SPSSX Discussion

Nice improvements to UNIANOVA in SPSS 25

Classic

List

Threaded

3 messages Options

Bruce Weaver

Nice improvements to UNIANOVA in SPSS 25

Administrator

While working on an assignment for a 4th year class (using v25), I noticed
that the PLOT dialog for UNIANOVA now allows the user to choose either a
line graph (the only option available previously) or a bar graph. About
time, I thought, as many of us teach students that line graphs are only
appropriate when the X-axis variable is quantitative in nature. It turns
out this is not the only nice improvement to UNIANOVA. See below for the
complete list of changes.

Changes to UNIANOVA in Release 25.0

TYPE, ERRORBAR, MEANREFERENCE, and YAXIS keywords added to PLOT
subcommand.
ROBUST subcommand added.
HCCOVB keyword added to OUTFILE subcommand. Any combination of COVB,
CORB, and HCCOVB keywords can now be specified on OUTFILE.
MBP, BP, F, and WHITE keywords added to PRINT subcommand.
MBPDESIGN, BPDESIGN, and FDESIGN subcommands added.

For more info:

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova.html

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_plot.html

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_robust.html

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_outfile.html

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_print.html

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_hc_designs.html

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.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

--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).

Bruce Weaver

Re: Nice improvements to UNIANOVA in SPSS 25

Administrator

As Jon Peck pointed out to me in a private message, all of these features
have also been added to GLM.

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/advanced/syn_glm_overview.html

Bruce Weaver wrote

> While working on an assignment for a 4th year class (using v25), I noticed
> that the PLOT dialog for UNIANOVA now allows the user to choose either a
> line graph (the only option available previously) or a bar graph. About
> time, I thought, as many of us teach students that line graphs are only
> appropriate when the X-axis variable is quantitative in nature. It turns
> out this is not the only nice improvement to UNIANOVA. See below for the
> complete list of changes.
>
>
> Changes to UNIANOVA in Release 25.0
>
> TYPE, ERRORBAR, MEANREFERENCE, and YAXIS keywords added to PLOT
> subcommand.
> ROBUST subcommand added.
> HCCOVB keyword added to OUTFILE subcommand. Any combination of COVB,
> CORB, and HCCOVB keywords can now be specified on OUTFILE.
> MBP, BP, F, and WHITE keywords added to PRINT subcommand.
> MBPDESIGN, BPDESIGN, and FDESIGN subcommands added.
>
> For more info:
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova.html
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_plot.html
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_robust.html
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_outfile.html
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_print.html
>
> https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_hc_designs.html
>
>
>
>
>
>
> -----
> --
> Bruce Weaver

> bweaver@

> http://sites.google.com/a/lakeheadu.ca/bweaver/
>
> "When all else fails, RTFM."
>
> NOTE: My Hotmail account is not monitored regularly.
> To send me an e-mail, please use the address shown above.
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

> (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

-----
--
Bruce Weaver
[hidden email]
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.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

Bruce Weaver

Re: Nice improvements to UNIANOVA in SPSS 25

Administrator

In reply to this post by Bruce Weaver

As pleased as I was to see the ROBUST sub-command added to UNIANOVA (and GLM)
in v25, I was quite quite surprised and ~disappointed~ to discover that
contrasts carried out via EMMEANS with COMPARE do not use the robust
standard errors. (I discovered this while cobbling together a toy example
for a student.) In order to obtain contrasts that use the robust SEs, one
must use LMATRIX. (And if you've ever used LMATRIX, I suspect you'll agree
it's considerably trickier than EMMEANS with COMPARE.)

What I expected to find was that if I use a ROBUST sub-command, all tests
and contrasts would use the robust SEs. But instead, UNIANOVA (and GLM)
give the usual output (with asymptotic SEs--ASEs) and then tack on some more
output using robust SEs. From the FM:

--- start of excerpt ---
ROBUST displays a table of parameter estimates, along with robust or
heteroskedasticity-consistent (HC) standard errors; and t statistics,
significance values, and confidence intervals that use the robust standard
errors.

- ROBUST enables specification of the HCCOVB keyword on the OUTFILE
subcommand, saving the robust covariance matrix estimates to a new file or
dataset.
- ROBUST displays a second set of output for any custom hypothesis tests
that are specified on LMATRIX subcommands, with results based on the
specified robust covariance matrix estimator.
--- end of excerpt ---

Source:
https://www.ibm.com/support/knowledgecenter/en/SSLVMB_26.0.0/statistics_reference_project_ddita/spss/base/syn_unianova_robust.html

IMO, when a user includes a ROBUST sub-command, it would be FAR more
convenient (and less confusing) to simply show the usual output using robust
SEs in place of the ASEs. This is what happens in Stata, for example, when
one uses the vce() option on -regress-.

Bruce

Bruce Weaver wrote

> bweaver@