Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.

# Re: st: Trend Regression Coefficients of Ordinal Predictor

 From William Buchanan To "statalist@hsphsun2.harvard.edu" Subject Re: st: Trend Regression Coefficients of Ordinal Predictor Date Tue, 15 Jan 2013 07:25:34 -0800

```Hi Karman,

This definitely makes it much clearer.  Do you have the underlying continuous variable or just the ordinal variable? If so, why not use the continuous variable rather than discarding a substantial amount of information by using an ordinal representation of it? If you only have the ordinal variable you should look at -contrast- to do what you were mentioning and what the paper was demonstrating.

HTH,
Billy

Sent from my iPhone

On Jan 15, 2013, at 7:15, Karman Tandon <karmantandon@gmail.com> wrote:

> Hi Billy & the rest of Statalist,
>
> In many papers, a continuous variable will be regressed across
> quintiles of another variable, resulting in a beta coefficient for
> each level of the predictor variable. Then, the authors will write a
> "p for trend" to show that the beta coefficients are significantly
> trending as the levels of the predictor variable increase or decrease.
> What does "p for trend" mean statistically, and how can I arrive at a
> "p for trend" using Stata?
>
> Here is an example of the above in a table from a journal article in
> Cancer Epidemiology, that anyone should be able to access:
> http://cebp.aacrjournals.org/content/14/12/2881/T3.expansion.html
>
> Thanks!
> Karman Tandon
>
> PS this is my first post, so please forgive any mistakes
>
>
> On Tue, Jan 15, 2013 at 1:04 AM, William Buchanan
> <william@williambuchanan.net> wrote:
>>
>> Please read the Statalist FAQ prior to submitting a query like you were asked when joining the list (specifically the section regarding the use of full names).  Your question is also fairly vague and you may want to illustrate what you want more explicitly to get an answer.
>>
>> HTH,
>> Billy
>>
>> Sent from my iPhone
>>
>> On Jan 14, 2013, at 21:45, karmantandon <karmantandon@gmail.com> wrote:
>>
>>> I am fitting a regression model with a normally distributed outcome (y)
>>> and an ordinal variable in 4 levels (x) as predictor, along with other
>>> variables (a, b), using the command
>>>
>>> regress y i.x a b
>>>
>>> I get separate coefficients for the levels of the predictor x while the
>>> first level is treated as reference. I would like to demonstrate that the
>>> coefficients of the ordinal variable levels are trending in a certain
>>> direction with the variable itself. Is there a way to demonstrate this?
>>>
>>> Thank you!
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/
```