educational site they worked at. The teachers were encouraged
to conduct informal peer observations and meet as regularly as they could
during the term time. The ten AfL guiding principles and their linked 50
AfL indicators were used as the basis for discussing teachers’ AfL practices
at the TLC and SB-TLC meetings. To further this, teachers were encouraged
to keep a portfolio so that they could keep track of their professional growth
over the project period.
THE IMPACT OF THE AFL PDP ON LEARNING
Pre- and Post-Tests
The two-group control group design was used as one research method to
investigate the impact of AfL PDP on learning. Pre–and post–AfL PDP
intervention tests were used to provide some measure of impact as a result
of the intervention. Shuttleworth (2009) stresses that this kind of research
method is, by far, the simplest and most common of the pre- and post-test
designs and is a useful way of ensuring strong level of internal validity.
Cohen, Manion and Morrison (2003), however, point out some potential
threats to internal validity of this research method. They say that, frequently
in educational research, events other than the experimental treatments
occur during the time between pre- and post-test observations. Such
events produce eff ects that can mistakenly be attributed to diff erences in
treatments. Also, it is very diffi cult, if not impossible, to judge whether the
process of pre-testing actually infl uenced the results even if a reasonable
time gap is given between the two administrations of the same test. There is
no guarantee on external validity, either. For example, it is both unethical
and impossible to isolate the two groups of students and also their teachers
in the same school during the whole investigative period. Therefore, it
is reasonable to assume that these students, and also their teachers, mix
outside lessons and share ideas, potentially contaminating the results. But,
if students are drawn from diff erent schools to prevent this, the chance of
selection bias arises. All in all, Shuttleworth (2009) contends that, as long
as its limitations are fully understood, the two-group control group design
is a very useful research method to study impact of treatment.
With the support of the BCA team at the Education Bureau (EDB), the
test items were adapted from TSA past papers used in 2008 for students
at the end of their primary 3 (Key Stage One). The paper assessed the
basic competencies of the students in the diff erent dimensions of three
subjects (i.e., English Language, Chinese Language and Mathematics).
There were four sub-domains in English Language and Chinese Language
(i.e., reading, listening, writing and speaking), each of which was tested
separately. There were four sub-domains in Mathematics (i.e., arithmetic