Evidence Synthesis Methods Interest Group

Hi, all,

I have faculty and students who are starting to use Rayyan for their systematic review screening, and looking for guidance from me. My initial understanding of a feasible workflow would be something like this, and would take place after they'd done all their searches, and imported all the records into a citation manager (RefWorks, generally, although some may be using Zotero):

1. Complete the deduplication process within their citation manager.
2. Import the deduplicated set into Rayyan.
3. Complete all screening.
4. Export the set of records they've selected for inclusion into a separate RefWorks project library or a Zotero Group (to ensure it remains walled off from their original search set and/or any other projects they may be working on).

It's mainly step 4 I'm trying to confirm. If you use Rayyan for screening, is this the process you'd use? I'm open to any recommendations you may have!

Thanks,

Hilary

------------------------------
Hilary Kraus
hilary.kraus@uconn.edu
Research Services Librarian
University of Connecticut
She/Her/Hers
------------------------------

Thanks, Amy, this is really helpful!
-Hilary

------------------------------
Hilary Kraus
hilary.kraus@uconn.edu
Research Services Librarian
University of Connecticut
She/Her/Hers
------------------------------

Original Message:
Sent: Jun 28, 2021 08:30 PM
From: Amy Riegelman
Subject: workflow for using Rayyan

Hi! I haven't used RefWorks in many years because we dropped our site license. Regardless, here is a common workflow for my teams---

1. Complete the deduplication process within their citation manager.
2. Import the deduplicated set into Rayyan.
3. Complete all screening  title/abstract screening in Rayyan
   
4. Export included items into a fresh, new Rayyan project.
5. Obtain fulltexts and complete fulltext screening
   
6. Export the set of records they've selected for inclusion into a separate RefWorks project library or a Zotero Group (to ensure it remains walled off from their original search set and/or any other projects they may be working on).

--

Amy Riegelman (she, her, hers)
Social Sciences Librarian

University of Minnesota
aspringe@umn.edu
http://orcid.org/0000-0003-4127-5222

New Article! Is the open access citation advantage real? A systematic review of the citation of open access and subscription-based articles 

Hillary,

I would add 2a: Rayyan will find additional duplicates, but many of them will be false dups because Rayyan seems to match for dups using the beginning of the title (i.e., articles that are Part 1, Part 2,  or have a one or two word difference toward the end of the title, will show up as dups). So someone on the team should take care of Rayyan's suggested duplicates (and who knows, Rayyan might catch something....but Rayyan has never caught a true duplicate if I used Bramer's method in Endnote to remove duplicates).

For #4, yes, export "included" records to citation manager and work on obtaining the PDFs. If there is a group citation manager library, everyone can access the PDFs from there (I've never attempted adding PDFs back into Rayyan). ALSO: I recommend that the team create a NEW Rayyan project with the same exported records for full-text screening.

^^^This method worked well with a large team with many students (i.e., doing title-abstract screening in one Rayyan project, then doing full-text screening in a separate Rayyan project, and THEN doing the data extraction). I worked with another, more advanced team of researchers, and that group's choice was to put all the PDFs included after TI-AB screening into a shared folder, and then the team members conducted full-text screening AND data extraction in one step.
​
Looking forward to reading other responses.

Jane

------------------------------
Jane Yatcilla
Health & Life Sciences Information Specialist
Purdue University Libraries
------------------------------

Hi, Jane,

Thanks so much for this reply! Your suggestions are really helpful.

-Hilary

------------------------------
Hilary Kraus
hilary.kraus@uconn.edu
Research Services Librarian
University of Connecticut
She/Her/Hers
------------------------------

Original Message:
Sent: Jun 29, 2021 06:46 AM
From: Jane Yatcilla
Subject: workflow for using Rayyan

Hillary,

I would add 2a: Rayyan will find additional duplicates, but many of them will be false dups because Rayyan seems to match for dups using the beginning of the title (i.e., articles that are Part 1, Part 2,  or have a one or two word difference toward the end of the title, will show up as dups). So someone on the team should take care of Rayyan's suggested duplicates (and who knows, Rayyan might catch something....but Rayyan has never caught a true duplicate if I used Bramer's method in Endnote to remove duplicates).

For #4, yes, export "included" records to citation manager and work on obtaining the PDFs. If there is a group citation manager library, everyone can access the PDFs from there (I've never attempted adding PDFs back into Rayyan). ALSO: I recommend that the team create a NEW Rayyan project with the same exported records for full-text screening.

^^^This method worked well with a large team with many students (i.e., doing title-abstract screening in one Rayyan project, then doing full-text screening in a separate Rayyan project, and THEN doing the data extraction). I worked with another, more advanced team of researchers, and that group's choice was to put all the PDFs included after TI-AB screening into a shared folder, and then the team members conducted full-text screening AND data extraction in one step.
​
Looking forward to reading other responses.

Jane

------------------------------
Jane Yatcilla
Health & Life Sciences Information Specialist
Purdue University Libraries
------------------------------

Thanks to everyone who responded to this (privately and via Connect). I wanted to post my revised, much more detailed workflow, in case it's helpful to others!

In your preferred citation management tool (such as RefWorks or Zotero):

1. Collect all references.
2. Record necessary numbers for your PRISMA diagram.
3. Perform deduplication of references.
4. Record number of references in deduplicated set.
5. Import the deduplicated set into a new Rayyan project.

In Rayyan:

1. Complete all title/abstract screening.
2. Record number of references in included set (if you found additional duplicates during title/abstract screening, be sure your numbers reflect this).
3. Export included items to a new Rayyan project.
4. Obtain full text of references and make available to screeners (a shared cloud storage folder outside of Rayyan is a good place to store full text for the team).
5. Complete full text screening.
6. Record final number of references included.
7. Export included records into your preferred citation manager as a separate project library (RefWorks) or a Group Library (Zotero) to ensure they remain walled off from your original set of references and any other projects you're working on.

------------------------------
Hilary Kraus
hilary.kraus@uconn.edu
Research Services Librarian
University of Connecticut
She/Her/Hers
------------------------------