st: Standard error of between/within subject factors.

[K C Wong <kcwong5@gmail.com>]

Dear all statalist users,

I'm studying GEE at the moment and I don't understanding the passage
below. Could if anyone kindly explain to me in perhaps in
non-statistical term or with examples/formula etc? Sorry for this is
not a STATA question.
Thanks in advance. Have a great weekend.

"Standard errors for between-subject factors are likely to be
underestimated under the independent assumption because treating
observations from the same subject as independent and thus increasing
the apparent sample size. On the other hand, standard errors
within-subject factors are likely to be overestimated since we do not
control for residual between-subject variability."
*
*   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/