The only recent question that I can see you posted (Mar 22) was tagged onto an existing thread so it is possible people assumed you were answering the question raised by the original poster.
Do not use html - learn how to send emails as plain text.
If you have a question, start a new thread, do not add it as a reply to an existing question.
And the usual, post a reproducible example with data and code. Your last question did that. Try posting again, in plain text and creating a new subject line. Use dput() to send your data to the list. Just listing it with print can leave out important information.
David L. Carlson
Department of Anthropology
Texas A&M University
From: R-help [mailto:[hidden email]] On Behalf Of Santiago Bueno
Sent: Monday, April 17, 2017 10:29 PM
To: [hidden email] Subject: [R] R help
I need help with R, and although I have posted my questions, no one seems
to care. Can some one coach me in formulating a correct question?
I am assuming that you are referring to your emails from last October and last month regarding nlme.
A) Read the Posting Guide, which mentions things like the fact that you should set your email program to send plain text when posting on this mailing list , and that there is a dedicated R-sig-mixed-models mailing list where questions about nlme would be more on topic. (The plain text thing is important to avoid us receiving a corrupted version of what you sent, so it is important if we are to understand you.)
B) Don't reply to an existing message on the list with a completely different question... start a fresh email with an informative subject line and all of the replies will show up together in many email programs and in the archives. Hijacked email threads tend to get overlooked, which is not in your best interest.
C) Make sure your example is reproducible. See for example  or  or  (or all three).
D) A course in linear regression should address when it is reasonable to remove terms, but it is beyond the scope of this list to go into that. Removing the intercept due to large p-values is hardly ever justified. Mixed models of the complexity you are trying to use typically require a lot more data than you showed last month to give significant results. You should probably consult with a local statistician on your experimental design.
On April 17, 2017 8:28:56 PM PDT, Santiago Bueno <[hidden email]> wrote:
>I need help with R, and although I have posted my questions, no one
>to care. Can some one coach me in formulating a correct question?
> [[alternative HTML version deleted]]
>[hidden email] mailing list -- To UNSUBSCRIBE and more, see
>https://stat.ethz.ch/mailman/listinfo/r-help >PLEASE do read the posting guide
>http://www.R-project.org/posting-guide.html >and provide commented, minimal, self-contained, reproducible code.