SPSSX Discussion

sample size for reliability analysis

Classic

List

Threaded

7 messages Options

Johnny Amora

sample size for reliability analysis

Dear all,

Your thoughts about sample size requirements in
conducting reliability analysis, particularly
cronbach's alpha are highly solicited.

Thanks in advance!

Johnny

____________________________________________________________________________________
Looking for last minute shopping deals?
Find them fast with Yahoo! Search. http://tools.search.yahoo.com/newsearch/category.php?category=shopping

=====================
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

Leah Quinlivan

Re: sample size for reliability analysis

Hi
Im only learning survey design at the moment, however, If your talking about
pre-testing a questionnaire and then conducting a reliability analysis, then

De Vaus, D. (2007). Surveys in Social Research suggests 75 to 100
respondents on a pre-test of a questionnaire (p. 117).

(I have to pretest my own and as its a sensitive sample / time
considerations am only using 30)

On 01/02/2008, John Amora <[hidden email]> wrote:

>
> Dear all,
>
> Your thoughts about sample size requirements in
> conducting reliability analysis, particularly
> cronbach's alpha are highly solicited.
>
> Thanks in advance!
>
> Johnny
>
>
>
> ____________________________________________________________________________________
> Looking for last minute shopping deals?
> Find them fast with Yahoo! Search.
> http://tools.search.yahoo.com/newsearch/category.php?category=shopping
>
> =====================
> 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
>

--
Leah Quinlivan

=====================
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

David Wasserman

Re: sample size for reliability analysis

In reply to this post by Johnny Amora

In my experience working with large-scale student assessments, we set a
minimum sample size of 250 to get stable results at the individual item
level.

----- Original Message -----
From: "John Amora" <[hidden email]>
To: <[hidden email]>
Sent: Friday, February 01, 2008 5:40 AM
Subject: sample size for reliability analysis

> Dear all,
>
> Your thoughts about sample size requirements in
> conducting reliability analysis, particularly
> cronbach's alpha are highly solicited.
>
> Thanks in advance!
>
> Johnny
>
>
>
> ____________________________________________________________________________________
> Looking for last minute shopping deals?
> Find them fast with Yahoo! Search.
> http://tools.search.yahoo.com/newsearch/category.php?category=shopping
>
> =====================
> 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

Leah Quinlivan

Re: sample size for reliability analysis

Ok, the most I can hope for is 60. Thus, for my own analysis and for
critically reviewing papers, when smaller than 250, can't be confident in
reliability? I have come across a number of journal articles using smaller
samples in pilots. Can you recommend any references as would help in my
studies

On 01/02/2008, David Wasserman <[hidden email]> wrote:

>
> In my experience working with large-scale student assessments, we set a
> minimum sample size of 250 to get stable results at the individual item
> level.
>
> ----- Original Message -----
> From: "John Amora" <[hidden email]>
> To: <[hidden email]>
> Sent: Friday, February 01, 2008 5:40 AM
> Subject: sample size for reliability analysis
>
>
> > Dear all,
> >
> > Your thoughts about sample size requirements in
> > conducting reliability analysis, particularly
> > cronbach's alpha are highly solicited.
> >
> > Thanks in advance!
> >
> > Johnny
> >
> >
> >
> >
> ____________________________________________________________________________________
> > Looking for last minute shopping deals?
> > Find them fast with Yahoo! Search.
> > http://tools.search.yahoo.com/newsearch/category.php?category=shopping
> >
> > =====================
> > 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
>

Matthias Spörrle

Re: sample size for reliability analysis

Charter, R. A. (2003). Study samples are too small to produce sufficiently
precise reliability coefficients. Journal of General Psychology, 130(2),
117-129.
HTH
Matthias

On Feb 1, 2008 6:36 PM, leah quinlivan <[hidden email]> wrote:

