Tweeting all things #EvidenceSynthesis, quotes from #SysRev consultations, & critiquing published non-systematic #SysRev searches. I am 🇨🇦 #canmedlib #medlibs
Libraryland
Joined October 2021
- Tweets1.9K
- Following148
- Followers856
- Likes1.8K
Pinned Tweet
Happy Friday #medlibs #canmedlibs #SRlibs. All of these examples have been coming through my twitter feed - I couldn't help myself 🙃 #SNARC-SIG @carrieprice78
SRLibrarianProblems 🦇 retweeted
Human librarian vs Chatgpt4 vs custom gpt librarian bot at constructing keyword search strategy. The human librarian won comfortably..
Exploring the use of generative artificial intelligence in systematic searching: A comparative case study of a human librarian, ChatGPT-4 and ChatGPT-4 Turbo - Xiayu Summer Chen, Yali Feng, 2024 bit.ly/4cTXuEB
Interesting result!
Folks that help/teach others to do/find research, help me out with this poll on citation hallucinations from #ChatGPT.
Have you helped someone to track down an article fulltext based on a legit-looking citation only to find out the article simply does not exist?
37%Haven't come across it
30%Once
33%More times than I'd like
30 votes • Final resultsFolks that help/teach others to do/find research, help me out with this poll on citation hallucinations from #ChatGPT.
Have you helped someone to track down an article fulltext based on a legit-looking citation only to find out the article simply does not exist?
37%Haven't come across it
30%Once
33%More times than I'd like
30 votes • Final results👇
And, don't use #ChatGPT for your medical research either. I spent an hour helping someone look up fulltext for their citations - turns out they did not exist (though the journal, volume and issue # were all solid - there just wasn't an article by that name or author in it).
Pro Tip: DON'T use ChatGPT to translate your systematic review searches. Just don't. Talk to your librarian if you're unsure. #systematicreviews #medlibs #canmedlib
🤔 moment of the day:
Was working on #DataExtraction for an #EvidenceSynthesis project, and came across an "interesting" inclusion criteria: articles that have perfect structure.
Hmmm, I wonder how they evaluated perfect structure.
SRLibrarianProblems 🦇 retweeted
Guideline for reporting systematic reviews of outcome measurement instruments (OMIs): PRISMA-COSMIN for OMIs 2024
@JClinEpi
sciencedirect.com/science/ar…
I've seen lots of reviews that use only MeSH terms in their #Medline searches. This is a fantastic thread that demonstrates why you should not just search with #MeSH for your #SystematicReview #SearchStrategy
This Post is from an account that no longer exists. Learn more
Interesting read with lots of practical suggestions.
Large scoping reviews: managing volume and potential chaos in a pool of evidence sources: jclinepi.com/article/S0895-4…
#EvidenceSynthesis #ScopingReviews #Guidance
SRLibrarianProblems 🦇 retweeted
Rapid reviews: “at JBI we believe the simplest definition is that they are reviews characterized by the omission, abbreviation, or simplification of the traditional steps in a systematic review.”
Read more in JBI position statement: ow.ly/FEPe50Nbxoa
#JBIMethodology
SRLibrarianProblems 🦇 retweeted
Our very own Anthea Sutton has co-written a piece with @ucl's Veronica Parisi on ChatGPT and systematic literature searching for the Journal of @EAHIL. Read it here, along with lots of other content about AI and library and information science: ojs.eahil.eu/JEAHIL/issue/vi…
#Reproducible search #Reporting is not just for #EvidenceSynthesis reviews.
It is also a requirement for any scientific study about searching methods and #search efficiency. In that context, the search strategy is your data collection instrument.
How can you #PeerReview ...
...a study about search methods if the searches used are not provided?
How can you as a #Reviewer (especially #Reveiwer2) evaluate the study's methods and whether the conclusions are justified, if you cannot see the data collection instrument?
Research study: My milkshake is better than yours & will satisfy all your needs
Me: OK, let me try it & show me the recipe
Study: We do not have a sample or recipe to share with you. We tasted it so take our word for it - it is better
Me: 😶
#TransparentMethods are not optional
SRLibrarianProblems 🦇 retweeted
The 2024 edition of the JBI Manual for Evidence Synthesis includes major revision to reduce duplication of effort, updated methodological guidance, new information & resources!
Go to synthesismanual.jbi.global
#JBIMethodology #JBIEBHC
What (free) tools/software do you all use when creating process diagrams or flow charts? #MedLibs #CanMedLibs #Librarian #AcademicChatter
SRLibrarianProblems 🦇 retweeted
Beyond Google Scholar, Scopus, and Web of Science: An evaluation of the backward and forward citation coverage of 59 databases' citation indices - Gusenbauer - Research Synthesis Methods - Wiley Online Library onlinelibrary.wiley.com/doi/…
Wise people (in the past): We must minimize bias, so let's have two humans do the same task & compare their results, despite the time costs
Researchers testing GenAI: The system can screen the records in a fraction of the time, but the cost is a slight loss of accuracy
Me: 🤔😟😵💫
Just completed a #PRESS for a peer librarian - it was in a database that I'm familiar with but don't use for my SR work & it refreshed things for me. It also forced me to go to the DB search help page to find out if something was possible. #SearchNerd #CanMedLibs #MedLibs #SysRev
Love the changes to the #JBI Manual of Evidence Synthesis. Thanks to @katemsaylor for sharing & alerting me to it on the other platform. I appreciate the links to previously published guidance for the review type in each chapter. Ex. scoping reviews --> jbi-global-wiki.refined.site… 😍