Dear Liz
Probably the best known way of doing this is the sweets in a bag
demonstration. This is described in full in the very useful Testing
Treatments book on the James Lind site page 56 Acrobat file (page 39 in
print version):
http://www.jameslindlibrary.org/pdf/testing-treatments.pdf
For consolidation you might find that the following for Learners and
Teachers respectively are useful. They are from the CMAJ Teaching Tips
and Learning Tips article series:
Learners:
http://www.cmaj.ca/cgi/content/full/171/6/611?etoc
Teachers:
http://www.cmaj.ca/cgi/data/171/6/611/DC1/1
I sometimes use the analogy of the fishermen when asked "How big was the
fish you caught?" If they have put the fish back you cannot
authoritatively establish its size. So in the absence of complete
information you would have to go on how wide apart their hands are . If
of course their estimates overlap then this collection of independent
samples (studies) lead you to believe that the true sixe of the fish lay
within the overlap range. (If you had 100 fishermen then your 95%
confidence interval would be where 95 of the estimates overlap ie. you
would eliminate the ones who obviously estimated too big and too small).
Let me know if you come up with anything better!
Andrew
Best wishes
Hunwick, Liz wrote:
>
> Has anyone out there devised a simple illustration/explanation of
> Confidence Intervals to use when doing CA courses? It’s the one
> concept I find very difficult to explain, probably because my own
> understanding is limited.
>
> I’d be grateful for any ideas.
>
> Thanks,
>
> *Liz Hunwick* MA, PG Dip LIS, MCLIP
>
> Librarian: Services
>
> Basildon Healthcare Library
>
> Education Centre
>
> Basildon University Hospital
>
> Nethermayne
>
> Basildon
>
> Essex SS16 5NL
>
> *[log in to unmask] <mailto:[log in to unmask]>*
>
> 01268 524900 ext 8272
>
> *Look inside your Library*
>
> Access all our services and resources at our redesigned website
> *_www.btuheks.nhs.uk_* *__*
>
> ============================================================================
> This email has been processed by the Basildon and Thurrock University Hospitals NHS Foundation Trust email gateway.
> Any executable files or known viruses have been removed.
> If this message has been sent to you in error please securely discard the email, notify the sender and do not open any attachments or act on the contents.
> This email may be subject to the Data Protection Act 1998, the Freedom of Information Act 2000 and other relevant acts and regulations. Unless the contents are clinically confidential or otherwise exempt they may be subject to disclosure.
> We do not accept or send emails larger than 10MB.
> For more information contact the IT Department, Basildon and Thurrock University Hospitals NHS Foundation Trust on +44 1268 82 19 19
> ============================================================================
>
>
--
Andrew Booth BA Dip Lib MSc MCLIP
Reader in Evidence Based Information Practice
Director of Information
School of Health & Related Research (ScHARR)
University of Sheffield, Regent Court,
30 Regent Street, SHEFFIELD,
S1 4DA
Tel: +44(0)114 244 3622
Fax: +44(0)114 272 4095
Email: [log in to unmask]
“First you're an unknown, then you write one book and you move up to obscurity"-- Martin Myers
|