# Re: st: AW: ANOVA help

 From David Airey To statalist@hsphsun2.harvard.edu Subject Re: st: AW: ANOVA help Date Mon, 18 May 2009 09:51:19 -0500

```.

```
Some things you left out, such as are your attitudinal variables ratio/ interval, ordinal, or categorical? You did say you calculated means, so I'm assuming your attitudinal variables are continuous, but you should say. Do you really have multiple nested levels (neighborhood, city, region) and do you care about difference in particular areas or do you just want to control for that? In addition to what Martin mentioned, -xtmixed- could be used.
```

On May 18, 2009, at 9:19 AM, Martin Weiss wrote:

```
```<>

```
Well, my initial reaction is that this is a case for -anova-, and you could
```obtain further information from

Hamilton (2009), chapter 5, http://www.stata.com/bookstore/sws.html

or

Rabe-Hesketh and Everitt (2007), chapter 4,
http://www.stata.com/bookstore/statanalyses.html

Among free resources, the UCLA website
```
http://www.ats.ucla.edu/stat/stata/faq/ has a whole section on - anova-.
```

```
Also note the command -ttest- to compare means, as well as its cousins like
```-signtest- et al. ...

HTH
Martin

-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu
```
[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Rahsaan Maxwell
```Gesendet: Sonntag, 17. Mai 2009 22:34
An: statalist@hsphsun2.harvard.edu
Betreff: st: ANOVA help

Hello-

```
I am interested in comparing mean scores on several attitudinal variables
```for
```
two groups: A and B. I have a dummy variable C (which is coded 0- if group
```A,
1- if group B).

```
There is a statistical significance for the difference in the overall means
```for
these two groups.  However, I want to see whether or not the means for
members
```
of A and B who live in the same neighborhood/city/region are the same. Some
```preliminary cross-tabs suggest that attitudes for these two groups are
similar
when analyzed at these geographic levels.

```
So, I am thinking back to my intro stats class and wondering if ANOVA can
```test
```
whether the difference in the mean attitude scores for members of groups A
```and
```
B in the same neighorhood/city/region are statistically significant. If so,
```how do I do this?  If not, is there another test I should be using?

Thanks,

-Rahsaan

Rahsaan Maxwell, Ph.D.
Assistant Professor
Department of Political Science
University of Massachusetts, Amherst

Postdoctoral Fellow
German Marshall Fund of the United States

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```
```

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/
```