SPSSX Discussion

Re: SPSSX-L Digest - 18 Sep 2009 to 19 Sep 2009 (#2009-263)

Classic

List

Threaded

6 messages Options

Jeff Galecki-2

Re: SPSSX-L Digest - 18 Sep 2009 to 19 Sep 2009 (#2009-263)

Hi Marta,
I run PASW 17 on Windows 7 x32 & x64 (two separate computers). I have
found no issues with running PASW on either system. I also run Viz
Designer on the 32 bit Windows 7 computer in case anyone wanted to
know what other PASW products I have had working on Windows 7.

Jeff

On Sep 20, 2009, at 12:02 AM, Automatic digest processor wrote:

> There are 6 messages totalling 393 lines in this issue.
>
> Topics of the day:
>
> 1. SPSS 15 or PASW 17 on Windows 7? (3)
> 2. Calculate probability of T value (2)
> 3. Teaching survey analysis with SPSS
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
> ----------------------------------------------------------------------
>
> Date: Sat, 19 Sep 2009 11:33:10 +0200
> From: =?ISO-8859-1?Q?Marta_Garc=EDa-Granero?= <[hidden email]
> >
> Subject: SPSS 15 or PASW 17 on Windows 7?
>
> Hi everybody on this cloudy Saturday morning (at least on Spain...):
>
> I am playing a bit with Windows 7, installed in the second hard
> drive of
> my old 32bits processor computer (yes, I know, no dual core, no gain
> in
> speed, I'm just testing improved stability), and I was wondering if
> anyone had any experience on installing/running SPSS15 or PASW17 (the
> two versions I have access to) on Windows 7, and would not mind
> sharing
> it with me. Is it possible, or should I wait for PASW18?
>
> Best regards,
> Marta GG
>
> --
> For miscellaneous SPSS related statistical stuff, visit:
> http://gjyp.nl/marta/
>
> ------------------------------
>
> Date: Sat, 19 Sep 2009 10:55:31 +0100
> From: Garry Gelade <[hidden email]>
> Subject: Re: SPSS 15 or PASW 17 on Windows 7?
>
> Dear Marta
>
> I'm also on SPSS15, and would be interested to know in general which
> versions of which SPSS products will run under Windows 7.
> Specifically,
> SPSS15, Clementine 12, and earlier versions of AMOS.
>
> Garry Gelade
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On
> Behalf Of
> Marta García-Granero
> Sent: 19 September 2009 10:33
> To: [hidden email]
> Subject: SPSS 15 or PASW 17 on Windows 7?
>
> Hi everybody on this cloudy Saturday morning (at least on Spain...):
>
> I am playing a bit with Windows 7, installed in the second hard
> drive of my
> old 32bits processor computer (yes, I know, no dual core, no gain in
> speed,
> I'm just testing improved stability), and I was wondering if anyone
> had any
> experience on installing/running SPSS15 or PASW17 (the two versions
> I have
> access to) on Windows 7, and would not mind sharing it with me. Is it
> possible, or should I wait for PASW18?
>
> Best regards,
> Marta GG
>
> --
> For miscellaneous SPSS related statistical stuff, visit:
> http://gjyp.nl/marta/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the
> command. To leave the list, send the command SIGNOFF SPSSX-L For a
> list of
> commands to manage subscriptions, send the command INFO REFCARD
>
> __________ Information from ESET NOD32 Antivirus, version of virus
> signature
> database 4439 (20090918) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
> __________ Information from ESET NOD32 Antivirus, version of virus
> signature
> database 4439 (20090918) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
> ------------------------------
>
> Date: Sat, 19 Sep 2009 11:27:49 +0100
> From: "Lemon, John S." <[hidden email]>
> Subject: Re: SPSS 15 or PASW 17 on Windows 7?
>
> As far as I know only versions post-16 will work on Vista and 7 - I
> think 15 would work with a 'fix'
>
> Best Wishes
>
> John S. Lemon
> DIT ( Directorate of Information Technology ) - Student Liaison
> Officer
> University of Aberdeen
> Edward Wright Building: Room G51
>
> Tel: +44 1224 273350
> Fax: +44 1224 273372
>
> Diary ( Free / Busy )
>
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On
> Behalf Of Garry Gelade
> Sent: 19 September 2009 10:56
> To: [hidden email]
> Subject: Re: SPSS 15 or PASW 17 on Windows 7?
>
> Dear Marta
>
> I'm also on SPSS15, and would be interested to know in general which
