Several possible methods, used to evaluate treatment effects in a randomize
d pretest-posttest trial with two treatment groups are the two-sample t tes
t, the paired t test, analysis of covariance I (ANCOVA I), the analysis of
covariance II (ANCOVA W), and generalized estimating equations (GEE). The A
NCOVA I includes treatment and baseline response as covariates in a linear
model and ANCOVA H additionally includes an interaction term between the ba
seline response and treatment indicator as a covariate. The parameters in t
he ANCOVA I and ANCOVA H models are generally estimated using ordinary leas
t squares. In this article, a semiparametric model, which makes no assumpti
ons about the response distributions, is used. The asymptotic properties of
the estimators derived from these five methods and their relative efficien
cies are discussed under this semiparametric model. We show that all these
methods yield consistent estimators for the treatment effect which have asy
mptotically normal distributions under the semiparametric model. The GEE an
d the ANCOVA H estimators are asymptotically equivalent and the most effici
ent. The estimators from other three methods are less efficient except unde
r some special conditions which are outlined in the article.