Reading and Evaluating Research: Paper 4

1
Reading and Evaluating Research Paper 4

• Yamashita, J. (2003). Processes of taking a
  gap-filling test Comparison of skilled and less
  skilled EFL readers. Language Testing, 20(3),
  267-293.
• GU Yongqi
• Victoria University of Wellington, New Zealand
• peter.gu_at_vuw.ac.nz

2
A top-down analysis of the study

• What should be studied, based on theory, as
  opposed to
• What is studied

3
Text processing ability in gap-filling
Person
Who is the student?
Background knowledge
Language proficiency
Text processing skill
What info is used?
Text processing
Extralinguistic context
Text
item
item
What info can be used?
Linguistic context
clause
sentence
text
4
Guiding principles for analysis

• How an item is answered is both person-dependent
  and text-dependent, therefore,
• Subject selection influences text processing
  (what info will be made use of, by whom)
• the more, the better
• The more representative, the better
• Text selection influences text processing (what
  info is available to be made use of)
• Selection of text
• Different texts contain different amount of
  info for the learner to make use of)
• Selection of items
• Rationale deletion what to delete
  influences what info is needed in filling the
  gap, e.g., rationale deletion vs. random deletion

5
This study

• The purpose of the study is construct validation
  of gap-filling tests
• i.e., filling in rationally deleted gaps in a
  text can be a good measure of higher order
  skills and global comprehension.
• If we deliberately delete items that require
  text-level processing, rather than clause and
  sentence-level processing (p.272, Materials), and
  design a study that elicits how readers process
  the gaps in reading, isnt the argument circular?
  In other words, arent the findings a result of
  the design?