UBC Theses and Dissertations
UBC Theses and Dissertations
Skimming strategy in reading as a function of familiarity with content and redundancy reduction in printed discourse Nacke, Phil L.
The validity of the theory of skimming as a process of looking only at the key words in continuous discourse was investigated in the present study. The primary research questions raised were whether skimming by looking only at key words is an effective reading strategy and to what extent skimming is essentially a process of disembedding key words. Attention was also given to the relationship of familiarity with the content of stimulus passages to skimming performance. The research design involved manipulating three independent variables: (1) Form of presentation of each of two stimulus passages (key words underlined; key words presented by masking the non-key words; original text with no key words identified; and no stimulus passage at all, the control group); (2) Levels of key words (three levels of redundancy reduction: syntactic, lexical and morphological, and anaphoric or discourse redundancy); and (3) Familiarity with content of the stimulus passage (pretest familiarity: Pretest and No« Pretest; and amount of background information: High, Middle, or Low Pretest scores). The criterion measure for each stimulus passage was a set of multiple-choice questions which were administered as a pretest (Pretest group only) and as a posttest following the skimming task(s). The dependent variables were the posttest raw scores and the information gain scores (posttest score minus pretest score). Scores on the VanWagenen Rate of Comprehension Scale were used as a covariate. Grade 11 subjects were randomly assigned to Pretest and No-Pretest groups in the first experimental session. Within these respective groups the Ss were randomly assigned to one of seven treatments for the second session at which the Ss were directed to skim two passages (science and history) under a time-limit condition. The results of this study indicate that: 1. Familiarity with the content of the reading materials is one of the important factors involved in the skimming process. It was observed that familiarity with the content could be induced by exposure to a related pretest. In fact, skimming was effective only when there was cueing via exposure to the pretest and then only on the science passage- Overall, skimming did not appear to be an effective reading strategy on the history passage. As predicted, having a greater amount of background information did facilitate skimming on the science passage. 2. Elimination of non-key words did not affect skimming performance on the science passage at any level of redundancy reduction. Therefore, since no significant effects were observed either due to having non-key words eliminated or to the amount of redundancy reduction (levels of key words), it was concluded that grade 11 Ss are able to gain information through skimming by looking only at key words in continuous discourse. Contrary to expectations, however, having key words identified did not facilitate skimming at any level of redundancy reduction. Consequently, skimming cannot be said to be essentially a process of disembedding key words. 3. There was no significant interaction between and among the three independent variables: forms of the passages, levels of key words, and amount of background information. Implications of the conclusions for methods and materials to be used for instruction are that (1) some materials may not be appropriately skimmed; (2) while readers may be capable of skimming-by capitalizing on the redundancy and predictability of the language, they apparently need instruction and practice in order to take advantage of these characteristics of the language in efficient information processing. The relationship between skimming and other factors such as immediate or short-term memory and practice effect should be investigated.
Item Citations and Data