Re: st: ANOVA for quartiles

[Nick Cox <njcoxstata@gmail.com>]

With -by:-, the command to the right of the colon must make sense, so  
that you could run it on its own. Here the -oneway- call is illegal as  
no factor variable is specified. As Jay suggests, your problem is not  
one-way anova any way.

Nick

On 18 Sep 2012, at 23:08, Vasan Kandaswamy <vasan.kandaswamy@ki.se>  
wrote:

> Dear Statalisters,
>
> I intend to perform an ANOVA to compare the means of a variables  
> across quartiles of BMI ( body mass index) in two genders (male and  
> female ).
>
> Example: Male    Female
> BMI Q1 :  5.2        5.1
> BMI Q2 :  5.6        5.4
> BMI Q3 :  6.3        6.8
> BMI Q3 :  7.7        7.9
> I want to compare these values across the quartiles using ANOVA, and  
> look for significance in both genders separately.
>
> bysort bmi_quartiles gender : oneway bmi
>
> There seems to be something wrong in this script.
*
*   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/