# RE: st: ANOVA

 From "Steichen, Thomas J." To Subject RE: st: ANOVA Date Tue, 21 Aug 2007 10:59:32 -0400

```Partial SS usually do not add up to the total SS... The only
(non-pathological) exception is when all your independent
variables are perfectly uncorrelated. Balanced data is not
immune to this.

Sequential SS will add up to the Total SS.

Partial SS treats each independent variable as if it were the
last variable considered. Thus any SS that might also be due
to some other correlated independent variable is not attributed
to the variable in question. In essence, any SS that might be
explained more than one way is not captured in Partial SS.
Thus the total of Partials is usually less than the Total SS.

Tom

-----------------------------------
Thomas J. Steichen
steicht@rjrt.com
-----------------------------------

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Aparna Mathur
Sent: Tuesday, August 21, 2007 10:00 AM
To: statalist@hsphsun2.harvard.edu
Subject: RE: st: ANOVA

Thanks for responding. Here is the result, and my data set is balanced.
All the variables uptil statepci are dummy variables.

.. anova annpremium alcman drugman mentalman mentparity chiro dental
crating gissue statepci unprt smallfrmemp1 pop medicare agedpop,
continuous (statepci unprt smallfrmemp1 pop medicare agedpop)

Number of obs =  137     R-squared     =
0.3686
Root MSE      = 782.485     Adj R-squared
=0.2962

Source |  Partial SS    df       MS           F
Prob >F
------------+----------------------------------------------------
Model |  43614642.7    14  3115331.62       5.09
0.0000
|
alcman |  1344769.68     1  1344769.68       2.20
0.1409
drugman |   2536376.6     1   2536376.6       4.14
0.0440
mentalman |  1869869.15     1  1869869.15       3.05
0.0831
mentparity |  610999.438     1  610999.438       1.00
0.3198
chiro |  4911205.77     1  4911205.77       8.02
0.0054
dental |  1663238.75     1  1663238.75       2.72
0.1019
crating |  1743577.33     1  1743577.33       2.85
0.0941
gissue |  2879808.29     1  2879808.29       4.70
0.0320
statepci |  19266358.9     1  19266358.9      31.47
0.0000
unprt |  2069599.21     1  2069599.21       3.38
0.0684
smallfrme~1 |  1557806.53     1  1557806.53       2.54
0.1133
pop |  1056893.25     1  1056893.25       1.73
0.1914
medicare |   4340.0111     1   4340.0111       0.01
0.9330
agedpop |  211267.685     1  211267.685       0.35
0.5580
|
Residual |  74698451.1   122  612282.386
------------+----------------------------------------------------
Total |   118313094   136   869949.22

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of David Airey
Sent: Tuesday, August 21, 2007 9:48 AM
To: statalist@hsphsun2.harvard.edu
Subject: re: st: ANOVA

..

You might post those results in text format and describe if your data
set is balanced.

> Hi,
>
> My problem is that when I do ANOVA, the sum of the partial SS for
> all the independent variables in the model does not add up to the
> Model SS. I have both categorical and continuous variables in my
> model. Could you help me understand why this might happen?
>
>
> Thanks,
> Aparna
>
>

--
David C. Airey, Ph.D.
Research Assistant Professor

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

-----------------------------------------
CONFIDENTIALITY NOTE: This e-mail message, including any
attachment(s), contains information that may be confidential,
protected by the attorney-client or other legal privileges, and/or
proprietary non-public information. If you are not an intended
recipient of this message or an authorized assistant to an intended
then delete it from your system. Use, dissemination, distribution,
or reproduction of this message and/or any of its attachments (if
any) by unintended recipients is not authorized and may be
unlawful.

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```

• Follow-Ups: