Selecting participants (Ss)
Above: the main square
in Tuscania; the Post
|Discussion Topic 5:
Selecting participant (Ss) for L2 research
As everyone knows, the random selection of participants (Ss) from the one or more
populations of interest in a study is a hallmark of behavioral research. A researcher who
randomly selects Ss for a study of height from the student body of Central High School will
certainly find that 16-year-old boys are taller than 16-year-old girls. The expected finding
might NOT be obtained, however, if the researcher recruits all of the female participants
for the study from the High School's girls basketball team and all of the male participants
from the school's chapter of the "Friends of Pygmies Society". This is an obvious example
of sampling bias. Alas, sampling bias is not always obvious.
Very few, if any, L2 speech studies have adhered to the requirement of random selection
of participants. This is usually done for reasons of necessity rather than out of ignorance
of how behavioral research should be conducted. In some cases, unfortunately, the
"shortcut" taken in participant selection may influence not only the outcome of the
research but also it's interpretation.
I'll begin with some critical comments regarding research in my lab, and point to what I
consider to be some of the limitations that exist. (to follow)
I am well aware of how hard it is to recruit participants (Ss) for L2 research. I was very
lucky, having time and financial resources for my research. Many researchers have little
time and even less money, which means that real ingenuity is needed to find page.