> versions of which SPSS products will run under Windows 7.
> Specifically,
> SPSS15, Clementine 12, and earlier versions of AMOS.
>
> Garry Gelade
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On
> Behalf Of
> Marta García-Granero
> Sent: 19 September 2009 10:33
> To: [hidden email]
> Subject: SPSS 15 or PASW 17 on Windows 7?
>
> Hi everybody on this cloudy Saturday morning (at least on Spain...):
>
> I am playing a bit with Windows 7, installed in the second hard
> drive of my
> old 32bits processor computer (yes, I know, no dual core, no gain in
> speed,
> I'm just testing improved stability), and I was wondering if anyone
> had any
> experience on installing/running SPSS15 or PASW17 (the two versions
> I have
> access to) on Windows 7, and would not mind sharing it with me. Is it
> possible, or should I wait for PASW18?
>
> Best regards,
> Marta GG
>
> --
> For miscellaneous SPSS related statistical stuff, visit:
> http://gjyp.nl/marta/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the
> command. To leave the list, send the command SIGNOFF SPSSX-L For a
> list of
> commands to manage subscriptions, send the command INFO REFCARD
>
> __________ Information from ESET NOD32 Antivirus, version of virus
> signature
> database 4439 (20090918) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
>
>
>
> __________ Information from ESET NOD32 Antivirus, version of virus
> signature
> database 4439 (20090918) __________
>
> The message was checked by ESET NOD32 Antivirus.
>
> http://www.eset.com
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>
>
> The University of Aberdeen is a charity registered in Scotland, No
> SC013683.
>
> ------------------------------
>
> Date: Sat, 19 Sep 2009 09:15:17 -0400
> From: Brock <[hidden email]>
> Subject: Calculate probability of T value
>
> I know this is a very basic question, but I have summary statistics
> only (not
> the raw data) and I want to do a one sample t-test. What is the
> best way to
> go about this?
>
> Calculate T and then find the probability of that statistic? I know
> I can use
> TDIST in Excel, but what is the equvialent in SPSS? I am trying to
> work
> through some examples online and I am not getting the same
> results....obviously it is me and not SPSS.
>
> In short, I actually have never calculated p by hand summarized data
> and
> simply want to give it a shot.
>
> Any help will be greatly appreciated!
>
> Brock
>
> ------------------------------
>
> Date: Sat, 19 Sep 2009 14:14:39 +0000
> From: Ruben van den Berg <[hidden email]>
> Subject: Re: Calculate probability of T value
>
> --_50c76fdc-2a4e-45e0-831f-950de689e0ff_
> Content-Type: text/plain; charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>
>
> Dear Brock=2C
>
> =20
>
> This seems to work (2 sided probability):
>
> =20
>
> data lis free/tvalue df.
>
> begin data
>
> 1 1 1=2C96 1000
>
> end data.
>
> comp sig=3D2*(1-cdf.t(tvalue=2Cdf)).
>
> exe.
>
>
> Kind regards=2C
>
> =20
>
> Ruben van den Berg
>
>
> =20
>
>
>
> =20
>> Date: Sat=2C 19 Sep 2009 09:15:17 -0400
>> From: [hidden email]
>> Subject: Calculate probability of T value
>> To: [hidden email]
>> =20
>> I know this is a very basic question=2C but I have summary
>> statistics onl=
> y (not
>> the raw data) and I want to do a one sample t-test. What is the
>> best way =
> to
>> go about this?
>> =20
>> Calculate T and then find the probability of that statistic? I know
>> I can=
> use
>> TDIST in Excel=2C but what is the equvialent in SPSS? I am trying
>> to work
>> through some examples online and I am not getting the same
>> results....obviously it is me and not SPSS.
>> =20
>> In short=2C I actually have never calculated p by hand summarized
>> data an=
> d
>> simply want to give it a shot.
>> =20
>> Any help will be greatly appreciated!
>> =20
>> Brock
>> =20
>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>> To manage your subscription to SPSSX-L=2C send a message to
>> [hidden email] (not to SPSSX-L)=2C with no body text
>> except th=
> e
>> command. To leave the list=2C send the command
>> SIGNOFF SPSSX-L
>> For a list of commands to manage subscriptions=2C send the command
>> INFO REFCARD
>
