Hi Joyce Typically we do a very quick cursory look P-values .25-.75 Discrimination (with our scanners algorithm) above .2 Harder question discriminating better hopefully :-) Basically its triage….if you are going to fix a couple of questions….do these ################## We use the Appersen scanners I believe their discrimination algorithm is on top 26% and bottom 26% jim -- Jim Sibley Director Centre for Instructional Support Faculty of Applied Science University of British Columbia 2205-6250 Applied Science Lane Vancouver, BC Canada V6T 1Z4 Phone 604.822.9241 Fax 604.822.7006 Email: [log in to unmask]<applewebdata:[log in to unmask]> Check out http://<http://cis.apsc.ubc.ca/>www.teambasedlearning.org © Copyright 2012, Jim Sibley, All rights reserved The information contained in this e-mail message and any attachments (collectively "message") is intended only for the personal and confidential use of the recipient (or recipients) named above. If the reader of this message is not the intended recipient, you are hereby notified that you have received this message in error and that any review, use, distribution, or copying of this message is strictly prohibited. If you have received this in error, please notify the sender immediately by e-mail, and delete the meesage. From: Joyce Vogler <[log in to unmask]<mailto:[log in to unmask]>> Reply-To: Joyce Vogler <[log in to unmask]<mailto:[log in to unmask]>> Date: Fri, 30 Nov 2012 08:23:23 -1000 To: <[log in to unmask]<mailto:[log in to unmask]>> Subject: Re: Item Analysis Hi Jim, How do you do item analysis when students can choose more than one choice? Do you use a Scantron machine. What do you do with the information afterward? Do you re-use questions for the next term if they meet a certain standard? Thank you for sharing. With warm aloha, Joyce Vogler, DrPH,APRN Associate Professor (808)956-5470 (808) 277-0877 cell(best number) On Nov 30, 2012, at 7:15 AM, Jim Sibley wrote: Hi we run item analysis on the whole cohort since our scanner pumps out discrimination, point bi-serial and p-value I don't see any problem with using only 10% of the group either jim On Fri, Nov 30, 2012 at 8:13 AM, Jenny Morris <[log in to unmask]<mailto:[log in to unmask]>> wrote: Dear All As this is the first time we have been using TBL I plan to run an item analysis on the questions we are using. We have around 300 students in the group and I was considering basing this analysis on 10% of the group. I’d welcome feedback from others who have undertaken similar work. Thanks and best wishes Jenny Dr Jenny Morris Associate Professor (Senior Lecturer) in Health Studies Faculty of Health, Education and Society University of Plymouth Knowledge Spa Treliske Truro TR1 3HD Cornwall Tel: 01872 256461 Web: http://www.plymouth.ac.uk/faculties/health ‘High quality education for high quality care’ -- Jim Sibley [log in to unmask]<mailto:[log in to unmask]> NEW Home Phone 604-564-1043 Work 604-822-9241