Y603 Lectures
Online
Lecture #15
1. Introduction to Randomized Block (RB)
and Generalized Randomized Block (GRB) designs
What makes a variable a blocking
variable? Give a few examples in your field.
What is the additional assumption
required by the randomized block design which is not necessary
with conventional CR-p or CRF-pq design?
What is the main difference between the
RB-p design and the GRB-p design? How does this difference affect
the way we carry out the F-test of the treatment
effect?
How to write SAS program code for RB-p
and GRB-p designs
How to interpret the post-hoc analysis of
mean differences based on these designs
How to compute 
for each significant treatment effect derived from RB or GRB
designs
2. Randomized Block Factorial (RBF)
Design
-- An extension of the randomized
factorial design.
-- You start with the usual CRF pq design
with two or more independent variables.
-- Then block subjects on a suitable
blocking variable such as IQ.
-- The underlying assumption is still no
interaction between the blocking variable and any
of the independent
variables.
3. Assignments:
(1) Review Sections 7.1, 7.2, and
7.4, in Kirk.
(2) Finish reading the handout on RB
design.
(3) Preview Sections 7.9 and 7.10 in
Kirk.
