A study of the summarizing strategies used by ESL first year science students at the University of Botswana

Chimbganda, A.B. (2007) A study of the summarizing strategies used by ESL first year science students at the University of Botswana. PhD thesis, Rhodes University.




One of the major problems faced by speakers of English as a second language (ESL) or non-native speakers of English (NNS) is that when they go to college or university, they find themselves without sufficient academic literacy skills to enable them to navigate their learning successfully, such as the ability to summarize textual material. This thesis examines the summarizing strategies used by ESL first year science students at the University of Botswana. Using multiple data collection methods, otherwise known as triangulation or pluralistic research, which is a combination of quantitative and qualitative methods, one hundred and twenty randomly sampled students completed questionnaires and summarized a scientific text. In order to observe the students more closely, nine students (3 high-, 3 average- and 3 low-proficiency) were purposively selected from the sample and wrote a further summary. The nine students were later interviewed in order to find out from them the kinds of strategies they had used in summarizing the texts. To obtain systematic data, the summaries and the taped interview were coded and analyzed using a hybrid scoring classification previously used by other researchers. The results from the Likert type of questionnaire suggest that the ESL first year science students are 'aware' of the appropriate reading, production and self-assessment strategies to use when summarizing. However, when the data from the questionnaire were cross-checked against the strategies they had used in the actual summarization of the text, most of their claims, especially those of the low-proficiency students, were not sustained. As a whole, the results show that high-proficiency students produce more accurate idea units and are more capable of generalizing ideas than low-proficiency students who prefer to "cut and paste" ideas. There are also significant differences between high- and low proficiency students in the manner in which they decode the text: low-proficiency students produce more distortions in their summaries than high-proficiency students who generally give accurate information. Similarly, high-proficiency students are able to sort out global ideas from a labyrinth of localized ideas, unlike average- and low-proficiency students who include trivial information. The same trend is observed with paraphrasing and sentence combinations: high-proficiency students are generally able to recast and coordinate their ideas, unlike low-proficiency students who produce run-on ideas. In terms of the discrete cognitive and meta-cognitive skills preferred by students, low proficiency students are noticeably unable to exploit pre-summarizing cognitive strategies such as discriminating, selecting, note-making, grouping, inferring meanings of new words and using synonyms to convey the intended meanings. There are also greater differences between high- and low-proficiency students when it comes to the use of meta-cognitive strategies. Unlike high-proficiency students who use their reservoir of meta-cognitive skills such as self-judgment, low-proficiency students ostensibly find it difficult to direct their summaries to the demands of the task and are unable to check the accuracy of their summaries. The findings also show that some of the high-proficiency students and many average- and low-proficiency students distort idea units, find it difficult to use their own words and cannot distinguish between main and supporting details. This resulted in the production of circuitous summaries that often failed to capture the gist of the argument. The way the students processed the main ideas also reveals an inherent weakness: most students of different proficiency levels were unable to combine ideas from different paragraphs to produce a coherent text. Not surprising, then, there were too many long summaries produced by both high- and low-proficiency students. To tackle some of the problems related to summarization, pre-reading strategies can be taught, which activate relevant prior knowledge, so that the learning of new knowledge can be facilitated. During the reading process students can become more meta-cognitively aware by monitoring their level of understanding of the text by using, for example, the strategy suggested by Schraw (1998) of "stop, read and think". Text analysis can also be used to help the students identify the main themes or macro-propositions in a text, and hence gain a more global perspective of the content, which is important for selecting the main ideas in a text. A particularly useful approach to fostering a deeper understanding of content is to use a form of reciprocal or peer-mediated teaching, in which students in pairs can articulate to each other their understanding of the main ideas expressed in the text. As part of the solution to the problems faced by students when processing information, we need to take Sewlall's (2000: 170) advice that there should be "a paradigm shift in the learning philosophy from content-based to an emphasis on the acquisition of skills". In this regard, both content and ESL teachers need to train their students in the explicit use of summarizing strategies, and to plan interwoven lessons and learning activities that develop the learners' intellectual ways of dealing with different learning problems so that they can make learning quicker, easier, more effective and exciting.

Item Type:Thesis (PhD)
Additional Information:Ph.D. (English Language & Linguistics)
Uncontrolled Keywords:English language, English as a second language, foreign speakers, Botswana, science students, summarizing science texts, summarizing strategies, pre-reading strategies, language and education
Subjects:Y Unknown > Subjects to be assigned
Divisions:Faculty > Faculty of Humanities > English Language and Linguistics
Supervisors:De Klerk, V.A. (Prof.) and Adendorff, R.D. (Prof.)
ID Code:891
Deposited By: Rhodes Library Archive Administrator
Deposited On:14 Dec 2007
Last Modified:06 Jan 2012 16:19
2598 full-text download(s) since 14 Dec 2007
486 full-text download(s) in the past 12 months
More statistics...

Repository Staff Only: item control page