> _________________________________________________________________
> Express yourself instantly with MSN Messenger! Download today it's
> FREE!
> http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/=
>
> --_50c76fdc-2a4e-45e0-831f-950de689e0ff_
> Content-Type: text/html; charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>
> <html>
> <head>
> <style>
> .hmmessage P
> {
> margin:0px=3B
> padding:0px
> }
> body.hmmessage
> {
> font-size: 10pt=3B
> font-family:Verdana
> }
> </style>
> </head>
> <body class=3D'hmmessage'>
> Dear Brock=2C 
> &nbsp=3B 
> This seems to work (2 sided probability): 
> &nbsp=3B 
> data lis free/tvalue df. 
> begin data 
> 1 1 1=2C96 1000 
> end data. 
> comp sig=3D2*(1-cdf.t(tvalue=2Cdf)). 
> exe. 
> Kind regards=2C 
> &nbsp=3B 
> Ruben van den
> Berg &nbsp=3B &nbsp=3B &gt=3B Da=
> te: Sat=2C 19 Sep 2009 09:15:17 -0400 &gt=3B From:
> [hidden email]<B=
> R>&gt=3B Subject: Calculate probability of T value &gt=3B To:
> SPSSX-L@LI=
> STSERV.UGA.EDU &gt=3B &gt=3B I know this is a very basic
> question=2C=
> but I have summary statistics only (not &gt=3B the raw data) and
> I want=
> to do a one sample t-test. What is the best way to &gt=3B go
> about this=
> ? &gt=3B &gt=3B Calculate T and then find the probability of
> that st=
> atistic? I know I can use &gt=3B TDIST in Excel=2C but what is
> the equvi=
> alent in SPSS? I am trying to work &gt=3B through some examples
> online a=
> nd I am not getting the same &gt=3B results....obviously it is me
> and no=
> t SPSS. &gt=3B &gt=3B In short=2C I actually have never
> calculated p=
> by hand summarized data and &gt=3B simply want to give it a
> shot. &g=
> t=3B &gt=3B Any help will be greatly appreciated! &gt=3B
> &gt=3B =
> Brock &gt=3B &gt=3B
> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
> =3D=3D=3D=3D=3D &gt=3B To manage your subscription to SPSSX-L=2C
> send a =
> message to &gt=3B [hidden email] (not to SPSSX-L)=2C
> with no =
> body text except the &gt=3B command. To leave the list=2C send
> the comma=
> nd &gt=3B SIGNOFF SPSSX-L &gt=3B For a list of commands to
> manage sub=
> scriptions=2C send the command &gt=3B INFO REFCARD <hr /
> >Expres=
> s yourself instantly with MSN Messenger! <a href=3D'http://
> clk.atdmt.com/AV=
> E/go/onm00200471ave/direct/01/' target=3D'_new'>MSN Messenger</a></
> body>
> </html>=
>
> --_50c76fdc-2a4e-45e0-831f-950de689e0ff_--
>
> ------------------------------
>
> Date: Sat, 19 Sep 2009 22:14:56 +0200
> From: John F Hall <[hidden email]>
> Subject: Teaching survey analysis with SPSS
>
> This is a multi-part message in MIME format.
>
> ------=_NextPart_000_00C1_01CA3976.9EADE0C0
> Content-Type: text/plain;
> charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>
> Lots of stats out there and one or two interesting logic and
> programming =
> problems, but is there anyone who teaches how to use SPSS, not for =
> statistics but for processing and analysing data from questionnaire =
> surveys. After all, that's how SPSS started. If so, I'd be happy
> to =
> swap experiences and materials, in my case from 1972 (blue manual)
> to 15 =
> for Windows today.
> ------=_NextPart_000_00C1_01CA3976.9EADE0C0
> Content-Type: text/html;
> charset="iso-8859-1"
> Content-Transfer-Encoding: quoted-printable
>
> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
> <HTML><HEAD>
> <META http-equiv=3DContent-Type content=3D"text/html; =
> charset=3Diso-8859-1">
> <META content=3D"MSHTML 6.00.6000.16890" name=3DGENERATOR>
> <STYLE></STYLE>
> </HEAD>
> <BODY bgColor=3D#ffffff>
> <DIV>Lots of stats out there and one or
> two =
> interesting=20
> logic and programming problems, but is there anyone who
> teaches=20
> how to use SPSS, not for statistics but for processing and =
> analysing=20
> data from questionnaire surveys.  After all, that's how SPSS =
> started. =20
> If so, I'd be happy to swap experiences and materials, in my case
> from =
> 1972=20
> (blue manual) to 15 for Windows today.</DIV></BODY></HTML>
>
> ------=_NextPart_000_00C1_01CA3976.9EADE0C0--
>
> ------------------------------
>
> End of SPSSX-L Digest - 18 Sep 2009 to 19 Sep 2009 (#2009-263)
> **************************************************************

