factors like literacy is becoming substandard in terms of research design. The
whole language learning business is based on learning with written support,
though some work on language learning with dyslexic learners also has
looked at the reduction of the role of script and written support.
Fred Genesee sees a growing interest in K-12 and minority language learners.
Both in North America and western Europe there is a tendency to do
research on immersion settings that go beyond assessing that there are now
negative effects. There is also more interest in the pedagogy of immersion
and Content and Language Integrated Learning (CLIL). As he says:“Immer-
sion is rich on research, CLIL is rich on pedagogy and they should be
brought together.”Lourdes Ortega argues that AL research has focused
almost exclusively on elite bilingualism, college FL learning and hardly on
minorities; this in contrast to what happens in Europe. She would like to see
more groups in society able to profit from AL research, in particular lin-
guistic minorities, heritage learners, but also for communication in hospitals
and courts. Susan Gass agrees:“What do we know about non-elite multi-
lingualism, the effects of migration?”At the same time Andrew Cohen argues
that applied linguists should be concerned not only with incipient learners
and basic skills but also with higher-level skills.
6.2.10 International comparisons
While for reading and math there are worldwide comparisons on how well
students do, no such comparisons on the scale of, for example, TIMSS
(Trends in International Mathematics and Science Study) and PIRLS (Pro-
gress in International Reading Literacy Studies) have been done for English
or any other language. Lyle Bachman speculates why that is the case. In his
view it would be technically possible to develop a test for making interna-
tional comparisons in English and other languages, but the only proposal he
is aware of for developing such a test for English failed because no one was
willing to spend money on it.
6.2.11 Generalizability
Generalizability is the holy grail of empirical research. In fact, it is a dream in
my view, because no study in ourfield can claim to be based on a sample
that is representative of the population, since we do not know the parameters
that define the population. Studying the typical second-year university students
does not help either. John Schumann:
Molenaar’s work distinguishing ergodic research from idiographic
inquiry that provides us with information that frees us from the notion
that truth is found only in group studies. He shows that what is true for
a group cannot be generalized to the individual except under very special
conditions. Here we have a nice symmetry –traditional views have70 Main trends I