Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: age: factor variables may not contain noninteger values


From   Lars Folkestad <lfolkestad@health.sdu.dk>
To   "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
Subject   Re: st: age: factor variables may not contain noninteger values
Date   Tue, 8 Nov 2011 12:14:16 +0100

This helps a lot.
Thank you.

But i have a question about the second point.
Is there a quick and easy way to round the age to the nearest integer
value? I could use the floor / ceil functions, but some of the ages
Should be rounded up and some should be rounded down.

lars

Den 08/11/11 11.40 skrev "Maarten Buis" <maartenlbuis@gmail.com>:

>On Tue, Nov 8, 2011 at 11:13 AM, Lars Folkestad wrote:
>> I am trying to investigate the interactions between age and a grouping
>>variable in a regression model, using this code:
>>
>> xi: regress Var1 age##i.group
>>
>> Var1: is a continuous variable
>> Age: is a continous variable / float
>>
>> When i run the code i get the following error message:
>> age:  factor variables may not contain noninteger values
>
>First, you are mixing -xi- and factor variable notation. You should
>just leave the -xi:- prefix away.
>
>Second, in most surveys I am aware of age will be measured in whole
>years. In those cases age should only contain integer values and this
>error message means I made an error while preparing the data. This is
>true regardless of how the data is stored. However, this is not
>necessarily an error, you could have fractional age if you had for
>instance exact birth dates and exact dates at which events of interest
>happen. Alternatively, I may have standardized age or rescaled it so
>it is measured in decades rather than years. However, in the latter
>two cases I would have created new with different variable names
>(probably z_age and age_d). So I would check whether non-integer age
>would make sense given the way your data was collected.
>
>Third, whether there is an error or not you probably want -regress
>var1 c.age##i.group- i.e. tread age as a continuous variable and not
>as a set of dummies.
>
>Hope this helps,
>Maarten
>
>--------------------------
>Maarten L. Buis
>Institut fuer Soziologie
>Universitaet Tuebingen
>Wilhelmstrasse 36
>72074 Tuebingen
>Germany
>
>
>http://www.maartenbuis.nl
>--------------------------
>
>*
>*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index