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Handel, Richard W.

McNemar Test

Re: SPSSX-L Digest - 18 Sep 2009 to 19 Sep 2009 (#2009-263)

Hi All,

A colleague has asked for assistance in proposing an analysis for a 2X2 contigency table for matched pairs. I know that the McNemar test is normally appropriate, but at most, 20 matched pairs (40 subjects total) will be available for analysis. I am concerned that some cells may have very low frequency counts. What are my options?

Thanks!

Rick

ajayohri

Re: McNemar Test

T test matched pairs

Sent from my iPhone

On Sep 20, 2009, at 12:59 PM, "Handel, Richard W." <[hidden email]> wrote:

Hi All,

A colleague has asked for assistance in proposing an analysis for a 2X2 contigency table for matched pairs. I know that the McNemar test is normally appropriate, but at most, 20 matched pairs (40 subjects total) will be available for analysis. I am concerned that some cells may have very low frequency counts. What are my options?

Thanks!

Rick

Marta Garcia-Granero

Re: McNemar Test

In reply to this post by Handel, Richard W.

Hi Richard

Handel, Richard W. wrote:
> A colleague has asked for assistance in proposing an analysis for a
> 2X2 contigency table for matched pairs. I know that the McNemar test
> is normally appropriate, but at most, 20 matched pairs (40 subjects
> total) will be available for analysis. I am concerned that some cells
> may have very low frequency counts. What are my options?
If you are concerned for the asymptotic (chi-square) test validity,
don't worry, because the program (SPSS) will compute an exact p-value if
necessary. It even computes exact p-values in conditions where
asymptotic testing would be OK (with at less 10 discordant pairs). Exact
p-values are bit too conservative.

If you are concerned for the actual power of the test with such a low
sample size, that's another problem, one with difficult solution, since
the sample size can't be modified...

HTH,
Marta GG

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

Marta Garcia-Granero

Re: McNemar Test

In reply to this post by ajayohri

Hi everybody

AJay Ohri wrote:
> T test matched pairs
>
> Sent from my iPhone

Definitely NOT a solution, since McNemar tests handles BINARY outcomes,
not continuous.

Sent from my desktop computer.

MGG

>
> On Sep 20, 2009, at 12:59 PM, "Handel, Richard W." <[hidden email]
> <mailto:[hidden email]>> wrote:
>
>> Hi All,
>>
>> A colleague has asked for assistance in proposing an analysis for a
>> 2X2 contigency table for matched pairs. I know that the McNemar test
>> is normally appropriate, but at most, 20 matched pairs (40 subjects
>> total) will be available for analysis. I am concerned that some
>> cells may have very low frequency counts. What are my options?
>>
>> Thanks!
>>
>> Rick
>>
>>

Evan Harrington, Ph.D.

Re: McNemar Test

Conover (1999) is a good source for the assumptions and/or limitations of nonparametric stats...but Conover says nothing about low sample size being problematic for McNemar tests.

Daniel (1978) in "Applied Nonparametric Statistics" notes that low sample size can cause distorted P values, and mentions that some favor "Gart's Test". I have never used Gart's test, but it appears to be intended to correct the issue of low sample sizes in McNemar.

Evan R. Harrington, Ph.D.
Associate Professor
Forensic Thesis Track Director
The Chicago School of Professional Psychology
Department of Forensic Psychology
325 North Wells Street
Chicago, IL 60654

Phone: 312 329-6693
Fax: 312 661-1272

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Marta García-Granero
Sent: Sunday, September 20, 2009 12:31 PM
To: [hidden email]
Subject: Re: McNemar Test

Hi everybody

AJay Ohri wrote:
> T test matched pairs
>
> Sent from my iPhone

Definitely NOT a solution, since McNemar tests handles BINARY outcomes,
not continuous.

Sent from my desktop computer.

MGG

--
For miscellaneous SPSS related statistical stuff, visit:
http://gjyp.nl/marta/

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD