Home » Predicting reading comprehension from authentic text. by Kristin Lee Swenson
Predicting reading comprehension from authentic text. Kristin Lee Swenson

Predicting reading comprehension from authentic text.

Kristin Lee Swenson

Published
ISBN : 9780549678915
NOOKstudy eTextbook
120 pages
Enter the sum

 About the Book 

Reading comprehension was predicted for graduate students reading authentic, expository text. Three factors were used to predict comprehension- these factors were Latent Semantic Analysis (LSA), Causal Cohesion, and Flesch Reading Ease scores.MoreReading comprehension was predicted for graduate students reading authentic, expository text. Three factors were used to predict comprehension- these factors were Latent Semantic Analysis (LSA), Causal Cohesion, and Flesch Reading Ease scores. Reading comprehension was measured as both accuracy and reading speed. The data was analyzed using hierarchical linear modeling and a separate analysis was conducted for each of the two dependent variables. The results showed that LSA and Flesch Reading Ease scores combined to predict 15% of the variance in accuracy and 10% of the variance in reading speed. These results extended to authentic text previous findings that showed LSA to predict comprehension in experimenter created. The results also indicated that measures of coherence and measures of traditional readability worked together to account for more of the variance in reading comprehension than either one could account for independently. No significant results were obtained for the causal measure. The findings in this study are interpreted within the framework of the construction-integration (CI) model and it is argued that they address some of the limitations of using propositional modeling to predict comprehension.