Reading Modality and Question-Type Performance At Three Levels of Passage Difficulty (Comprehension, Text, Oral-Silent, Literal-Inferential).

Recent evidence suggests that differences in the ability to comprehend when reading orally and silently and in the ability to answer literal and inferential questions are a function of reader competence: high, average or low. This argument is limited, however, by the use of grade-placement passages as the measure; that is, task difficulty has not been accounted for as a factor influencing performance differences among less-skilled readers. The present study is an attempt to determine to what extent, if any, differential reading modality and question-type performance are affected by level of passage difficulty. Eighty-six third, fourth and fifth grade students were administered passages from the Analytical Reading Inventory (Woods and Moe, 1977). Three levels of difficulty were used: (1) two levels below grade-placement, (2) at grade placement and (3) three levels above grade-placement. One passage at each level was read orally, one silently. Questions were classified as literal or inferential. Only those whose total score on the grade-placement passages was higher than 60% continued on to read at the highest passage level. Four levels of competence were found. Data were analyzed using a repeated measures analysis of variance. Results indicate that passage difficulty does not affect (1) the ability to comprehend when reading orally vs. silently or (2) the ability to answer literal vs. inferential questions. Consequently, this evidence provides indirect support for claims that evaluations which fail to take into account variables related to skill level, such as test format (oral vs. silent) or type of question asked (literal vs. inferential), may either overestimate or underestimate the reading comprehension abilities of students at specific levels of proficiency.Ph.D.Reading instructionUniversity of Michiganhttp://deepblue.lib.umich.edu/bitstream/2027.42/161096/1/8621289.pd