SPSSX Discussion

Eta^2 in ANOVA

Classic

List

Threaded

5 messages Options

Johnny Amora

Eta^2 in ANOVA

Eta-squared is the proportion of variation in the dependent variable that is explained by the factor(independent variable). In the ANOVA outputs, Eta-squared can be easily computed by hand using the formula Eta^2 = SS(between) / SS(total).

I wonder why spss does not provide Eta^2 as part of the ANOVA output. Your comments are welcome.

Thanks.

Johnny

Bring your friends to the fun.
Invite your friends from Hotmail, Gmail to Yahoo! Mail today!

Marta Garcia-Granero

Re: Eta^2 in ANOVA

Johnny Amora wrote:

> Eta-squared is the proportion of variation in the dependent variable
> that is explained by the factor(independent variable). In the ANOVA
> outputs, Eta-squared can be easily computed by hand using the formula
> Eta^2 = SS(between) / SS(total).
>
> I wonder why spss does not provide Eta^2 as part of the ANOVA output.
> Your comments are welcome.
>
>
>
> ------------------------------------------------------------------------
> <http://sg.rd.yahoo.com/ph/mail/trueswitch/mailtagline/*https://secure5.trueswitch.com/yahoo-intl/?country=ph&language=en>

For oneway layouts, SPSS computes eta-square:

MEANS
TABLES=depvar BY factor
/CELLS MEAN COUNT STDDEV
/STATISTICS ANOVA .

For factorial designs, SPSS computes partial eta-square (not the same, I
know):

UNIANOVA
depvar BY factor1 factor2....
/PRINT = ETASQ
/DESIGN = factor1 factor2 factor1*factor2........ .

Anyway, have you tried using OMS?

Marta García-Granero

--
For miscellaneous 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

Johnny Amora

Re: Eta^2 in ANOVA

In reply to this post by Johnny Amora

Thanks, Marta,

I also got it using point-click. It is in analysis > compare means >...

Johnny

--- On Thu, 4/2/09, Marta García-Granero <[hidden email]> wrote:

From: Marta García-Granero <[hidden email]>
Subject: Re: Eta^2 in ANOVA
To: [hidden email]
Date: Thursday, 2 April, 2009, 3:07 PM

Johnny Amora wrote:

> Eta-squared is the proportion of variation in the dependent variable
> that is explained by the factor(independent variable). In the ANOVA
> outputs, Eta-squared can be easily computed by hand using the formula
> Eta^2 = SS(between) / SS(total).
>
> I wonder why spss does not provide Eta^2 as part of the ANOVA output.
> Your comments are welcome.
>
>
>
> ------------------------------------------------------------------------
> <http://sg.rd.yahoo.com/ph/mail/trueswitch/mailtagline/*https://secure5.trueswitch.com/yahoo-intl/?country=ph&language=en>
For oneway layouts, SPSS computes eta-square:

MEANS
TABLES=depvar BY factor
/CELLS MEAN COUNT STDDEV
/STATISTICS ANOVA .

For factorial designs, SPSS computes partial eta-square (not the same, I
know):

UNIANOVA
depvar BY factor1 factor2....
/PRINT = ETASQ
/DESIGN = factor1 factor2 factor1*factor2........ .

Anyway, have you tried using OMS?

Marta García-Granero

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

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Start chatting with friends on the all-new Yahoo! Pingbox today!!
It's easy to create your personal chat space on your blogs

Beckstead, Jason

Re: Eta^2 in ANOVA

In reply to this post by Johnny Amora

For the more complex ANOVA designs, use the MANOVA procedure with METHOD=SEQUENTIAL

and you will get eta2, partial-eta2 as well as omega2 for all sources.

_____________________________________________________________
Jason W. Beckstead, Ph.D.
Associate Professor/Quantitative Methodologist
University of South Florida College of Nursing
12901 Bruce B. Downs Blvd., MDC22, Tampa, FL 33612, USA
phone: +1.813.974.7667 fax: +1.813.974.5418
website: http://personal.health.usf.edu/jbeckste/

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Johnny Amora
Sent: Thursday, April 02, 2009 1:40 AM
To: [hidden email]
Subject: Eta^2 in ANOVA

I wonder why spss does not provide Eta^2 as part of the ANOVA output. Your comments are welcome.

Thanks.

Johnny

Bring your friends to the fun.
Invite your friends from Hotmail, Gmail to Yahoo! Mail today!

Johnny Amora

Re: Eta^2 in ANOVA

In reply to this post by Johnny Amora

This problem has already been solved. Thanks to everyone.

Johnny

--- On Thu, 4/2/09, Beckstead, Jason <[hidden email]> wrote:

From: Beckstead, Jason <[hidden email]>
Subject: Re: Eta^2 in ANOVA
To: [hidden email]
Date: Thursday, 2 April, 2009, 10:10 PM

For the more complex ANOVA designs, use the MANOVA procedure with METHOD=SEQUENTIAL

and you will get eta2, partial-eta2 as well as omega2 for all sources.

_____________________________________________________________
Jason W. Beckstead, Ph.D.
Associate Professor/Quantitative Methodologist
University of South Florida College of Nursing
12901 Bruce B. Downs Blvd., MDC22, Tampa, FL 33612, USA
phone: +1.813.974.7667 fax: +1.813.974.5418
website: http://personal.health.usf.edu/jbeckste/

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Johnny Amora
Sent: Thursday, April 02, 2009 1:40 AM
To: [hidden email]
Subject: Eta^2 in ANOVA

Eta-squared is the proportion of variation in the dependent variable that is explained by the factor(independent variable). In the ANOVA outputs, Eta-squared can be easily computed by hand using the formula Eta^2 = SS(between) / SS(total).

I wonder why spss does not provide Eta^2 as part of the ANOVA output. Your comments are welcome.

Thanks.

Johnny

Bring your friends to the fun.
Invite your friends from Hotmail, Gmail to Yahoo! Mail today!

Open emails faster.
Yahoo! recommends that you upgrade your browser to the new Internet Explorer 8 optimized for Yahoo!.Get it here! (It's free)