Your CIOS file request: Q-METHOD/06059133.007 hotline item

-
Received:  by CIOS Mailer; Friday 5 Jun 2009 13:30:07
Date:         Fri, 5 Jun 2009 18:28:34 +0100
From:         Sam Hopper 
Subject: Re: Analysis?
To:           Q-METHOD@LISTSERV.KENT.EDU
List-Help: ,              
List-Unsubscribe: 
List-Subscribe: 
List-Owner: 
List-Archive: 

--_4d8a617e-d9d4-4ce2-bb88-f99245962e11_
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable


Hi peter and Bob


Thank you so much for your replies! I rotated only 4 factors=2C but I think=
 what bob says could accunt for my data. Also=2C if Im thinking correctly (=
Its a friday evening afterall)=2C the pre rotated factors account for more =
than the 4 factors that are rotated... so thinking about PQMethod=2C you ob=
tain the eigenvalues first=2C then move onto choosing how many to rotate ba=
sed on eigenavalues more than 1 in statistical terms=2C so your both correc=
t I think? I may be wrong=2C as Im not at home and my data is else where so=
 cant check it!

=20

Another question can now crept into my mind about strong v's weak factors? =
I've not heard of this before=2C please could I be enlightened=2C again thi=
nking about viva time and what could come up!

=20

Thank you so much!

Sam



=20



=20

> Date: Fri=2C 5 Jun 2009 10:53:36 -0500
> From: bob.braswell  at AGMD.ORG
> Subject: Re: Analysis?
> To: Q-METHOD@LISTSERV.KENT.EDU
>=20
> Peter Schmolck wrote:
> > That's not possible=2C or should not be possible=2C in fact. If the
> > cumulative percentages of the unrotated factors above and the %'s
> > explained variance below originate from the same analysis the figures
> > must be the same=2C i.e. 41 + 26 + 4 + 5 (=3D76) should add up to 72.55=
=2C
> > within rounding error. Even if you rotated more than 4 factors but kept
> > only 4=2C the sum of expl. variances could be smaller only but not larg=
er
> > than the cumulative variance explained by the first 4 unrotated factors=
.=20
> >=20
> Sam's original post left out the number of factors rotated and the total
> of variance explained by that number=2C but it is clear=2C as Peter point=
s
> out=2C that the total variance at the beginning of the rotation has to be
> the same as the total variance at the end. If the data reported are
> correct and from the same analysis=2C then we must assume that Sam rotate=
d
> more than four factors.
>=20
> From that assumption=2C what Peter has written is almost correct=2C but
> overlooks one possibility. It is true that each of the first four
> factors extracted would have pulled out the most possible variance at
> each iteration=2C but it does not necessarily follow that there is no
> combination of four rotated factors that could add up to more total
> variance explained than the originally extracted four. Consider the
> case of PCA extraction followed by rotation of all 34 extracted
> components. During extraction=2C components 2-34 had to compete for 39%
> that remained after the first was extracted. But during rotation=2C with
> the first factor only pulling out 41% of the variance=2C 59% is available
> for 2-34. From this adjusted starting point=2C we can be certain that
> factor 2 of the rotated solution has the opportunity to explain more
> variance than factor 2 of the unrotated solution=2C and so on for the
> other factors. Thus what Sam reports is conceivable.
>=20
> Bob Braswell

_________________________________________________________________

MSN straight to your mobile - news=2C entertainment=2C videos and more.

http://clk.atdmt.com/UKM/go/147991039/direct/01/=

--_4d8a617e-d9d4-4ce2-bb88-f99245962e11_
Content-Type: text/html; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable






Hi peter and Bob


Thank you so much for your replies! I rotated only 4 factors=2C but I t=
hink what bob says could accunt for my data. Also=2C if Im thinking correct=
ly (Its a friday evening afterall)=2C the pre rotated factors account for m=
ore than the 4 factors that are rotated... so thinking about PQMethod=2C yo=
u obtain the eigenvalues first=2C then move onto choosing how many to rotat=
e based on eigenavalues more than 1 in statistical terms=2C so your both co=
rrect I think? I may be wrong=2C as Im not at home and my data is else wher=
e so cant check it!

 =3B

Another question can now crept into my mind about strong v's weak factors? =
I've not heard of this before=2C please could I be enlightened=2C again thi=
nking about viva time and what could come up!

 =3B

Thank you so much!

Sam



 =3B



 =3B

>=3B Date: Fri=2C 5 Jun 2009 10:53:36 -0500
>=3B From: bob.brasw=  at 
ell@AGMD.ORG
>=3B Subject: Re: Analysis?
>=3B To: Q-METHOD@LISTSE=
RV.KENT.EDU
>=3B 
>=3B Peter Schmolck wrote:
>=3B >=3B Tha=
t's not possible=2C or should not be possible=2C in fact. If the
>=3B =
>=3B cumulative percentages of the unrotated factors above and the %'s>=3B >=3B explained variance below originate from the same analysis th=
e figures
>=3B >=3B must be the same=2C i.e. 41 + 26 + 4 + 5 (=3D76)=
 should add up to 72.55=2C
>=3B >=3B within rounding error. Even if =
you rotated more than 4 factors but kept
>=3B >=3B only 4=2C the sum=
 of expl. variances could be smaller only but not larger
>=3B >=3B t=
han the cumulative variance explained by the first 4 unrotated factors. >=3B >=3B 
>=3B Sam's original post left out the number of factor=
s rotated and the total
>=3B of variance explained by that number=2C b=
ut it is clear=2C as Peter points
>=3B out=2C that the total variance =
at the beginning of the rotation has to be
>=3B the same as the total =
variance at the end. If the data reported are
>=3B correct and from th=
e same analysis=2C then we must assume that Sam rotated
>=3B more than=
 four factors.
>=3B 
>=3B From that assumption=2C what Peter has =
written is almost correct=2C but
>=3B overlooks one possibility. It is=
 true that each of the first four
>=3B factors extracted would have pu=
lled out the most possible variance at
>=3B each iteration=2C but it d=
oes not necessarily follow that there is no
>=3B combination of four r=
otated factors that could add up to more total
>=3B variance explained=
 than the originally extracted four. Consider the
>=3B case of PCA ext=
raction followed by rotation of all 34 extracted
>=3B components. Duri=
ng extraction=2C components 2-34 had to compete for 39%
>=3B that rema=
ined after the first was extracted. But during rotation=2C with
>=3B t=
he first factor only pulling out 41% of the variance=2C 59% is available>=3B for 2-34. From this adjusted starting point=2C we can be certain th=
at
>=3B factor 2 of the rotated solution has the opportunity to explai=
n more
>=3B variance than factor 2 of the unrotated solution=2C and so=
 on for the
>=3B other factors. Thus what Sam reports is conceivable.<=
BR>>=3B 
>=3B Bob Braswell

View your Twitter and Flic=
kr updates from one place - Learn more!
=

--_4d8a617e-d9d4-4ce2-bb88-f99245962e11_--