Testing Foreign Language Reading Comprehension: The Immediate Recall Protocol

Abstract

Traditional methods of testing reading comprehension are problematic. True/false and multiple choice tests, for example, are often not passage dependent. In a 1975 study conducted on multiple choice questions in standardized reading tests in the native language, Pyrczak found no significant difference between the scores of students who read a passage and selected an answer to comprehension questions and scores of students who simply selected a, b, c, or d in answer to the same comprehension questions without reading the text.' Pyrczak cited three reasons for the results: 1) prior knowledge; 2) the interrelatedness of the questions; and 3) the general construction of multiple choice tests. In a later publication, Pyrczak with Axelrod concluded that teachers should pretest a test before administering it, in order to determine whether it is indeed passage dependent.2 They concluded that if reading tests are passage independent and students do not have to read carefully to peiform at criterion level, the testing system defeats the purpose of reading instruction. Even if problems of passage dependence vs. passage independence were resolved with multiple choice or true/false statements carefully worded and randomized, another stumbling block remains. Johns questions whether it is productive at all to test silent reading since passage dependence for some readers may be passage independence for others. It is difficult to know what knowledge the reader already has at hand.3

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 84 citations exist upstream.