlap between boards.
I believe the numbers should match, and that this is a bug in EbmsSearch (so, the number on the report page is right, and the result set for the search is wrong). We're checking for the state, and we're checking for the cycle, but we're not correlating the two. I'll either go ahead and try to fix this myself, or wait until I can go over it with Alan on Tuesday.
Good catch!
I have fixed the bug on DEV. I'll promote it to QA next week after I've had a chance to review my extensive rewrite of the search logic code to address the problem, and everyone agrees that it's working as expected on DEV.
However, there's one more wrinkle. The numbers are closer, but they don't match exactly. Now the search comes back with 1670 articles, compared with the 1671 given as the total rejected for the report. Here's why. For the search, we were given the requirement that we should only use 'rejected in initial review' states which are current. For the report, we weren't given that restriction. This actually seems to make sense, as the report and the search are answering two different questions ("Which articles were rejected in initial review for this cycle?" for the report, versus "Which articles are currently stuck at the state of having been rejected by the librarian?" for the search), the latter being more of an historical question, compared to the search, which is more interested in the current state of a particular article. The article which is reflected in the report, but which is not picked up by the search, is EBMS 320572, "NIH-defined graft-versus-host disease and evidence for a potent graft-versus-leukemia effect in patients with acute lymphoblastic leukemia." For the Pediatric Treatment board, Cynthia approved the article for the topic Childhood Acute Lymphoblastic Leukemia on January 6, but rejected it on the same day for the topic Childhood Hematopoietic Cell Transplantation. In spite of that initial rejection, Sharon approved the article for the Childhood Hematopoietic Cell Transplantation topic based on examination of the abstract on the 19th of January, Bonnie fetched the PDF for the full text of the article two days later, and then on the 24th Sharon looked at the full text and decided Cynthia was right originally, rejecting the article for that topic. So the report is correct in saying that it was rejected by the initial review, but the search is correct in saying that the article isn't in the set of articles whose current state is "rejected in initial review."
Does this make sense?
Please try some testing of the search page on DEV to make sure I haven't broken something.
Yes this makes sense. and I think it also makes sense that the report and search are asking different questions.
I did a bit of testing in Dev. I am getting the same results you are getting - 1670 & 1671. I did a few other searches and compared to the reports and am not seeing that you have broken anything. Minaxi can you do some testing as well to double check.
I did testing and everything seems to be fine.
Thanks to both of you for checking.
Verified on QA.
Verified on Prod.
Elapsed: 0:00:00.000949