Pinpointing the classifiers of English language writing ability: A discriminant function analysis approach

The  major  aim  of  this  paper  was  to  investigate  the  validity  of  language  and intelligence  factors  for  classifying  Iranian  English  learners`  writing  performance. Iranian  participants  of  the  study  took  three  tests  for  grammar,  breadth,  and  depth  of vocabulary, and two tests for verbal and narrative intelligence. They also produced a corpus  of  argumentative  writings  in  answer  to  IELTS  specimen.  Several  runs  of discriminant function analyses were used to examine the classifying power of the five variables  for  discriminating  between  low  and  high  ability  L2  writers.  The  results revealed that among language factors, depth of vocabulary (collocational knowledge) produces  the  best  discriminant  function.  In  general,  narrative  intelligence  was  found to  be  the  most  reliable  predictor  for  membership  in  low  or  high  groups.  It  was  also found  that,  among  the  five  sub-abilities  of  narrative  intelligence,  emplotment  carries the  highest classifying  value.  Finally, the applications and implications of the results for  second  language  researchers,  cognitive  scientists,  and  applied  linguists  were discussed.