> Ok, the most I can hope for is 60. Thus, for my own analysis and for
> critically reviewing papers, when smaller than 250, can't be confident in
> reliability? I have come across a number of journal articles using smaller
> samples in pilots. Can you recommend any references as would help in my
> studies
>
>
> On 01/02/2008, David Wasserman <[hidden email]> wrote:
> >
> > In my experience working with large-scale student assessments, we set a
> > minimum sample size of 250 to get stable results at the individual item
> > level.
> >
> > ----- Original Message -----
> > From: "John Amora" <[hidden email]>
> > To: <[hidden email]>
> > Sent: Friday, February 01, 2008 5:40 AM
> > Subject: sample size for reliability analysis
> >
> >
> > > Dear all,
> > >
> > > Your thoughts about sample size requirements in
> > > conducting reliability analysis, particularly
> > > cronbach's alpha are highly solicited.
> > >
> > > Thanks in advance!
> > >
> > > Johnny
> > >
> > >
> > >
> > >
> >
> ____________________________________________________________________________________
> > > Looking for last minute shopping deals?
> > > Find them fast with Yahoo! Search.
> > > http://tools.search.yahoo.com/newsearch/category.php?category=shopping
> > >
> > > =====================
> > > 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
> >
>
>
>
> --
> Leah Quinlivan
>
> =====================
> 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
>

Leah Quinlivan

Re: sample size for reliability analysis

Thank-you kindly

leah

On 01/02/2008, Matthias Spörrle <[hidden email]> wrote:

>
> Charter, R. A. (2003). Study samples are too small to produce sufficiently
> precise reliability coefficients. Journal of General Psychology, 130(2),
> 117-129.
> HTH
> Matthias
>
>
>
> On Feb 1, 2008 6:36 PM, leah quinlivan <[hidden email]> wrote:
>
> > Ok, the most I can hope for is 60. Thus, for my own analysis and for
> > critically reviewing papers, when smaller than 250, can't be confident
> in
> > reliability? I have come across a number of journal articles using
> smaller
> > samples in pilots. Can you recommend any references as would help in my
> > studies
> >
> >
> > On 01/02/2008, David Wasserman <[hidden email]> wrote:
> > >
> > > In my experience working with large-scale student assessments, we set
> a
> > > minimum sample size of 250 to get stable results at the individual
> item
> > > level.
> > >
> > > ----- Original Message -----
> > > From: "John Amora" <[hidden email]>
> > > To: <[hidden email]>
> > > Sent: Friday, February 01, 2008 5:40 AM
> > > Subject: sample size for reliability analysis
> > >
> > >
> > > > Dear all,
> > > >
> > > > Your thoughts about sample size requirements in
> > > > conducting reliability analysis, particularly
> > > > cronbach's alpha are highly solicited.
> > > >
> > > > Thanks in advance!
> > > >
> > > > Johnny
> > > >
> > > >
> > > >
> > > >
> > >
> >
> ____________________________________________________________________________________
> > > > Looking for last minute shopping deals?
> > > > Find them fast with Yahoo! Search.
> > > >
> http://tools.search.yahoo.com/newsearch/category.php?category=shopping
> > > >
> > > > =====================
> > > > 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
> > >
> >
> >
> >
> > --
> > Leah Quinlivan
> >
> > =====================
> > 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
>

--
Leah Quinlivan

====================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

Simon, Steve, PhD

Re: sample size for reliability analysis

In reply to this post by Johnny Amora

John Amora writes:

> Your thoughts about sample size requirements in
> conducting reliability analysis, particularly
> cronbach's alpha are highly solicited.

I have a different take on this compared to the other respondents. Since
the goal of most reliability studies is to produce some sort of
correlation, the sample size is adequate if the confidence limits around
that correlation are reasonably narrow.

I elaborate a bit more about this at

www.childrensmercy.org/stats/weblog2007/ValidityReliabilitySampleSize.as
p

Steve Simon, [hidden email], Standard Disclaimer
Evidence Based Medicine gives my book 4/4.5 stars out of five!
Full text is at http://ebm.bmj.com/cgi/content/full/12/2/59

=====================
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