Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
As others have suggested, the models could have used the actual BMI,
and not reduced it to three categories. The appropriate functional
form of BMI may not be linear. In that situation one would have to
determine the functional form and fit it. The task would be more
complicated if the appropriate functional form of BMI differed among
the outcome measures.
Does the article report the values that were assigned to the
adjustment variables to get the means shown in the table? The most
likely choice is the overall mean of each of the adjustment variables.
That information is needed in interpreting the means in the table.
Also, the authors should have checked that the values assigned jointly
to the assignment variables are compatible with the data.
On Tue, Jan 15, 2013 at 10:15 AM, Karman Tandon <email@example.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:
> Karman Tandon
* For searches and help try: