[prev in list] [next in list] [prev in thread] [next in thread] 

List:       sas-l
Subject:    Re: Signal Detection analyses
From:       Jay Weedon <jweedon () EARTHLINK ! NET>
Date:       1999-10-29 15:28:59
[Download RAW message or body]

On 14 Oct 1999 23:07:32 GMT, campbell@acs.ryerson.ca (Kent Campbell)
wrote:

>Francis C. Dane, Ph.D. (dane_fc@MERCER.EDU) wrote:
>: This is a multi-part message in MIME format.
>
>: ------=_NextPart_000_00D3_01BF164A.DC85A400
>: Content-Type: multipart/alternative;
>:         boundary="----=_NextPart_001_00D4_01BF164A.DC85A400"
>
>
>: ------=_NextPart_001_00D4_01BF164A.DC85A400
>: Content-Type: text/plain;
>:         charset="iso-8859-1"
>: Content-Transfer-Encoding: quoted-printable
>
>: One of our faculty members submitted the following request for help:
>
>: "I am looking for software to conduct a statistical procedure called a
>: signal detection analysis. Such an analysis allows an investigator to
>: get a measure of response bias (beta) and sensitivity (d prime) in a
>: variety of tasks. I am examining human memory performance on a
>: recognition test and would like to conduct the analysis based on hit and
>: false alarm rates."
>
>: I couldn't find anything in SAS help about d-prime or beta in this =
>: context.  Anyone have any suggestions?
>
>: Thanks,
>
>: Francis C. Dane, Ph.D.
>: Kilpatrick Professor & Chair
>: Mercer University Psychology
>[snip]
>
>     hi Francis et al.,
>                        this is sometimes called an ROC curve analysis.
>I don't have any stats software on my computer in this office so I can't
>be certain but I think SPSS has this function.  Don't know about SAS.
>For the details - many books on epidemiology cover this topic (it's used
>to evaluate diagnostic tests) or you could try a psychophysics book.  One
>of the classic psychophysics intro books is Green and Swets.  You could
>also try an intro biostatistics book.
>     If you need some exact titles and biblio info, send me an e-mail.

I've recently been told that ROC analysis is definitely available in
SPSS 9.

FWIW,
Jay.

[prev in list] [next in list] [prev in thread] [next in thread]