An Investigation Into The Effects Of Background Knowledge Intervention On English Reading Comprehension

This study was an attempt to investigate whether different interventions varied in the effects on the processes as well as the products of reading, whether they had different impacts on different types of textual information, and whether these effects varied with the reading proficiency.The effects of 2 pre-reading interventions were explored in this study:1) T1, where the participants were provided with a text (Text 1) introducing the topic of the test material (Text 2); and 2) T2, where the participants were engaged in a discussion after reading Text 1. The effects of the 2 experimental conditions were compared with the control condition To. The experiment was a 3×3 factorial design, the independent variables being reading proficiency (low, medium and high levels) and interventions (To, T1 and T2). Three intact classes were randomly sampled from each proficiency level (N= 322), provided with conditions of To, T1 and T2, respectively, before reading Text 2. Both immediate and delayed (2 days later) written recalls were collected from each participant, which were graded against a weighted scoring system. Quantitative analyses were conducted based on the scores from both recalls. In addition, qualitative coding was done based on the erroneous idea units and those inconsistent with Text 2 in the immediate recall protocols, and the relationships between the coded categories and interventions were examined through a number of non-parametric tests.The quantitative data analyses indicated that, in both immediate and delayed recalls, both Ti and T2 facilitated the comprehension and retention of the main idea of Text 2, but their effects on the total amount of information varied with the participants'reading proficiency levels. On the whole, in terms of the quantity of information correctly recalled, the interventions benefited the medium-level participants more than the high-level, and detracted from the low-level participants' performance; T1 was more facilitative than T2. The results of the qualitative analyses corroborated the quantitative results on the one hand, and revealed the advantage of T2 over T1, on the other hand, in the quality of local processing for the low-level participants, and in the global and extra-textual processing for the medium-and high-level participants. The findings were interpreted from the information-processing perspective.